READING PERFORMANCE WITH STAND MAGNIFIERS IN AGE-RELATED … · Chapter 1 Literature review 1 1.1 Low vision 3 1.2 Age-related macular degeneration 10 1.3 Reading with age-related

READING PERFORMANCE WITH STAND

MAGNIFIERS IN AGE-RELATED

MACULAR DEGENERATION

Allen Ming Yan Cheong Bachelor of Science (Honours) in Optometry

This thesis is submitted for the degree of Doctor of Philosophy

(2003)

School of Optometry

Queensland University of Technology

Brisbane, Australia

Keywords

i

Keywords

Age-related macular degeneration (AMD)

Low vision

Low vision aids

Low vision rehabilitation

Magnification

Magnifiers

Reading

Training

Abstract

ii

Abstract This research was designed to address important issues for the effective prescription

of, and training in the use of, magnifiers for reading patients with visual impairment.

The emphasis was on the development of simple methods of assessment and training

that could be easily implemented, at no great cost, by low vision practitioners in

clinical practice. To ensure that the results would be widely applicable, the research

focused on subjects with age-related macular degeneration (AMD) using stand

magnifiers (being the most common cause of low vision and the most commonly

prescribed magnifiers respectively). From this research, modifications to the current

methods of reading rehabilitation are suggested to more effectively improve low

vision reading for the millions of people with low vision around the world.

The magnification and reading performance achieved with the magnifier determined

by the fixed acuity reserve method was as valid as that achieved with the magnifier

determined by the individual acuity reserve method. The fixed acuity reserve is a

simpler method to calculate the required magnification, as it requires only near visual

acuity and the patient’s goal reading task. This method was primarily used to select

the appropriate illuminated stand magnifiers for the subjects participating in the

subsequent studies and is recommended for use as the starting point in clinical low

vision practice.

The main study of this thesis was a longitudinal investigation of the benefit of large

print reading practice on reading performance with stand magnifiers. Instead of the

intensive training programs on magnifier use which have been suggested by previous

studies, this study aimed to investigate the effect of simple large print reading

practice, under either full or restricted field of view (the latter simulated by a practice

stand), on reading rate with stand magnifiers for subjects with AMD. The

experimental hypothesis was that reading practice prior to the prescription of stand

magnifiers would improve reading performance with the stand magnifiers for

subjects with AMD. As previous studies have shown, reading rate reduced when a

stand magnifier was first introduced. One week of reading practice on large print,

with or without a reduced field of view, gave an improvement in reading rate with

the stand magnifier for passages of text (such that the reading rates with and without

Abstract

iii

magnifiers were not significantly different). There was a suggestion that this practice

may give a more rapid improvement in reading rate than that achieved by the control

subjects who did not do any large print reading practice, but this did not reach

statistical significance. Even very brief reading with the stand magnifiers by the

control subjects gave some improvement in reading rate. Therefore, home or in-

office reading practice on large print or with magnifiers is recommended for patients

with AMD before magnifiers are prescribed.

Subjects who had neither reading practice nor exposure to the magnifier prior to its

prescription required two weeks practice using their stand magnifiers to achieve their

maximum reading rate. This suggests that home practice in using stand magnifiers is

beneficial and a follow up visit is recommended two weeks after the provision of a

magnifier to assess any change in reading rate. If no improvement in the magnifier

reading rate is found or the rate is less than the reading rate on large print without a

magnifier, further investigations of the patients’ vision and/or their magnifier

manipulation strategy are necessary.

In the last study, a simple method aimed at alleviating difficulties with magnifier

manipulation and navigation, the attachment of a line guide to the base of the stand

magnifier, was investigated using both objective methods (recording magnifier

movements and reading rate measures) and subjective methods (simple

questionnaire). Although there was no improvement in the objective measures of

reading or navigation performance with the line guide, more than half of the subjects

with low vision preferred to have the line guide on their stand magnifiers. This

suggests that the objective measures might not be sensitive enough to predict the

subjective response, or that other factors that were not measured in this study

influenced subjects’ preferences in selecting the line guide (e.g., psychological

support provided by the line guide in reading orientation). Clinically, the subjective

response of patients to the use of low vision aids as well as their motivation are

important criteria for success in low vision rehabilitation. There was a tendency for

less experienced users to prefer the line guide to assist their use of the stand

magnifier for reading. Therefore, a line guide could be offered as a preliminary

training aid when stand magnifiers are first prescribed for AMD patients. Possible

improvements to the design of the line guide were identified. Further research is

Abstract

iv

required to assess the benefits of this or similar devices for new magnifier users and

to understand the difficulties that people with visual impairment have with page

navigation in order to determine improved methods of training navigation strategies.

The unique contribution of this study to the field of low vision rehabilitation is that

the benefit of short-term reading practice, on large print or with magnifiers, as

simple, cheap methods of enhancing reading performance with stand magnifiers was

demonstrated. The results of this study have led to the development of

recommendations for assessing and training AMD patients who are prescribed stand

magnifiers.

Table of contents

v

Table of contents

Keywords iAbstract iiTable of contents vList of figures viiList of tables xiList of abbreviations xiiiStatement of original authorship xivAcknowledgments xv

Chapter 1 Literature review 1 1.1 Low vision 3 1.2 Age-related macular degeneration 10 1.3 Reading with age-related macular degeneration 26 1.4 Reading rehabilitation for people with age-related macular

degeneration (AMD) 50 1.5 Low vision training and rehabilitation for reading 70 1.6 Objectives of this study 77

Chapter 2 Validation of reading chart for measuring

reading performance 78 2.1 Introduction 79 2.2 Subjects 85 2.3 Methods 86 2.4 Analysis 90 2.5 Results 91 2.6 Discussion 102 2.7 Conclusion 105

Chapter 3 Validation of the method of calculating

magnification for reading with low vision 106 3.1 Introduction 107 3.2 Subjects 112 3.3 Methods 115 3.4 Analysis 124 3.5 Results and Discussion 125 3.6 Conclusion 136

Table of contents

vi

Chapter 4 Effect of practice on reading rate with stand

magnifiers 137 4.1 Introduction 139 4.2 Subjects 143 4.3 Methods 148 4.4 Analysis 168 4.5 Results 172 4.6 Discussion 210 4.7 Conclusion and recommendations 226

Chapter 5 Does a line guide improve reading performance

with stand magnifiers? 228 5.1 Introduction 229 5.2 Subjects 232 5.3 Methods 233 5.4 Analysis 250 5.5 Results 251 5.6 Discussion 262 5.7 Conclusion 269

Chapter 6 Conclusions and recommendations 271 6.1 Introduction 272 6.2 Main findings 272 6.3 Further research 279 6.4 Clinical recommendations 282 6.5 Summary 285

References 286

Appendices 326

List of figures

vii

List of Figures

Figure 1.1 Prevalence of visual impairment by age from the Blue Mountains Eye Study (1996) in Australia……………….….. 9

Figure 1.2 Prevalence of AMD by age………………………………….. 15Figure 1.3 Ray diagram of a hand-held magnifier where the object was

located within the object distance…………………………… 58 Figure 2.1 Example of the determination of maximum reading rate

(MRR) critical print size (CPS) and text visual acuity (VA)... 83Figure 2.2 Bailey-Lovie text reading chart……………………………… 87Figure 2.3 Mean and standard errors of maximum reading rate (MRR)

across 10 trials for each chart for the 18 subjects…………… 93Figure 2.4 Difference in maximum oral reading rate (MRR) between

the first and second trials (2 versions for each chart) for the MNRead and Bailey-Lovie text reading charts for the 18 subjects………………………………………………………. 94

Figure 2.5 Comparison of the mean and standard errors of critical print size (CPS) across 10 trials for each chart for the 18 subjects.. 95

Figure 2.6a Correlation of maximum reading rates (MRR) measured by the two reading charts. ……..……………………………….. 96

Figure 2.6b Correlation of critical print sizes (CPS) measured by the two reading charts………………………………………………... 96

Figure 2.7 Difference in maximum reading rate (MRR) between Bailey-Lovie text reading and MNRead charts measured at the first trial…………………………………………………………... 97

Figure 2.8 Differences in critical print size (CPS) measured using the Bailey-Lovie text reading and MNRead charts at the first trial for the 18 subjects………………………………………. 98

Figure 2.9a Correlation of maximum reading rates (MRR) determined by the two analysis methods at the first trial for the 18 subjects.. 99

Figure 2.9b Correlation of critical print sizes (CPS) determined by the two different methods at the first trial for the 18 subjects…... 100

Figure 2.10a Comparison of difference in maximum reading rates (MRR) determined by the MNRead Analysis 0.3 and graphical method at the first trial for Bailey-Lovie text reading chart… 101

Figure 2.10b Comparison of difference in critical print size (CPS) determined by the MNRead Analysis 0.3 and graphical method at the first trial for Bailey-Lovie text reading chart… 102

Figure 3.1 Example of one of the nine short reading passages…………. 117Figure 3.2 Measurement of reading rate without low vision aids using

Bailey-Lovie text reading chart……………………………… 119Figure 3.3 Example of reduced reading rate with low vision aids when

reading large print sentences………………………………… 123Figure 3.4 Log reading rates (RR) with low vision aids on passages at

target print size for both visits…………………….………… 126

List of figures

viii

Figure 3.5 Equivalent viewing distance (cm) determined by each

method for each subject……………………………………… 129

Figure 3.6 Log reading rates (RR) with low vision aids on passages at target print size for each subject at the first visit…………….. 129

Figure 3.7 Difference in log reading rates calculated by the fixed and individual acuity reserve methods versus mean of the log reading rates…………………………………………...…….. 130

Figure 3.8 Difference in log reading rates without LVAs at the first and second visits as a function of mean reading rates without LVAs………………………………………………………… 131

Figure 3.9a Difference in log reading rates with LVAs determined by fixed acuity reserve method at the first and second visits as a function of mean reading rates with LVAs………………….. 132

Figure 3.9b Difference in log reading rates with LVAs determined by the individual acuity reserve method at the first and second visits as a function of mean reading rates with LVAs……………... 133

Figure 3.10 Example of log reading rates (RR) as a function of print size with and without low vision aids (LVA) for sentences using the Bailey-Lovie text reading chart………………………….. 135

Figure 4.1 Flow chart of experimental interventions for each group…… 149 Figure 4.2 Example of a reading passage for large print reading……….. 156 Figure 4.3 Example of a reading passage for reading with stand

magnifiers...………………………………………………….. 157 Figure 4.4 Practice stand to simulate reduced field of view…………….. 159 Figure 4.5 Example of a large print reading book (N24 print)………….. 167 Figure 4.6 Log reading rate with STM on passages at CPS across time

(log-scale)…………………………………………………..... 174 Figure 4.7 Change in log reading rate across time for subjects from

control (N), large print practice (P1), large print under reduced field of view practice (P2) and clinical groups (C).... 176

Figure 4.8 Change of log passage reading rate with STM as a function of time (log-scale) for the experimental groups……………... 177

Figure 4.9 Distance visual acuity across time (log-scale)………………. 179 Figure 4.10 Near (word) visual acuity across time (log-scale)...…………. 180 Figure 4.11 Word threshold print size achieved with STM across time

(log-scale)…...……………………………………………….. 181 Figure 4.12 Log reading rate with STM across time (log scale) excluding

subjects whose distance visual acuity reduced by 2 lines or more………………………………………………………….. 183

Figure 4.13 Log reading rate without STM across time (log-scale)..…….. 186 Figure 4.14 Comparison of log reading rates with (solid line) and without

STM (dashed line) across time (log-scale) for each group….. 189 Figure 4.15 Log reading rate at week 20 for different reading assessment

at CPS………………………………………………………... 192 Figure 4.16 Comparison of log reading rates with STM for sentences

(represented in solid line) and passages (represented in dash line) across time (log-scale) for each group…………………. 194

List of figures

ix

Figure 4.17a Frequency of reading across time..…………………………... 198Figure 4.17b Duration of reading without STM across time..……………... 199Figure 4.17c Duration of reading with STM across time..………………… 199Figure 4.18a Reading material without STM read by subjects……...…….. 201Figure 4.18b Goal reading material defined by subjects……….………….. 201Figure 4.18c Primary reading material with STM at weeks 4 and 8...…….. 202Figure 4.19 Usefulness of STM across time reported by subjects..……… 203Figure 4.20 Limitations of the STM reported by the subjects…...……….. 204Figure 4.21 Correlation of log reading rates with and without STM at

week 2 for passages and sentences…….…………………….. 205Figure 4.22a Correlation between log reading rate for passages and near

visual acuity at week 2...…………………………………….. 207Figure 4.22b Correlation between log reading rate for sentences and near

visual acuity at week 2...…………………………………….. 207 Figure 5.1 The design of the line guide attached on the stand magnifier.. 237Figure 5.2 3-SPACE Isotrak system…………………………………….. 239Figure 5.3 Calibration traces of 2 mm steps movement (x-direction)…... 239Figure 5.4 Traces of movements of a stand magnifier in horizontal and

vertical directions during reading……………………………. 241Figure 5.5 Traces of a stand magnifier movement moved during reading 243Figure 5.6 Example of a reading passage for reading assessment…...….. 248Figure 5.7 Comparison of the change in log reading rate due to the use

of line guide and the length of STM use (log-scale)………… 253Figure 5.8a Difficulty in using stand magnifiers (subjective response

reported by the subjects)…………...………………………... 256Figure 5.8b Frequency of missing lines in using stand magnifiers………. 256Figure 5.9a Subjective preferences for the line guide and fixation patch... 258Figure 5.9b Reasons for preference for the line guide and fixation patch... 258Figure 5.9c Reasons for rejection of the line guide and fixation patch…... 259Figure 5.10a Distribution of reported difficulty in using STM for subjects

who preferred the line-guide………………………………… 260Figure 5.10b Distribution of reported difficulty in using STM for subjects

who did not prefer the line-guide……………………………. 260 Figure A1.1 Bailey-Lovie text reading chart……………………………... 327Figure A1.2 Example of a passage used in Chapter 3 to measure reading

rate with magnifiers ………………………….……………... 328Figure A1.3 Example of a passage used in Chapter 4 to measure reading

rate on large print without stand magnifiers……...………….. 328Figure A1.4 Example of a passage used in Chapter 4 to measure reading

rate with magnifiers ………………….……………………… 329Figure A1.5 Example of a passage used in Chapter 5 to measure reading

rate with magnifiers ………………..………………………... 330

List of figures

x

Figure A2.1 Reading rates with and without stand magnifiers for the

control group (N)…...………………………………………... 334 Figure A2.2 Reading rates with and without stand magnifiers for the large

print practice group (P1)…………………………………….. 335 Figure A2.3 Reading rates with and without stand magnifiers for the large

print with reduced field of view practice group (P2)………... 336 Figure A2.4 Reading rates with and without stand magnifiers for the

clinical group (C).…………………………………………… 337 Figure A3.1 Example of calculation of visual field loss for subject 40..…. 339 Figure A4.1 Ray diagram of a stand magnifier…………………………… 341 Figure A4.2 Conjugate object-image method………………………….….. 342 Figure A4.3 Real image method…………………………………………... 343 Figure A4.4 Determination of the dioptric power of a stand magnifier…... 344 Figure A6.1 Area of reading stand across which calibration measurements

were performed……………………………………………… 354 Figure A6.2 Layout of the page to measure the sensitivity and linearity of

the Isotrak in x-plane by step interval movements (e.g. 2 cm), at 17 cm below top of reading stand…………………… 355

Figure A6.3 An example of a plot of sensor position along x-plane for interval step movements of 2 cm……………………………. 356

Figure A6.4 Layout of the page to measure the sensitivity and linearity of the Isotrak in y-plane by step interval movements (e.g. 2 cm), at 27 cm in from the right side of the reading stand…… 357

Figure A6.5 An example of a plot of sensor position in the y-plane for interval step movements of 2 cm. Plot demonstrates good linearity across the full vertical extent assessed……………... 357

Figure A6.6 An example of an x-y plot of sensor position for continuous movements across a page……………………………………. 359

Figure A6.7 Isotrak position as a function of marker position (step intervals) on the paper……………………………………….. 361

List of tables

xi

List of Tables

Table 1.1 The relation between disease, impairment, activity limitation and participation restriction for vision (Dickinson, 1998)…………… 4

Table 1.2 Epidemiological studies on the prevalence of AMD (Smith et al., 2001)…………………………………………………………...… 14

Table 1.3 Longitudinal study of the visual acuity of people with non-exudative AMD (Sunness et al., 2002)…………………………. 18

Table 1.4 Types of reading for people with low vision (Whittaker and Lovie-Kitchin, 1993)……………………………………………. 27

Table 1.5 Summary of previous studies on the relationship between visual acuity and reading rate…………………………………………... 32

Table 1.6 Required field of view (FOV) for optimum reading rate from previous studies……..…………………………………………… 47

Table 1.7 Summary of previous studies on factors predicting reading rate of people with low vision……………………………………….. 51

Table 1.8 Overview of low vision aids …………………………………… 56Table 1.9 Summary of previous studies on the reading rates with and

without low vision aids on adults ……………………………….. 64Table 1.10 Summary of previous studies on training eccentric viewing

and/or the use of low vision aids.………………………………... 71 Table 2.1 Comparison of formats and layouts of the reading charts……….. 81Table 2.2 Subjects’ details…………………………………………………. 86Table 2.3 The mean of maximum reading rate (MRR) and critical print

size (CPS) of 10 repeated measures for each chart……………… 92Table 2.4 Correlation of two methods from different charts……………….. 100 Table 3.1 Previous formulae for calculating magnification………………... 109Table 3.2 Subject details…………………………………………………… 113Table 3.3 Reading measures for first and second visits……………………. 122Table 3.4 Acuity reserve achieving maximum reading rate for individual

subject……………………………………………………………. 126Table 3.5 Comparison of the reading performance of subjects at two visits. 127 Table 4.1 Subjects’ details………………………………………………….. 144Table 4.2 Subjects’ contrast sensitivity and visual field results……………. 153Table 4.3 Mean number of characters and standard words for passages of

different print sizes used for measuring reading rate with stand magnifiers………………………………………………………... 158

Table 4.4 Aperture width of practice stand required for each print size to give approximately 6 characters of horizontal field of view…….. 159

Table 4.5 Eschenbach illuminated stand magnifiers – Optical parameters measured…………………………………………………………. 161

Table 4.6 Summarised assessment of different groups at each visit……….. 163Table 4.7 Comparison of vision and reading measures for different groups

at week 2 for all groups………………..………………………… 173

List of tables

xii

Table 4.8 Summary of the mean vision measures across time for each

group……………………………………………………………... 178 Table 4.9 Summary of the measures of mean log reading rate (log wpm)

across time for each group………………………………………. 188 Table 4.10a Comparison of log reading rates with STM using sentences and

passages………………………………………………………….. 195 Table 4.10b Comparison of log reading rates without STM using sentences

and passages……………………………………………………... 196 Table 4.10c Comparison of log reading rates with and without STM using

sentences and passages…………………………………………... 196 Table 4.11 Percentage of subjects whose threshold print size with STM was

N8 or better at different visits……………………………...……. 202 Table 4.12 Correlations (Pearson r) between log reading rates with and

without STM and clinical measures without STM at week 2…… 206 Table 4.13 Summary of the multiple regression analyses…………………… 209 Table 5.1 Details of subjects’ vision measures…………………………….. 233 Table 5.2 Eschenbach illuminated stand magnifiers………………...……... 235 Table 5.3 Categorisations of magnifier movements………………………... 246 Table 5.4 Definitions of navigation errors…………………………………. 247 Table 5.5 Descriptive statistics of navigation and reading measures using

stand magnifier with and without a line guide…………………... 252 Table 5.6 Strategy of magnifier movements with and without a line guide

attached…………………………………………………………... 254 Table 5.7 Spearman correlations between subjective response of difficulty

in using magnifier and clinical measures without the attachment of line guide……….……………………………………………... 257

Table 5.8 Comparison of objective performance measures using stand magnifier with and without a line guide for subjects who preferred and rejected the line guide…………………………….. 262

Table A2.1 Results of power analysis: minimum number of subjects per

group for various levels of alpha and beta (for Chapter 2)…...…. 331 Table A2.2 Results of power analysis: minimum number of subjects per

group for various levels of alpha and beta (for Chapter 4)…...…. 338 Table A4.1 Results of the optical parameter measures for the stand

magnifiers used in this study……………………………….……. 347 Table A4.2 Optical parameters of the stand magnifiers used in this study,

calculated from the measured values for equivalent power (Fe) and image distance (l’)…………………………………………... 348

Table A6.1 The mean distance (cm) moved by the sensor between interval

markers along x- and y-planes…………………………………... 360 Table A6.2 The mean distance (cm) measured along x-and y-planes using

continuous lines………………………………………………….. 362

List of abbreviations

xiii

List of abbreviations

AMD Age-related macular degeneration

CCTV Closed circuit television

CFL Central field loss

CPS Critical print size

CS Contrast sensitivity

EVD Equivalent viewing distance

FOV Field of view

HHM Hand held magnifier

LVA Low vision aid

LVC Low vision clinic

MRR Maximum reading rate

RR Reading rate

PRL Preferred retinal locus

STM/s Stand magnifier/s

TPS Target print size

VA Visual acuity

VRC Vision rehabilitation centre

wpm words per minute

Statement of authorship

xiv

Statement of original authorship

The work contained in this thesis has not been previously submitted for a degree or

diploma at any other higher education institution. To the best of my knowledge and

belief, the thesis contains no material previously published or written by another

person except where due reference is made.

Signed: _______________________

Date: _________________________

Acknowledgments

xv

Acknowledgments

A journey is easier when you travel together. The completion of this PhD thesis is

the result of several years of work whereby I have been accompanied and supported

by many people. It is to my great delight that I now have the opportunity to express

my gratitude to all of them. First of all, this research was supported by an

International Postgraduate Research Scholarship granted by the Queensland

University of Technology (QUT). It has been a great pleasure to work in the Centre

for Eye Research and the university.

The first person I would like to thank is my principal supervisor, Associate Professor

Jan Lovie-Kitchin, who gave me invaluable advice and assistance. I am so grateful

for her thoughtful discussion on the experimental designs with me, as well as

guidance on the interpretation of the results and critical review on my writing. Jan

also showed her full support and continuing encouragement during the time of this

study. She is not only my supervisor for my PhD study, but also my mentor, who

played an important role to provoke and train my critical thinking.

I would also like to thank Dr Alex Bowers, my associate supervisor, who gave me

valuable advice on the planning of experiments, the interpretation of results and in

the preparation of this thesis and scientific papers. Alex’s sharing of her experience

as a PhD student and continuing encouragement were particularly important when I

encountered problems. I would also like to thank Dr Brian Brown, my associate

supervisor, for his valuable statistical advice and guidance in the preparation of this

thesis.

I am grateful to Eschenbach and European Eyewear for their generous provision of

the illuminated stand magnifiers to the participants in support of this study. I would

also like to thank Professor Mark Pearcy of the School of Mechanical,

Manufacturing and Medical Engineering, QUT, for the loan of the 3 SPACE Isotrak

system (Polhemus Navigation Sciences Division, Kaiser Aerospace, Vermont, USA).

Acknowledgments

xvi

I would also like to thank all the full-time and part-time staff and the postgraduate

students in the School of Optometry, the Centre for Eye Research and the Optometry

Clinic for their support and encouragement in many different areas.

I am also grateful to the following people for their valuable assistance and support:

Professor Ian Bailey of the School of Optometry, University of Berkeley for

supplying the two versions of Bailey-Lovie text reading charts; Professor Gordon

Legge of the Department of Psychology, University of Minnesota for supplying the

sentences used to produce different versions of the Bailey-Lovie text reading charts;

Professor Leo Carney; Associate Professor Joanne Wood and Dr Lawrence Stark, of

the Department of Optometry, QUT, for their helpful suggestions on thesis

preparation; Associate Professor David Atchison, Dr Chitra Avudainayagam and Dr

Avuday Avudainayagam, also from the Department of Optometry, for their useful

advice on optics; Mr Peter Hendicott, of the Centre for Eye Research, for valuable

statistical advice; Miss Jennifer Chen, of the Centre for Eye Research, for being a

good companion in our research studies; Professor Ian Bailey of the School of

Optometry, University of Berkeley and Dr Sue Leat of the School of Optometry,

University of Waterloo for being the external examiners and providing valuable

comments on thesis.

Most importantly, I would like to acknowledge the people who participated as

subjects in the various stages in this study. Their generous support and time are the

major drive for the completion of this study.

Last but not least, I would not have been able to finish my PhD without support from

my family and friends from Hong Kong and Australia, who have formed part of my

vision and taught me the good things that really matter in life. I feel a deep sense of

gratitude to my mum (Mrs Irene Oi-Yuk Cheong) for her love and care throughout

my life. Thanks also to my close friend, Mr. Bryan Kin-Lung Cheung for his mutual

support and encouragement.

Chapter 1 Literature review

1

CHAPTER 1

Literature review

1.1 Low vision ...................................................................................................... 3

1.1.1 Prevalence of low vision ......................................................................... 5 1.1.2 Gender distribution.................................................................................. 8 1.1.3 Age distribution....................................................................................... 8 1.1.4 Ocular disorders .................................................................................... 10

1.2 Age-related macular degeneration ............................................................ 10

1.2.1 Aetiology ............................................................................................... 11 1.2.1.1 Non-exudative (dry or atrophic AMD)........................................... 11 1.2.1.2 Exudative (wet or neovascular AMD)............................................ 12

1.2.2 Prevalence of AMD............................................................................... 13 1.2.3 Risk factors............................................................................................ 15 1.2.4 Progression of early AMD .................................................................... 16 1.2.5 Effect on vision functions...................................................................... 17

1.2.5.1 Visual acuity ................................................................................... 17 1.2.5.2 Contrast sensitivity ......................................................................... 19 1.2.5.3 Visual field...................................................................................... 20 1.2.5.4 Scotopic function ............................................................................ 20 1.2.5.5 Colour vision .................................................................................. 21 1.2.5.6 Glare recovery ................................................................................ 21

1.2.6 Reading performance............................................................................. 22 1.2.7 Mobility ................................................................................................. 22 1.2.8 Medical Treatment ................................................................................ 23

1.3 Reading with age-related macular degeneration (AMD)........................ 26

1.3.1 Reading with low vision........................................................................ 27 1.3.2 Assessment of reading performance...................................................... 27

1.3.2.1 Reading rate .................................................................................... 28 1.3.2.2 Comprehension ............................................................................... 29 1.3.2.3 Eye movements............................................................................... 29 1.3.2.4 Questionnaires ................................................................................ 30

1.3.3 Factors affecting reading performance.................................................. 30 1.3.3.1 Visual acuity ................................................................................... 31 1.3.3.2 Contrast sensitivity ......................................................................... 35 1.3.3.3 Central scotoma .............................................................................. 37 1.3.3.4 Saccade length ................................................................................ 39 1.3.3.5 Age.................................................................................................. 41 1.3.3.6 Text variables ................................................................................. 42

1.3.4 Predicting reading performance from vision variables ......................... 49


2

1.4 Reading rehabilitation for people with age-related macular

degeneration (AMD) ..................................................................................50

1.4.1 Types of magnification in low vision ....................................................53 1.4.1.1 Relative size magnification (RSM).................................................53 1.4.1.2 Relative distance magnification (RDM) .........................................54 1.4.1.3 Relative image magnification (RIM) or angular magnification......55 1.4.1.4 Projection magnification .................................................................55

1.4.2 Determination of required magnification ..............................................55 1.4.3 Optical and electronic aids.....................................................................56

1.4.3.1 High near addition...........................................................................57 1.4.3.2 Hand-held magnifier .......................................................................57 1.4.3.3 Stand magnifier ...............................................................................59 1.4.3.4 Near telescope .................................................................................61 1.4.3.5 Closed circuit television (CCTV) ...................................................61

1.4.4 Non-optical aids.....................................................................................62 1.4.4.1 Lighting...........................................................................................62 1.4.4.2 Reading stand..................................................................................63

1.4.5 Reading performance with low vision aids............................................63

1.5 Low vision training and rehabilitation for reading..................................70

1.5.1 Training..................................................................................................70 1.5.1.1 Improvement of visual skills...........................................................70 1.5.1.2 Training for the use of low vision aids ...........................................75

1.6 Objectives of this study...............................................................................77


3

1.1 Low vision

Traditionally, visual performance is classified as legally “seeing” and legally

“blind”. This definition is artificial and discriminatory, neglecting the person whose

vision is substantially impaired, but still retains some visual ability for daily

activities. The term “Low Vision” was officially incorporated in the ninth revision of

the International Classification of Diseases (ICD-9) in 1974 by the World Health

Organization (WHO) (1980). For the purpose of this thesis low vision is defined as

vision loss due to ocular disease that is not correctable with glasses, contact lenses or

surgical treatments. Although visual impairment has a slightly different definition

(see below), the terms low vision and visual impairment are often used

interchangeably.

People with low vision require not only medical assistance to treat their eye diseases

but also rehabilitative assistance to address the consequences of their vision loss.

Based on the tenth revision of the International Classification of Diseases (ICD-10),

an aetiological framework to formalise the relationship between disorder and

impairment, activity and participation limitation has been recommended (World

Health Organization, 2001 (a)). This framework describes different aspects of the

visual and functional problems which are consequences of ocular disorders

(Colenbrander, 2000).

Ocular disorder is defined as “deviation from the normal structure or physiology of

the visual system” (Colenbrander, 1977). Any physiological or pathological disorder

of the visual pathway can result in visual impairment. To cite an example, age-

related macular degeneration is a pathological deviation from the normal structure of

the retina primarily due to ageing (Table 1.1). Visual impairment refers to the

consequences of an ocular disease and tends to be measured in terms of changes in

vision "function" relative to the normal range of vision function in healthy normal

eyes (Colenbrander, 1977). For a person with age-related macular degeneration,

vision functions such as visual acuity and visual field are impaired. Ocular disorder

and visual impairment describe the organ’s condition, which are medically related. In

contrast, activity limitation and participation restriction are the terms that refer to the


4

whole person. Activity limitation is a consequence of the visual impairment, such

that an individual’s ability to perform a particular visual task is limited. For example,

a person may have difficulty reading newspaper print with his/her habitual reading

lenses due to the visual impairment (which is the consequence of having age-related

macular degeneration). Although the term “activity limitation” has replaced

disability, the term “disability” is still more commonly used in the rehabilitation

literature. Because of the limitation in performing daily activities, people with low

vision attend low vision clinics (LVC) seeking assistance to alleviate the disability.

For example, a person with reading difficulty requires a low vision aid from the LVC

to achieve his/her reading goal. As a consequence of the activity limitation, social,

economic and psychological life of the person may be affected leading to

participation restriction (Colenbrander, 1977). This may be reflected as a loss of

physical independence, self-esteem or employment (Lovie-Kitchin and Bowman,

1985). In order to manage vision loss and its associated limitations, multi-

disciplinary rehabilitation services to minimise the activity limitation and to reduce

the participation restriction through appropriate interventions are required (Table

1.1)

Table 1.1 The relation between disease, impairment, activity limitation and participation restriction for vision (Dickinson, 1998)

halla

This table is not available online. Please consult the hardcopy thesis available from the QUT Library


5

Although the WHO has defined low vision in terms of impairment and activity

limitation (World Health Organization, 1980 (b)), there is no consensus for defining

low vision for the purpose of eligibility for financial benefitsa from government or

related associations (Leat et al., 1999). As a consequence, there is a wide range in the

levels of vision which constitute low vision among different countries for this

purpose.

In Australia, there is no separate definition of low vision for legal purposes. Thus

people with low vision have to be registered as "legally blind" before they can obtain

pension or benefits from the government. People who have best corrected visual

acuity (Snellen scale) of worse than 6/60 in both eyes, or a visual field of less than

100 of fixation in the better eye irrespective of corrected visual acuity, or a

combination of visual defects resulting in the same degree of visual impairment due

to the reduction of visual acuity and constriction of visual field, are eligible to be

registered as legally blindb.

1.1.1 Prevalence of low vision

The WHO estimated in 1995 that there were 110 million people worldwide with low

vision (Thylefors et al., 1995). Of these, only 40 to 45 million people were blind

(Thylefors et al., 1995). This indicates that there is a large number of people with

vision between near normal and blindness, which can be termed as “low vision”. The

prevalence of low vision increased to approximately 180 million in 2001 (World

Health Organization, 2001 (b))c. This indicates that the number of people with low

vision is increasing significantly. The WHO Task Force on Data on Blindness

reported that 90% of all low vision was found in developing countries with 75% of

world blindness occurring in Asia and Africa (Thylefors et al., 1995).

a These benefits include income tax allowance or pension, transport benefits or other assistance for individuals whose vision prevents them from seeing well enough to work. b http://www.facs.gov.au/guide/ssguide/11p210.htm (December, 2002) c http://www.who.int /pbd/pbl/pbl_home.htm (September, 2001)


6

The prevalence of visual impairment in different regions varies among different

epidemiological studies. In Australia, there have been two population studies

estimating the prevalence of visual impairment. In the Blue Mountains Eye Study

(BMES), 3654 people aged 50 years or older were examined of whom 5.3% had

visual impairment (Attebo et al., 1996). In the Melbourne Visual Impairment Project

(MVIP), the prevalence of visual impairment was 1.32% among 3268 people aged 40

years or older (Taylor et al., 2000). There are two possible explanations for the

discrepancy in the findings between the two studies. Despite similar numbers of

participants in each study, the age distribution of the samples was different. Fifty-

five percent of people participating in the MVIP were aged 60 years or below while

45% of people who participated in the BMES were aged below 65 years. Numerous

studies indicate that the prevalence of visual impairment increases with age. As more

people of younger age participated in the MVIP, the prevalence of visual impairment

reported would be expected to be smaller. Additionally, the definition of visual

impairment in each study was different. The MVIP defined visual impairment as best

corrected visual acuity in the better eye of worse than 6/12. In contrast, people with

best-corrected visual acuity in the better eye of 6/12 were included in the calculation

of prevalence of visual impairment in the BMES.

In the United States of America, the National Institute on Disability and

Rehabilitation Research (NIDRR) (1993) estimated that the number of people with

visual impairment ranged from 6 to 11.4 million. NIDRR showed that vision loss

was the third most common chronic disease. The Lighthouse National Survey on

Vision Loss (The Lighthouse Inc, 1995) found that 16.5 million (17%) Americans

aged 45 years or older reported some form of visual impairment with their glasses or

contact lenses. This higher prevalence is probably because of the broader definition

of visual impairment used in the Lighthouse National Survey. The Baltimore Eye

Survey (Katz and Tielsch, 1996) interviewed 6850 residents aged 40 years and older

found that 25% reported limitations in activities due to visual impairment. The

Beaver Dam Eye Study (BDES) (Klein et al., 1992) examined 4897 participants aged

between 43 and 86 years and reported that 5.23% of people had visual impairment

(best-corrected visual acuity in the better eye of 6/12 or worse). This finding agreed

remarkably with that reported in the BMES in Australia for a similar age profile.


7

However, the prevalence of visual impairment found in both the BDES and BMES

was relatively lower compared to those studies which estimated the prevalence of

visual impairment subjectively by asking the subjects to report any limitations of

their daily activities.

In the United Kingdom, data on the prevalence of low vision are based on

registrations rather than surveys. The Government Statistical Service (1994) reported

that more than 265,000 people of all ages were visually impairedd. This figure was

considerably less than the finding from a Royal National Institute of the Blind

(RNIB) survey, which estimated that more than 757,000 people were visually

impaired (Bruce et al., 1991). However, a more recent report from the Royal

National Institute for the Blind (1997) indicated that an estimated 1.1 million people

of all ages (1.82%) were blind or partially sighted. According to the figures provided

by these studies, the prevalence of visual impairment is increasing.

In some developed Asian countries, the prevalence of low vision appears to be less

than that in the developed Western countries. According to a survey conducted in

Malaysia in 1996, the prevalence of low vision was 2.44% (Zainal et al., 2002). A

similar figure was found in a Taiwanese population-based survey which reported that

2.72% of people aged 50 years or older had visual impairment (best-corrected visual

acuity in the better eye of worse than 6/18) (Liu et al., 2001).

In developed countries, where more ophthalmological expertise is available, more

resources and awareness have been directed towards the treatment of eye diseases

and vision rehabilitation for low vision. Therefore, the prevalence of visual

impairment in the developed world tends to be lower than that in developing

countries. Thylefors et al. (1995) estimated that by the year 2020, the global

population of visual impaired would have increased threefold. At least 80% of these

people with visual impairment will be in developing countries. This percentage may

increase up to 90% (World Health Organization, 1997), as cataract surgery is

generally not available, or only available to the elite few who live in the developing

d People with best corrected visual acuity (Snellen scale) worse than 6/60 with full field, or worse than 6/24 with moderate field constriction, or 6/18 or better with gross field defect are eligible to register as visually impaired (Dickinson, 1998).


8

countries (Taylor, 2001). Therefore, treatment as well as rehabilitation services are

essential for these developing countries.

1.1.2 Gender distribution

Some studies have found that visual impairment in adults is significantly more

prevalent in females than males (Attebo et al., 1996; Elliott et al., 1997a; Taylor et

al., 1997). In addition, low vision clinics report a greater proportion of females seeking

services (Knave et al., 1984; Nilsson, 1986; Nilsson and Nilsson, 1986; Nilsson, 1988).

The cause for the higher prevalence of visual impairment in females is unclear.

Mitchell (1991) suggested that this might be due to a longer life expectancy in women

than men. As life expectancy of females increases, the associated increase in visual

impairment with increasing age increases.

1.1.3 Age distribution

Rates of visual impairment sharply increase with age, particularly in people beyond 65

years of age (Tielsch et al., 1990; Klein et al., 1991; Salive et al., 1992; Leonard,

1999). The BMES (Attebo et al., 1996) in Australia found that the prevalence of

vision loss increased with age from 0.8% of persons aged 49 to 54 years to 42% of

persons 85 years of age or older (Figure 1.1). The Lighthouse National Survey on

vision loss (The Lighthouse Inc, 1995) found that only 15% of American people aged

45 to 64 years reported some form of visual impairment, however, the percentage

increased to 17% and 30% in the population aged 65 to 74 years and people beyond

74 years respectively. The result from the BDES which was a longitudinal study

(Klein et al., 2001) also agreed with the results from previous cross-sectional studies

(The Lighthouse Inc, 1995; Thylefors et al., 1995; Attebo et al., 1996). The BDES

(Klein et al., 2001) monitored 3684 American people for ten years and found a

significant decrease in visual acuity among people of 75 years and older. The

incidence of vision loss was consistent with the higher frequency of age-related eye

diseases such as age-related macular degeneration (refer to Table 1.2) and cataract


9

among people aged 75 years and older. Older people (> 75 years old) were 15 times

more likely to have impaired vision than people less than 75 years of age (Klein et

al., 2001). The data collected from the Government Statistical Service (1994) and

Royal National Institute for the Blind (1997) in the United Kingdom supported the

increased prevalence of visual impairment with increased age. They found that the

percentage of older people (≥ 75 years) estimated to be blind or low vision was

higher than the percentage of people to be blind or low vision in the younger-age

groups.

0

5

10

15

20

25

30

35

40

45

49-54 55-64 65-74 75-84 85+

Age groups (years)

Prev

alen

ce (%

)

Figure 1.1 Prevalence of visual impairment by age from the Blue Mountains Eye Study (1996) in Australia.

Rates of visual impairment sharply increase with age.

These results emphasize the fact that, in the "developed" world, the vast majority of

visual impairment is acquired in the later years of life. Around the world, the number

of older people will double in the next 20 years which is likely to result in a

significant increase in the number of people with visual impairment (Taylor, 2001).

Demand for low vision rehabilitation will therefore increase dramatically in the near

future.


10

1.1.4 Ocular disorders

Many ocular or neurological disorders, such as age-related macular degeneration

(AMD), cataract, glaucoma, diabetic retinopathy, optic atrophy, retinitis pigmentosa

(RP), trachoma, inherited retinal disorders can result in low vision. The main causes

of vision loss in developed countries are AMD, cataract, glaucoma, and diabetic

retinopathy (Grey et al., 1989; Klein et al., 1992; Sterns, 1994). Among these

diseases, AMD is the most common cause of permanent vision loss in developed or

industrialised countries (Lovie-Kitchin and Bowman, 1985; Thompson et al., 1989;

Howe, 1995; Starr et al., 1998; Van Newkirk et al., 2001; Kocur and Resnikoff, 2002;

Munoz and West, 2002). For developing countries, cataract is the most common cause

of visual impairment (Liu et al., 2001; Zainal et al., 2002). Importantly, unlike AMD,

cataract is a treatable eye disease.

Among registered blind and low vision populations, AMD is the most common cause

of visual impairment (Lovie-Kitchin and Bowman, 1985; Cooper, 1990; Attebo et

al., 1996), particularly in people aged 75 years and older. It is also the most common

cause of new cases of visual impairment (Prevent Blindness America, 1998). A

number of studies have predicted that the prevalence of AMD will increase and the

visual impairment associated with this disorder will double in the next two decades

(Evans and Wormald, 1996; Taylor et al., 1997; Weih et al., 2000; Taylor, 2001).

1.2 Age-related macular degeneration

Age-related macular degeneration (AMD) is the most common cause of severe

central vision loss among people aged 50 years and over in Australia (Mitchell et al.,

1995; Attebo et al., 1996), the United States (Klein et al., 1995; Klein et al., 1997;

Prevent Blindness America, 1998), the United Kingdom (Evans, 1995; Evans and

Wormald, 1996; Gass, 1997) and other developed countries (Klaver et al., 1998;

Klaver et al., 2001). Age-related macular degeneration is a disease involving typical

lesions in the macula in older people, which is not caused by infections or

inflammations (Evans, 2001). The disease is characterised by the presence of some

degree of vision loss in association with drusen and atrophy of the retinal pigment


11

epithelium or changes associated with subretinal neovascularisation among

individuals over 50 years old. The condition is usually bilateral, although both eyes

may not be affected to the same extent (Klein, 1999). This section presents an

overview of the clinical manifestations of AMD and its effect on vision functions

and daily activities.

1.2.1 Aetiology

Even though AMD is the leading cause of blindness in most western industrialised

countries, its aetiology is not clearly understood (Hyman, 1992). At present, there are

no medical interventions that can prevent the incidence or progression of AMD.

Some medical treatments are available to slow the progression and so reduce the risk

of severe vision loss (see section 1.2.8).

AMD is a disease of the macula photoreceptors, retinal pigment epithelium (RPE)

and Bruch’s membrane. It is generally classified into either one of the two distinct

forms - non-exudative (dry or atrophic) form or exudative (wet or neovascular) form.

Approximately 10% of the cases of AMD are of the exudative type (Dickman, 1982;

Hyman et al., 1983; Ferris et al., 1984; Chisholm, 1996; Varmus, 1997) and account

for 80 to 90% of severe vision loss or blindness in patients with AMD (Bressler et

al., 1988; Varmus, 1997).

1.2.1.1 Non-exudative (dry or atrophic AMD)

Approximately 90% of AMD is non-exudative (Dickinson and Rabbitt, 1991;

Chisholm, 1996), resulting in moderate loss of vision (Lewis, 1992). Its aetiology is

not well understood but it usually occurs in one eye first and later in the other eye. Its

progression may be slow with a number of signs identified during fundus

examination.

The earliest clinically detectable feature of AMD is the appearance of asymptomatic

yellow excrescences beneath the RPE called drusen, which are characterised by

sharply delineated round or oval areas of hypopigmentation and are distributed at the


12

posterior pole close to macula region. Drusen are an accumulation of

mucopolysaccharides, lipids and metabolic by-products from photoreceptors, which

are deposited onto Bruch’s membrane (Lovie-Kitchin and Bowman, 1985). The

exact role of drusen in the pathogenesis of AMD is still unclear (Edwards et al.,

1999). Normally, drusen alone do not cause vision loss, but result in mild

metamorphopsia which in turn reduces reading rate (Legge et al., 1985 (b);

Whittaker et al., 1988; Legge, 1991; Rumney and Leat, 1994; Bullimore and Bailey,

1995) and contrast sensitivity (Brown and Kitchin, 1983; Brown et al., 1986 (c);

Midena et al., 1997). When clusters of drusen start to accumulate, round or oval

patches of retinal atrophy may form, resulting in partial scotoma in central vision and

gradual vision loss (Swann and Lovie-Kitchin, 1990).

In addition to drusen, areas of increased pigment (hyperpigmentation) in the outer

retina or choroid or areas of decreased pigment (hypopigmentation) of the RPE are

the other clinical signs of AMD (Zimmerman, 1992; Bird et al., 1995).

1.2.1.2 Exudative (wet or neovascular AMD)

Exudative AMD is a quickly progressing disease in which subretinal neovascular

membranes, consisting of proliferations of new fragile vessels which begin to grow

from the choriocapillaris. The blood vessels grow through Bruch’s membrane and

into the sub-RPE space and later into the subretinal space and haemorrhage into the

macula and foveal areas. This may give the clinical appearance of sub-retinal fluid,

macula oedema, retinal, subretinal or sub-RPE haemorrhages which may in turn

result in formation of choroidal neovascular membranes (Bressler et al., 1988;

Edwards et al., 1999). Complications such as haemorrhagic RPE detachment,

sensory detachment, vitreous haemorrhage or exudative retinal detachment ending in

disciform scarring may result. For this reason, immediate medical treatment for

choroidal neovascularisation is essential where possible to reduce the growth of the

subretinal neovascular membranes.


13

1.2.2 Prevalence of AMD

A number of population-based studies have reported different estimates of the

prevalence of AMD (Table 1.2). The discrepancies are mainly due to the differences

in the definition of AMD, methods of classifying AMD, interpretation of the

classification and characteristics of the sample population.

In the major population studies, the prevalence of non-exudative AMD has been

estimated at 0.4-0.7% for all ages (Table 1.2). For example, 0.45% and 0.44% of

people in the BMES (Mitchell et al., 1995) and BDES (Klein et al., 1992)

respectively were reported to have non-exudative AMD. The higher prevalence of

non-exudative AMD reported in the Rotterdam Study (0.66%) (Vingerling et al.,

1995) was probably because more people of older age were recruited in this study

(55 –106 years). Research studies have widely supported the positive correlation

between the prevalence of AMD and age (Klein et al., 1992; Mitchell et al., 1995;

Van Newkirk et al., 2000; La Heij et al., 2001).

The prevalence of exudative AMD in the BMES was 1.2% (Mitchell et al., 1995)

which was slightly higher than that found in the BDES (0.9%) (Klein et al., 1992) or

the Rotterdam Study (0.72%) (Vingerling et al., 1995). Interestingly, the prevalence

of exudative AMD in the Rotterdam study (1995) was relatively lower than that in

the BMES and BDES. Smith et al. (2001) attributed the lower prevalence of

exudative AMD in the Rotterdam Study to the difference in the photographic

documentation. Because of the reduced photo quality in the Rotterdam Study for

defining the appearance of exudative AMD, it was difficult to identify the clinical

signs during the interpretation of the fundus photos. As a consequence, some

participants with less definite signs shown on the photos might have been

inappropriately classified, resulting in a lower prevalence of exudative AMD (Smith

et al., 2001). In addition, a lower prevalence of exudative AMD (0.5%) was reported

in the Chesapeake Bay study (Bressler et al., 1989) probably because of the different

age profile of subjects.


14

Table 1.2 Epidemiological studies on the prevalence of AMD (Smith et al., 2001)

halla



15

1.2.3 Risk factors

The prevalence of AMD increased significantly with age in all studies (Klein et al.,

1992; Mitchell et al., 1995; VanNewkirk et al., 2000; La Heij et al., 2001) (Table 1.2

and Figure 1.2). Age has been found to be the most important risk factor for AMD.

For people aged 90 years and above, the risk of developing AMD is 8 to 10-times

higher than for people aged 50 years (Evans, 2001). Similarly, the BDES found that

people between 75 and 84 years of age had a significantly higher prevalence of AMD

than people 43 to 54 years of age (Klein et al., 1992). Edwards and colleagues

(1999) pointed out that over one third of people aged 90 years or above were affected

by AMD.

02468

101214161820

55-64 65-74 75-84 85+

Age group (years)

Prev

alen

ce (%

)

BeaverDam EyeStudy

BlueMountainsEye Study

RotterdamStudy

Figure 1.2

Prevalence of AMD by age.

The prevalence of AMD increased significantly with age

In addition to age, some studies have indicated that AMD is more prevalent among

whites than blacks (Gregor and Joffe, 1978; Klein and Klein, 1982; Friedman et al.,

1999; Friedman, 2000), in particular the more severe form of AMD (Friedman et al.,

1999). In contrast, other studies indicate no statistically significant difference in the

prevalence of AMD among different ethnic groups (Bressler and Bressler, 1995;

Hawkins et al., 1999). As no conclusion can be drawn from these studies, further


16

research on larger sample sizes is needed to confirm whether race is a risk factor

associated with AMD.

People with a family history of AMD are at increased risk of having this eye disease

(Hyman et al., 1983; Smith and Mitchell, 1998). Hyman et al. (1983) reported that

family history of AMD was the strongest predictor for the presence of AMD. Other

family based studies also concluded that siblings of an affected person had a higher

risk of developing AMD compared to the general population (Silvestri et al., 1994;

Yoshida et al., 2000). Moreover, a few studies have examined the existence of

genetic predisposition for AMD by using twin siblings (Dosso and Bovet, 1992;

Meyers et al., 1995; Gottfredsdottir et al., 1999). Results show a higher concordance

of AMD between monozygotic twins than dizygotic twins (Meyers and Zachary,

1988; Dosso and Bovet, 1992; Grizzard and Beck, 1994; Meyers et al., 1995;

Gottfredsdottir et al., 1999). All these studies indicate that hereditary factors are

important in the aetiology of AMD, but the mechanism for the genetic predisposition

is not known.

1.2.4 Progression of early AMD

AMD is a progressive eye disease resulting in vision loss over time. The progression

of non-exudative AMD is slow and gradual compared to exudative AMD. A number

of epidemiological studies have investigated the progression of AMD by monitoring

the change in vision function for a number of years.

In the BDES, Klein and colleagues (1997) defined the participants with small hard

drusen (< 63 µm in diameter) but no pigmentary abnormalities and no reduction in

visual acuity as having early AMD and monitored their vision function for 5 years.

Among 3684 participants, they reported that 11.7% developed late AMD (presence

of soft indistinct drusen and pigmentary abnormalities) within 5 years, of which

7.1% were exudative AMD and 4.6% were non-exudative AMD. Similarly, the

Chesapeake Watermen study (Bressler et al., 1995) followed 483 patients with AMD

for 5 years and found that 10% of people progressed to exudative AMD, which was

the more severe form of AMD leading to significant visual impairment (Bressler et


17

al., 1995). Klaver and co-workers (2001) followed 1244 people with early AMD and

found that in 21.5% of people the AMD had progressed to a more severe stage within

two years. The progression rate from this study was significantly higher than the

BDES (Klaver et al., 2001) or the Chesapeake Watermen Study (Bressler et al.,

1995).

People with unilateral AMD had an incidence rate of 29% developing AMD in the

fellow eye within two years (Klaver et al., 2001). This figure was comparatively

higher but close to the percentage reported in the BDES (Klein et al., 1997) in which

22% of the second eye developed AMD within 5 years. The discrepancy in the

incidence rate may be attributed to the smaller number of older participants at the

baseline in the BDES (Klein et al., 1997) than that in the Rotterdam study (Klaver et

al., 2001).

1.2.5 Effect on vision functions

As AMD is a progressive condition, there can be a wide range of vision findings,

depending on the severity. Apart from reduced visual acuity, there are other deficits

of vision function associated with AMD, such as reduced contrast sensitivity,

presence of central scotoma, impaired dark adaptation, abnormal scotopic sensitivity

and impaired colour discrimination (Eisner et al., 1987; Eisner et al., 1991).

1.2.5.1 Visual acuity

Typically, visual acuity decreases only slightly, in a range of 0.3 to 0.5 logMAR

(6/12 to 6/18) in the earliest stages of AMD (Fischer, 2000). Over time, the decrease

in visual acuity progresses slowly as the geographic atrophy of the RPE progresses

(Lovie-Kitchin and Bowman, 1985; Schuchard et al., 1999).

Sunness and colleagues (1997; 1999) monitored the vision functions of a group of

AMD people with non-exudative geographic atrophy for two years. They divided the

participants into two groups – group 1 subjects had visual acuity of better than 0.4

logMAR (better than 6/15) while group 2 people had visual acuity between 0.4 and


18

1.0 logMAR (6/15 and 6/60). After two years, the visual acuity for half of the

subjects from group 1 declined to 0.70 logMAR or worse (6/30 or worse), a

reduction of 0.3 log-unit or more. In addition, visual acuity for another one quarter of

the subjects from group 1 declined to 1.0 logMAR or worse (≥ 6/60), a reduction of

0.6 log unit or more (Sunness et al., 1997). In contrast, only one fifth of the people

from group 2 showed a reduction in visual acuity by 0.3 log unit or more in the past

two years. This indicated that the progression rate of AMD and the consequent

vision deterioration for people with better initial visual acuity was higher than for

people with moderate visual acuity. The results from two other studies by Sunness et

al. (1999; 2002), in which vision function was monitored for longer durations,

confirmed this result. Eyes with better visual acuity had a higher rate of acuity loss

than eyes with median visual acuity (Sunness, 1999; Sunness et al., 2002) (Table

1.3).

Table 1.3 Longitudinal study of the visual acuity of people with non-exudative AMD (Sunness et al., 2002)

The progression rate (and reduction in visual acuity) for exudative AMD is different

from that of non-exudative AMD. It progresses suddenly with sensory haemorrhagic

detachment and fibrous disciform scarring in the macula region (Swann and Lovie-

Kitchin, 1990; Bird et al., 1995) resulting in severe vision loss (Lovie-Kitchin and

Bowman, 1985; Alexander, 1993; Vinding, 1995). It is common for visual acuity of

halla



19

patients with exudative AMD to reduce to less than 1.0 logMAR (6/60). Scupola and

colleagues (1999) monitored patients with exudative AMD with subretinal

haemorrhage over two years and discovered that in 80% of the eyes the Snellen

visual acuity reduced by approximately five times, which implied the vision reduced

from 1.1 logMAR to 1.8 logMAR on the logMAR scale. They further pointed out

that the recurrence of haemorrhages due to exudative AMD had a major impact on

the final visual acuity. This indicates that unless surgical treatment is applied,

patients with exudative AMD and submacular haemorrhages have a poor prognosis.

Because of the reduction in visual acuity due to AMD, rehabilitative assistance to

improve these patients’ resolution ability is necessary in order to perform the tasks

necessary for survival and for enjoying life.

1.2.5.2 Contrast sensitivity

Contrast sensitivity reflects a person’s ability to differentiate fine details in the

environment at low and high contrasts. Contrast sensitivity has been shown to be

impaired in patients with AMD even when the disease is in the early stage (Brown

and Garner, 1983; Kleiner et al., 1988; Collins and Brown, 1989). The magnitude of

contrast sensitivity reduction varies according to the type of AMD and its severity.

For early non-exudative AMD, contrast sensitivity is significantly reduced at high

spatial frequencies (Kleiner et al., 1988; Midena et al., 1997; Sunness et al., 1997).

In contrast, for people who have late-stage AMD, the reduction in contrast sensitivity

occurs across all spatial frequencies and the peak contrast sensitivity is shifted to

lower spatial frequencies (Woo, 1985; Hampton and Nelsen, 1992; Vinding, 1995).

For patients with AMD and reduced contrast sensitivity, the qualitative aspects of

vision become worse leading to limitations in activities such as mobility and reading.


20

1.2.5.3 Visual field

Central visual field sensitivity starts to decrease for patients with AMD in the early

stage. According to the phase of the disease, there are varying degrees of central

field loss and different sizes and locations of scotomas result. As the geographic

atrophy of RPE usually starts in the perifoveal region, a typical para-central field loss

with foveal sparing can be plotted in the early phase of AMD (Swann and Lovie-

Kitchin, 1990). A para-central field defect (or scotoma) is generally found within the

central 20° diameter of the visual field where light sensitivity is reduced with no

foveal involvement. Although the visual acuity is still reasonably good for patients

with paracentral scotoma, their reading rate is dramatically affected (Sunness et al.,

1997). Reasons for the reduction in reading rate are discussed in section 1.2.6 and

1.3.3.3.

As the AMD progresses, the extent of the central visual field defect grows both

inward and outward, forming a horseshoe-shaped and then a ring shaped scotoma

around the fovea (Schuchard et al., 1999). Finally, it coalesces to produce an

absolute central visual field defect in which the central retinal areas are no longer

sensitive to objects of any light intensity (Lovie-Kitchin and Bowman, 1985). For

patients with exudative AMD, presence of sub-retinal fluid and neovascularisation

may result in absolute central scotoma that may extend beyond that central 20°.

1.2.5.4 Scotopic function

Studies have shown scotopic dysfunction in the early stage of AMD (Brown and

Kitchin, 1983; Brown et al., 1986 (a); Steinmetz et al., 1993; Owsley et al., 2000).

Patients with early AMD show more deficits in the scotopic (rod-mediated)

sensitivity than the photopic (cone-mediated) sensitivity. As a consequence, people

with AMD may complain of poor night vision because of the delay in dark

adaptation (Steinmetz et al., 1993). Brown and Kitchin (1983) found that the results

of a test of dark adaptation correlated well with the adaptation difficulties reported

by patients with AMD.


21

1.2.5.5 Colour vision

Reduction of colour discrimination in eyes with early AMD has been well

documented (Bowman, 1980; Collins, 1986; Eisner et al., 1987; Eisner et al., 1991;

Cheng and Vingrys, 1993). S-cones (short-wavelength cones), one of the three types

of photoreceptor cells, are the first group of cells to degenerate in early macula

disease (Hart, 1987; Nork, 1995) resulting in a blue-yellow colour defect.

Conversely, Midena and colleagues (1997) did not find impaired colour vision

function in patients with early AMD, but this was probably because the subjects in

their study had very early AMD compared to subjects in the other studies. Lovie-

Kitchin and Brown (1986) compared the reaction times to red lights for people with

and without AMD. They showed that subjects with AMD had prolonged reaction

time compared to the age-matched subjects with normal vision (Lovie-Kitchin and

Brown, 1986). If the vision associated with the eye disease deteriorates, colour

discrimination becomes worse. Measuring colour vision function is thus a useful

functional tool to differentiate early AMD.

1.2.5.6 Glare recovery

A number of studies have reported significant delays in glare recovery for patients

with AMD. This functional change is present even in people with early AMD

(Collins and Brown, 1989; Midena et al., 1997; Richer, 1999). Marshall (1985)

attributed the reduction in glare recovery function to the dysfunction of the

photoreceptor membranes in AMD. Midena et al. (1997) found a high correlation

between the functional deterioration in glare recovery and a worsening of the typical

retinal lesions characteristic of early AMD. They further pointed out that the

reduction in macula recovery function was due to the geographic atrophy of the RPE

and focal hyperpigmentation in AMD (Midena et al., 1997). Similar to the

assessment of colour vision function, monitoring the function of glare recovery is a

sensitive indicator of functional loss in patients with early AMD.

As a consequence of the impairments in vision function arising from AMD, activities

of every day visual tasks, such as reading and mobility can be severely limited.


22

1.2.6 Reading performance

In the literature, numerous studies have reported that the status of central vision is

one of the most important factors affecting reading performance (Legge et al., 1985

(a); Legge et al., 1985 (b); Whittaker et al., 1988; Bailey et al., 1989; Legge, 1991;

Whittaker and Lovie-Kitchin, 1993; Rumney and Leat, 1994; Bullimore and Bailey,

1995). Therefore, any loss of central visual acuity (Legge et al., 1985 (b);

Timberlake et al., 1986; Whittaker et al., 1988; Bullimore and Bailey, 1995) and/or

central visual field (Legge et al., 1985 (b); Cumming et al., 1991; Whittaker and

Lovie-Kitchin, 1993; Bullimore and Bailey, 1995) can severely affect reading.

Patients with AMD may maintain reasonably good visual acuity if the eye disease is

still at its early stage with no foveal involvement. However their reading rate can be

dramatically affected if there is a paracentral scotoma in the central visual field

(Sunness et al., 1997). Sunness et al. (1997) assessed the reading rate of AMD

people with geographic atrophy with visual acuity of 0.4 logMAR and better (≥

6/15). They found that 50% of subjects had maximum reading rates of less than 100

words per minute (wpm) while 17% had less than 50 wpm. This indicates that the

parafoveal scotoma due to the geographic atrophy can significantly compromise

reading rate even when the visual acuity is good. Therefore, it is not surprising that

the reading rate for people with AMD is comparatively slower than people with other

low vision causes (Legge et al., 1985 (b); Legge et al., 1992).

As the central scotoma and impaired macula function progress (section 1.2.5.3),

instead of using the non-functioning fovea, the visual system of the majority of

patients with AMD selects a preferred retinal location in the paracentral area to

perform visual tasks - “preferred retinal locus (PRL)” (von Noorden and Mackensen,

1962; Cummings et al., 1985; Timberlake et al., 1986; Whittaker et al., 1988). A

more detailed review of reading with low vision and with AMD in particular is given

in section 1.3 and training of eccentric fixation is discussed in section 1.5.1.1.

1.2.7 Mobility

A number of studies have shown that mobility is compromised in people with AMD

(Brown et al., 1986 (b); Lovie-Kitchin et al., 1990; Kuyk et al., 1998; Kuyk et al.,


23

1998 (b)). Brown et al. (1986 (b)) reported that subjects with AMD experienced

greater problems with mobility compared to subjects with normal vision under low

luminance. This finding was supported by a recent study by Hassan et al. (2002)

which found that the size of the binocular central scotoma was highly related to the

mobility performance. They further pointed out that visual field and contrast

sensitivity measures were predictive variables of mobility performance of patients

with AMD. Rehabilitative training programs for orientation and mobility have been

suggested for people who are visually impaired and have difficulty moving with

confidence and safety (Zimmerman, 1992).

1.2.8 Medical Treatment

Currently there is no treatment for patients with the more common dry or non-

exudative AMD (Dickinson and Rabbitt, 1991; Chisholm, 1996). Although medical

treatments are available for exudative AMD, only a minority of patients benefit from

these treatments. Most of these treatments are aimed at slowing the progression of

AMD and reducing the risk of further vision loss. A brief outline of the available

treatments for AMD and a review of some of the studies for each treatment are

discussed below.

Thermal laser photocoagulation of new blood vessels remains the standard treatment

for patients with exudative AMD. The purpose of this treatment is to slow down the

vision loss due to the subretinal neovascularization, but it does not cure the disease

itself. Although laser photocoagulation causes damage to the overlying

photoreceptors on the retina which results in an immediate loss of vision, large

multi-centre clinical studies have found that the majority of patients benefit from this

treatment, experiencing less vision deterioration over time compared to the untreated

patients (Macular Photocoagulation Study Group, 1986; 1991; 1994 (a); 1994 (c);

1996). However, the clinical trials show that there is at least a 50% chance that the

choroidal neovascularisation will recur within two years after the treatment and about

half of the treated patients experience some initial leakage beneath the centre of the

fovea (Macular Photocoagulation Study Group, 1991; Macular Photocoagulation

Study Group, 1994 (a); Macular Photocoagulation Study Group, 1994 (b); Macular


24

Photocoagulation Study Group, 1994 (c)). The main disadvantage of laser

photocoagulation is the resultant scars on the retina which lead to corresponding

defects in the central field.

The use of ionising radiation to reduce regression and/or to promote inactivation of

the choroidal neovascularization is another possible therapeutic approach (Brady et

al., 1997). This treatment is aimed at destroying the vascular tissue by inhibiting the

growth of new blood vessels and thus reduces the risk for further leakage or bleeding

(Flaxel, 2002). A few clinical studies have been undertaken to investigate the

outcome of this treatment, but the results are not consistent. Some studies have found

that the treatment was beneficial in reducing the degree of vision loss (Chakravarthy

et al., 1993; Brady et al., 1997; Postgens et al., 1997), while more recent studies

found that the treatment did not reduce the degree of vision loss (Stalmans et al.,

1997; Bergink et al., 1998; Hart et al., 2002). Although the reasons for the

differences in these results are still under investigation, external beam radiation

therapy in subjects with subfoveal choroidal neovascularization in AMD has not

been commonly used.

Photodynamic therapy (PDT) with the photosensitive dye Verteporfin (Visudyne;

Novartis Ag, Bulach, Switzerland) is a new treatment for exudative AMD, which

was approved by the USA Food and Drug Authority in 2000 (12 April, 2000)f. This

therapy can be used to treat AMD where the blood vessels have already reached the

fovea. Unlike thermal laser photocoagulation, it has been claimed that PDT causes

no damage to the overlying photoreceptors and gives no further reduction in vision

(Franks, 2002). By injecting the photosensitive dye via a peripheral vein, the

photosensitised subretinal membrane can be ablated with low levels of infrared laser

light. As verteporfin is a photosensitive dye, patients receiving this therapy must stay

away from any exposure to bright sunlight for 24 hours post therapy to avoid the dye

escaping to the surrounding tissue (Arnold and Sarks, 2000). Recent studies have

shown that PDT is an effective therapy in reducing the risk of moderate and severe

vision loss in patients presenting with sub-foveal lesions and choroidal

f http://www.centerwatch.com/patient/drugs/dru616.html (November, 2000)


25

neovascularisation (Verteporfin in Photodynamic Therapy Study Group, 2001 (a);

Verteporfin In Photodynamic Therapy Study Group, 2001 (b)). Rechtman (2002)

found that the therapy could reduce the risk of vision loss due to neovascular AMD

by approximately 20 to 30%. To date, the majority of studies which have examined

the effectiveness of PDT in reducing the risk of vision loss in patients with AMD

have been limited to a maximum of 2 years. Further longitudinal research examining

the long term benefit of PDT is needed.

Transpupillary thermotherapy is the treatment of subretinal neovascular membranes

by diode laser power which generates hyperthermic energy onto the retina (Ip et al.,

1999). Karel and his colleagues (2002; 2002) suggested that transpupillary

thermotherapy was a possible treatment for neovascular AMD. However few clinical

studies with large sample sizes had been conducted to evaluate its outcome on

reducing the vision loss for subjects with AMD (Auer et al., 2002; Karel and

Zahlava, 2002; Karel et al., 2002; Lanzetta et al., 2002). Research studies

investigating the effectiveness of this treatment and the long-term effects are still on

going.

The subfoveal choroidal neovascular membranes (CNV) in patients with exudative

AMD are sight threatening. In addition to the above-mentioned medical treatments,

surgical excision of the CNV has been considered to be an important therapeutic

approach (Thomas and Kaplan, 1991; Lambert et al., 1992). Additional methods

involving the transplantation of the RPE cells after surgical excision of CNV (Gass,

1994; Algvere et al., 1997) and the translocation of the macula by repositioning the

fovea from an area of subretinal pathology to an area of normal subretinal tissue

(Machemer and Steinhorst, 1993 (a); Machemer and Steinhorst, 1993 (b)) have been

proposed. Various clinical and experimental trials have been undertaken to

investigate the benefits of these surgical treatments. However, they are still in the

early stages of development and have not yet been accepted as therapies for AMD

(Toth and Machemer, 1999; Pertile and Claes, 2002).

Instead of directly treating AMD (either medically or surgically), supplementary

antioxidant micronutrients have been proposed as one way to reduce the incidence


26

rate and early progression of AMD. The hypothesis is that the nutritional

supplements reduce the generation of free radicals and then decrease the lipid

accumulation, which is thought to cause oxidative damage to photoreceptor cells and

proliferation of drusen (Bressler et al., 1988; Beatty et al., 1999). There have been a

number of large-scale randomised trials conduced by the Age-Related Eye Disease

Study (1996; 1996; 2001) examining the role of antioxidant supplementation on

AMD. Their results have suggested a moderate beneficial effect of beta carotene,

vitamin C, and zinc supplementation in reducing the progression to severe AMD.

However, a longitudinal study to monitor the longer-term progression rate of AMD

is needed before this therapy can be considered effective in reducing the risk of

AMD and its progression.

1.3 Reading with age-related macular degeneration (AMD)

Throughout the world, reading is one of the most highly valued activities in human

society. It is not only essential to effectively cope in society but also is an important

form of recreation. This may be particularly valuable for older people who are more

subject to physical limitations. For example, reading may compensate for hearing

loss, which increases with ageing (Lott et al., 2001). Any ocular disorder which

deprives people of reading causes severe vision disability. Many studies have

demonstrated that older people with visual impairments have difficulty with reading

(Farrall, 1991; Elliott et al., 1997a; Watson et al., 1997; Wolffsohn and Cochrane,

1999). The retrospective study of clinical records conducted by Elliott et al. (1997a)

indicated that the most frequent primary objective of older people attending low

vision clinics was to seek assistance to alleviate their reading disability (75%).

Therefore, reading is the most important rehabilitation goal of patients with low

vision (Krieger, 1967; Faye, 1970; Hall et al., 1987; Farrall, 1991; Elliott et al.,

1997a; Wolffsohn and Cochrane, 1999). Before further discussion on how reading

rehabilitation helps people with low vision to retain their reading ability, it is

important to have a general review on reading and the factors that affect low vision

reading particularly in people with AMD.


27

1.3.1 Reading with low vision

Whittaker and Lovie-Kitchin (1993) reviewed a number of research studies on the

visual psychophysics of low vision reading and suggested four visual requirements

for reading – acuity reserve, contrast reserve, field of view and central scotoma size.

Each of the visual requirements is discussed in detail in section 1.3.3. In their review,

they pointed out the visual requirements for reading depend on the nature of the

reading task. Based on different reading rates and reading requirements, they divided

reading into three categories (Table 1.4). Spot reading is reading for activities of

daily living or for survival, which does not require a fast reading rate. For reading

books or passages of text, a faster reading rate is required so as to attain meaning

from reading. Fluent reading rate for people with low vision refers to approximately

80 wpm, which is approximately the reading rate for second grade reading level

(Carver, 1990). A high fluent reading rate is approximately 160 wpm (Whittaker and

Lovie-Kitchin, 1993), which is equivalent to that for approximately sixth grade

reading level (Carver, 1990).

Table 1.4 Types of reading for people with low vision (Whittaker and Lovie-Kitchin, 1993)

1.3.2 Assessment of reading performance

In clinical low vision practice, near visual acuity is usually the only measure of

"reading performance" that is recorded by practitioners. It is the person’s threshold

print size measured with a test chart containing sentences or words in typeset print at

a defined working distance such as 40 cm. Assessment of reading should provide

more information about overall performance than that provided by a simple measure

of a person's resolution limit. Reading performance can be evaluated by reading rate

(speed), reading comprehension, eye movements and questionnaires (National

Research Council, 2002). Among these available measures, reading rate and reading

halla



28

comprehension are commonly used as the tools to assess reading performance

(Legge et al., 1985 (b); Carver, 1990; Whittaker and Lovie-Kitchin, 1993; Rubin and

Turano, 1994; Watson and Whittaker, 1994).

1.3.2.1 Reading rate

Reading rate refers to the number of words read as a function of time. It is usually

expressed in words per minute (wpm). Legge and co-workers (1985 (a); 1985 (b))

proposed that reading rate should take reading accuracy into account to reflect the

quality of reading. They defined the reading rate as number of words correctly read

in a unit time.

Reading rate has been widely used as the primary index for measuring reading

performance for a number of reasons (Legge et al., 1985 (b); Legge et al., 1989 (b);

Legge, 1991; Legge et al., 1992; Rubin and Turano, 1994). Firstly, it is relatively

easy to measure and provides a straightforward objective measurement of reading

performance. Secondly, it is sensitive to variations in stimulus parameters (Legge et

al., 1985 (a); Legge et al., 1985 (b); Legge et al., 1989 (b); Whittaker and Lovie-

Kitchin, 1993). Lastly, it is highly reproducible provided that the same testing

conditions and level of reading material are used (Legge, 1991). Difficulty of the

reading material, such as the complexity of the content or structure of the sentences,

has an impact on reading rate. Provided that the readability of reading material does

not exceed the subject’s reading level, reading rate does not vary significantly

(Carver, 1990). Carver (1990) suggested that the differences in reading rate due to

different level of difficulty of reading materials could be minimised by measuring the

reading rate in terms of “standard words” per unit time. A standard word is defined

as a word length of six characters. For all experiments reported in this thesis, reading

rate was calculated as the number of standard words correctly read per minute.

Reading rate is a measure that reflects the dynamic nature of reading but does not

include an assessment of the subject’s ability to interpret the information from

reading (Legge et al., 1989 (b); Dickinson and Rabbitt, 1991). In order to eliminate

this drawback, comprehension is an additional measure of reading performance.


29

1.3.2.2 Comprehension

Comprehension is the process of integrating visual information from the printed page

with cognitive information stored in the memory to achieve understanding (Smith,

1978). It depends more on non-visual linguistic processing or cognitive factors than

does reading rate and may be more useful in revealing interactions between cognitive

capacities and visual deficits (Carver, 1990). Research has shown that there is little

relationship between reading rate and reading comprehension; therefore these two

measures (reading rate and reading comprehension) have been suggested as separate

indices of reading performance (Legge et al., 1989 (b)). Legge and co-workers (1989

(b)) found that comprehension was not significantly affected by an individual’s

vision and reading rate. In contrast, Dickinson and Rabbitt (1991) found that free

recall comprehension was impaired for subjects with simulated low vision. Further

research study is required to investigate the relationship between comprehension and

low vision reading (National Research Council, 2002).

1.3.2.3 Eye movements

Recording eye movements during reading enables changes in reading rate to be

interpreted in terms of the underlying parameters of oculomotor behaviour (e.g.

saccade length and fixation durationg). Comparison of the eye movements of people

with normal and low vision can help to identify the reasons for poor reading

performance or slower reading rate for people who are visually impaired (Trauzettel-

Klosinski et al., 1994; Bullimore and Bailey, 1995). As the research reported in this

thesis did not include the recording of eye movements during reading, a

comprehensive literature review on the relationship between eye movements and

reading was not undertaken.

g Reading eye movements consist of a sequence of saccades separated by fixation. Saccade is a high-velocity eye movement that brings the fovea to the object without information extraction. Fixation duration is the period that information is gathered (measured in milliseconds).


30

Questionnaires

Since the 1990s, a number of research studies have focussed on designing

questionnaires as psychometric instruments used to measure visual impairment

(Mangione et al., 1992; Mangione et al., 1995; Mangione et al., 1998; Haymes et al.,

2000; Haymes et al., 2001; Mangione et al., 2001) and its impact on daily function

(Massof and Rubin, 2001). Some of these questionnaires, such as the National Eye

Institute Visual Function Questionnaires (Mangione et al., 1998) and Activities of

Daily Vision Questionnaires (Friedman et al., 1999 (b)), include a series of questions

related to a participant’s reading behaviour. Their results showed that the objective

reading measures were highly correlated with the subjective responses of the

participants. This suggests that the use of questionnaires can be a useful measure to

reflect participants’ reading performance.

In this study, reading rate (with and without magnifier) was adopted as the objective

method to assess subjects’ reading performance with no comprehension measured or

eye movements recorded. In the reading rate measure, subjects were requested to

read for understanding which involved the complete thoughts in the sentences of

textual materials (Carver, 1990). In addition, a simplified questionnaire was used to

elicit the subjective response to subjects’ habitual reading behaviour and the

problems encountered in reading that might not be detected by objective

assessments.

1.3.3 Factors affecting reading performance

There are a number of factors that can affect reading performance. Legge (1991)

broadly categorised these factors into visual, text and non-visual variables. Visual

variables include visual acuity, contrast sensitivity and visual field status while text

variables include print size, illuminance, contrast, field of view and text structure.

The non-visual variables mainly reflect the reader’s background, such as age,

cognitive or linguistic capacity (intelligence) and education. In this section, only

factors that are related to low vision reading especially for people with central field

loss (due to AMD) are discussed.


31

1.3.3.1 Visual acuity

Visual acuity measures the spatial resolution of a person’s visual system and is the

primary functional test of vision. It indicates the angular size of the smallest detail of

a recognition task (letters, symbols or words) that a person can resolve. Visual acuity

can be measured at different working distances, but is typically assessed for distance

(e.g. 6 m) and for near (e.g. 40 cm). Distance visual acuity is usually expressed in

Snellen notation in the form of a fraction with the test distance (e.g. 6 m) as the

numerator and the distance at which the letters subtend 5 minutes of arc (MAR) as

the denominator. Decimal notation (the reciprocal of the Snellen notation) and

logMAR notation (logarithm of the MAR) are two alternative expressions of distance

visual acuity. For near reading or text visual acuity, logMAR notation is commonly

used to express the threshold print size that a person can resolve at a standardised

test distance. M units and N-notation are two other alternatives used to represent near

visual acuity. M unit is the distance in metres at which the height of the lower case

letters subtends 5 minutes of arc while N notation refers to the size of the typeface

where each point represents 1/72 part of an inch (0.353 mm). N notation is a British

notation that is commonly used for standard near vision test type (Time Roman).

Usually, N-notation can be converted to M units by dividing by eight, provided that

the same font and style are used (Bailey, 1991).

Numerous studies have found that when foveal vision is degraded, reading rates as

well as reading comprehension are reduced (Rayner, 1979; Rubin and Turano, 1994;

Chung et al., 1998 (a)). Table 1.5 summarises the studies in which the relationship

between distance and near visual acuities and reading rate have been examined.

Although the relationships reported in the studies vary with different correlation

coefficients (r-value) and different levels of significance, there is a general

agreement that visual acuity is positively correlated to reading rate. Thus people with

better visual acuity read faster than those with poorer acuity.


32

Table 1.5 Summary of previous studies on the relationship between visual acuity and reading rate

Research study Sample size

Materials of measurement

Correlation with distance acuity

Correlation with near acuity Remarks or limitation of the study

Sloan and Habel (1973) 22 LVA Not measured Positive relationship (r-

value was not reported)

Goodrich et al. (1976) 27 CCTV r = 0.46 (NR) Not measured

12 CCTV

r = 0.34 (NS) r = 0.53 (<0.001) Goodrich et al.

(1977) 12 Optical LVA r = 0.49 (NS) r = 0.45 (NS)

For these two studies, the correlation on day 1 was reported in this table. Methodology and reading task used to measure RR were not reported.

Goodrich et al. (1980) 96 CCTV and optical

LVA r = 0.29 (<0.05) r = 0.21 (NS)

Krisher and Meissen (1983) 72 Scrolled text on TV

screen (5x TPS) Positive relationship Not measured

Legge et al. (1985 (b)) 16 Scrolled text

(optimal print size) r = 0.28 (NR) r = 0.57 (NR)

Krischer et al. (1985) 178 Scrolled text (5x

threshold print size)

r = 0.83 (NR) (Retinal diseases) r = 0.81 (NR) (Refraction anomalies) r = 0.53 (NR) (optic nerve atrophy) r = 0.24 (NR) (AMD)

Not measured Acuity measured by gratings was probably over-estimated in this study.

Rubin (1987) 28 Scrolled text on TV monitor (Optimal print size)

Not measured r = 0.63 (NR)

Lovie-Kitchin and Woo(1987) 10 CCTV (Scrolled

text) r = 0.29 (NS) r = 0.29 (NS) Sample size was too small for analyses.


33

Table 1.5 Summary of previous studies on the relationship between visual acuity and reading rate (continued)





Rubin and Legge (1989) 19 Scrolled text with

6° characters Not measured r = 0.37 (NR)

Near visual acuity combined with contrast sensitivity (0.3 c/deg) improved the regression coefficient of RR. The RR might not reflect the reading performance in real world, as the size of the text was very large.

Lowe and Drasdo (1990) 9

CCTV (scrolled text at optimum print size)

r = 0.59 (NR) Not measured Sample size was too small for analyses.

McMahon et al. (1991)

19 (AMD)

Static text (1.5x TPS) Not measured r = 0.72 (NR)

Legge et al (1992) 141

Static text with 6° characters (sentences or unrelated words)

r = 0.31 (<0.001 - All) r = 0.33 (0.04) (cloudy M and CFL) r = 0.39 (0.005) (clear M and CFL) Not significant (intact CF)

Not measured The RR might not reflect the reading performance in real world, as the size of the text was very large.

Leat and Woodhouse (1993)

30 Optical LVA (unrelated words of 1.6x TPS)

r = 0.51 (p=0.004) Not measured RR of unrelated words was slower than RR for text (Legge

et al., 1989 (a)).

Ahn and Legge (1995) 40 Static text of N12

with LVAs r = 0.30 (NS) Not measured

Bullimore and Bailey (1995)

13 (AMD)

Static text (optimal print size) r = 0.30 (NR) r = 0.72 (NR)

Sunness et al.(1996)

30 eyes (AMD)

Random words (1.3 to 6x threshold print size) on TV

Not significant Not measured


34

Table 1.5 Summary of previous studies on the relationship between visual acuity and reading rate (continued)





Song et al. (1996) 94 Scrolled text

(sentence)

r = 0.34 (0.0008) (Whole group) r = 0.55 (<0.01) (CFL) Not significant (PFL)

Not measured

Bowers (1998 (a))

20 (AMD)

Static text (passage) of 16 point 0.74 (<0.01) r = 0.75 (<0.01) RR and reading comprehension were assessed.

Lovie-Kitchin et al. (2000 (a))

22 (MD)

Static text (passage) at critical print size r = 0.79 (<0.001) r = 0.79 (<0.001)

Aquilante et al. (2001)

15 (AMD)

Static text on projector (optimal print size)

Not significant Not measured

Sample size included subjects with mixed causes of low vision unless otherwise indicated. Closed circuit television (CCTV) and low vision aids (LVA) were the major aids used for the measurement. AMD - age-related macular degeneration MD - macular degeneration TPS - Threshold print size RR - reading rate Cloudy M - Cloudy media Clear M - Clear media CFL - Central field loss CF- Central field PFL - Peripheral field loss (NR) - Significance level not reported (NS) - Not statistically significant The correlation coefficients (r-value) which achieved statistical significance are bolded.


35

The wide discrepancy in the results is mainly due to the wide range of experimental

methods used in different studies. Many different tests were used to measure visual

acuity. For example, Krischer et al. (1985) measured visual acuity using gratings

which over-estimates the visual acuity of subjects with low vision compared with

acuity measured by letters or symbols (Kitchin and Bailey, 1981). In addition, not all

studies measured acuity to threshold nor was it apparent whether the reading

materials selected for measuring reading rate were at or above critical print sizeh to

measure maximum reading rate (Bullimore et al., 1990; Song et al., 1996).

For studies in which the relationships of reading rate with both distance and near

visual acuities were reported, reading rate appears to be more strongly correlated to

near word reading visual acuity than to distance visual acuity (Goodrich et al., 1977;

Legge et al., 1985 (b); Bullimore and Bailey, 1995). This might be because near

visual acuity measured with unrelated words or coherent text is more related to the

process of reading.


For many years, visual acuity has been the main test for patients with low vision in

assessing their reading performance. However, many clinical and research studies

have shown that contrast sensitivity also plays an important role in reading

performance (Brown, 1981; Rubin and Legge, 1985; Legge et al., 1987; Rubin,

1987; Rubin and Legge, 1989; Leat and Woodhouse, 1993; Leat and Woo, 1997).

Contrast sensitivity is a measure of the ability to detect objects of different spatial

frequencies and at different contrasts. There have been many types of charts that are

commercially available to measure contrast sensitivity, such as Pelli-Robson,

Cambridge Gratings, Regan Low Contrast and University of Waterloo Low Contrast

charts. Leat and Woo (1997) investigated these charts and reported that the Pelli-

Robson chart was the best predictor of low vision reading rate because of its high

repeatability and wide range of spatial frequency being measured. In the study

reported in this thesis, the Pelli-Robson chart was used to measure the subject’s

h Critical print size is the smallest print size at which a person can achieve maximum reading rate.


36

contrast sensitivity. The design of this chart and the measuring procedures are

discussed in Chapter 4 (section 4.3.2.3).

Previous studies have revealed a correlation between reading rate and contrast

sensitivity at low spatial frequencies among people with media opacities and visual

field loss (Brown, 1981; Legge et al., 1987; Rubin, 1987; Rubin and Legge, 1989;

Leat and Woodhouse, 1993). In general, deficits in contrast sensitivity result in

reduced reading rate although the specific spatial frequency that correlated with

reading rate differed between studies. Brown (1981) found a high correlation

between reading rate and contrast sensitivity at 0.2 and 0.4 cycles/degree (cpd)

measured with the Arden plates, but found a poor correlation when the contrast

sensitivity was measured by oscilloscope. Brown (1981) suggested that the visual

field assessed with the oscilloscope monitor was relatively larger than the field used

for reading, whereas the Arden plates measured contrast sensitivity over a similar

visual field to that used for reading. Therefore, the result reported by the Arden

plates might be more appropriate to compare with reading rate.

Rubin and Legge (1989) and Leat and Woodhouse (1993) agreed with Brown (1981)

that contrast sensitivity at low spatial frequencies was strongly correlated with

reading rate. Rubin and Legge (1989) stated that contrast sensitivities at both low

spatial frequencies (0.1 to 0.3 cpd) and the peak contrast sensitivity were good

predictors of reading rate for subjects with media opacities and field loss (central or

peripheral). Leat and Woodhouse (1993) found that contrast sensitivity at 0.5 cpd

was strongly correlated with low vision reading rate for participants with different

ocular disorders. They suggested that considerable loss of contrast sensitivity at high

spatial frequencies could be sustained without significantly affecting reading rate

since it can be compensated for by magnification. This is because magnification can

possibly reduce the effects of low contrast by increasing the acuity reserve for

patients to read small print to achieve optimal reading rate (Mohammed and

Dickinson, 2000). However similar loss of contrast sensitivity at low spatial

frequencies might result in reduction in reading rate in people with visual

impairment (Leat and Woodhouse, 1993).


37

In contrast, Bullimore and Bailey (1995) found a non-significant correlation between

contrast sensitivity and reading rate for subjects with AMD. The results reported in

this study only reflected the relationship between contrast sensitivity and reading rate

for people with AMD, which might not apply to people with causes of low vision

other than AMD. In addition, only a small sample size (n = 13) and a more restricted

range of spatial frequencies were used by Bullimore and Bailey (1995) compared

with the studies by Rubin and Legge (1989) and Leat and Woodhouse (1993). These

could be reasons for the differences found in the strength of the correlation between

reading rate and contrast sensitivity between studies.

1.3.3.3 Central scotoma

It is evident that the status of the central visual field is also very important in reading

(Legge et al., 1985 (a); Legge et al., 1985 (b); Whittaker et al., 1988; Bailey et al.,

1989; Legge, 1991; Rumney and Leat, 1994; Bullimore and Bailey, 1995). For

people with low vision and central field loss (due to AMD), the maximum reading

rate is substantially slower than that for people with other low vision causes (Legge

et al., 1985 (b); Elliott et al., 2001; Patel et al., 2001). This is mainly because fewer

letters can be read per forward saccade, more regressions are made and there is a

reduced visual span in the peripheral visual field which is essential for reading with

central field loss (Legge et al., 1985 (b); Rumney and Leat, 1994; Bullimore and

Bailey, 1995; Fine and Rubin, 1999).

Instead of using the non-functioning macula, patients with AMD usually develop a

PRL at extra-foveal locations to perform visual tasks. The location of this PRL

relative to the scotoma may be an important factor affecting reading rate (Rayner et

al., 1982; Guez et al., 1993; Fletcher, 1999). If the PRL is located in the field to the

left of the scotoma (i.e. the scotoma is placed to the right of fixation in visual field),

both the visual acuity and reading rate may improve, as the person is not using a non-

functioning area for fixating. However, the increase in reading rate may be less than

predicted because of the location of the scotoma and the structure of English text

which is read from left to right, requiring rightward eye movements along a line. A

scotoma which is placed to the right of fixation in the visual field interferes with the


38

rightward movements for correct orientation of the fixated word or the beginning of

the next word along the line (Guez et al., 1993). Despite the impediment of the right

scotoma (relative to the PRL) on reading performance, a number of studies have

shown that people with AMD often develop PRLs such that the scotoma is to the

right rather than to the left of fixation (Guez et al., 1993; Trauzettel-Klosinski et al.,

1994; Sunness et al., 1996; Fletcher and Schuchard, 1997; Fletcher et al., 1999).

Also, patients with AMD may locate their PRL above or below the scotoma (Fletcher

et al., 1999; Schuchard and Fletcher, 2000). The choice of the position for the PRL is

still not well understood and there have been no studies investigating the relationship

between the locations of PRL and the scotoma, to define the best orientation for

reading performance (Fletcher et al., 1999). Further research related to this field is

needed.

It has previously been established that some people with central scotomas use more

than one PRL during reading (Duret et al., 1999; Safran et al., 1999; Deruaz et al.,

2002). For example, a person can use one PRL to have a global view of text but

another PRL to discriminate the letters more precisely. This implies that the use of

different visual tasks is important in mapping the location of PRL. Those studies

which reported that the majority of people with central field loss developed a PRL on

the left of the scotoma (in visual field space) presented a simple stimulus by a

scanning laser ophthalmoscope (SLO) which did not require discrimination of fine

detail (Guez et al., 1993; Trauzettel-Klosinski et al., 1994; Sunness et al., 1996;

Fletcher and Schuchard, 1997; Fletcher et al., 1999). Thus, the location of the PRL

and the scotoma might not be in the same position when the subjects were requested

to undertake reading tasks requiring fine detail and discrimination of horizontal

strings (Cummings et al., 1985; Cummings et al., 1988).

Lei and Schuchard (1997) used a SLO to compare the fixation pattern and the

location of the PRL of subjects with macular diseases for different illumination

levels. They reported that two well-defined PRLs were developed by some subjects,

depending on the brightness of the stimuli used for the visual tasks. The location of

the PRL used for high illuminance was closer to the fovea compared with the PRL

used for low illuminance. This suggests that if low illuminance in the SLO is used


39

for scanning the retinal areas of patients with AMD, the locations of PRL and

scotoma observed through the SLO might not be the same PRL that is used for

reading which requires higher illumination (Lei and Schuchard, 1997). Therefore,

improved techniques in mapping the position of PRL and the scotoma are necessary

such that the results reported from the SLO or other instruments can better reflect the

location of the PRL used for reading.

The measurement of the visual field of patients with AMD is usually for functional

rather than for diagnostic purposes. Visual fields are currently assessed by

confrontation, arc perimeters, tangent screens, Amsler charts and computerised

perimetry (Lovie-Kitchin and Bowman, 1985). Among these techniques,

computerised perimetry provides accurate detection of visual field defects as it uses

standardised target parameters, a uniform background with large database of age-

matched normative values for comparison. However, this test is too difficult for

patients with low vision to perform, especially those with AMD because they have

difficulty maintaining steady fixation on the fixation target which is usually smaller

than their resolution limit (National Advisory Eye Council, 1993). Instead of using a

threshold measure of the visual field, Lovie-Kitchin and Whittaker (1998 (b))

recommended the use of a modified Bjerrum tangent screen technique to measure the

central (25 degrees) visual field of patients with AMD. Detailed procedures of this

visual field assessment are described in Chapter 4 (section 4.3.2.2).

1.3.3.4 Saccade length

Reading eye movements consist of a sequence of saccades and fixations. The

majority of reading time is determined by the number and the length of fixations. If

the saccades are shorter and fixations remain constant in duration, the number of

fixations increases and the time to read the text will therefore increase. This

argument has been supported by the results from previous eye-movement studies,

which showed that people with low vision make shorter saccades than people with

normal vision (Rumney and Leat, 1994; Bullimore and Bailey, 1995). Rumney and

Leat (1994) and Bullimore and Bailey (1995) studied the eye movements of subjects

with low vision and suggested that saccade size (in terms of numbers of characters)


40

was the limiting factor affecting low vision reading rates. Rumney and Leat (1994)

reported a mean of 3.5 characters per forward saccade for subjects with low vision

compared with a mean of 6.8 characters for subjects with normal vision. However,

the comparison of saccade sizes was made between the eye movements of subjects

with low vision reading with magnifiers and the eye movements of subjects with

normal vision without magnifiers. Because of the different parameters used in

measuring the saccades sizes of subjects with normal vision and low vision, the

shorter forward saccades shown in the subjects with low vision could have been due

to the use of magnifiers. Bowers et al. (2001 (b)) compared the reading rates and eye

movements with and without magnifiers for subjects with low vision and central

field loss. They found no significant difference in forward saccades for subjects with

low vision reading with and without magnifiers. This indicates that the reduction in

saccade length (forward saccade) is an adaptation to low vision and not due to the

magnifier. Reduction in saccade size will, by definition, limit reading rates. In

addition, Whittaker and colleagues (1990; 1991) reported that reduction in saccadic

accuracy, elevation in latencies and increase in fixation durations were reasons for

the reduction in reading rate in their investigations of saccadic eye movements.

Rubin and Turano (1992) eliminated the need for saccadic eye movements during

reading by the use of rapid serial visual presentation (RSVP). In RSVP, text is

presented sequentially on the screen, one word at a time at the same location in the

visual field. Although the same retinal eccentricities were used for all subjects,

reading rates for subjects with central field loss were significantly slower than for

subjects with normal vision. Rubin and Turano (1992) suggested that there must be

factors other than reduced saccade length which reduced low vision reading rates.

Raasch and Rubin (1993) attributed the reduction in low vision reading rate to the

degraded vision bordering the scotoma and the unsteadiness of the eccentric fixation.

Culham et al. (1992) used a SLO to compare the reading performance for subjects

with AMD and normal vision at the same eccentricity. Their results supported the

suggestion by Raasch and Rubin (1993) that the quality of vision function which

borders the margin of scotoma is a factor reducing reading rate in people with low

vision. However, Timberlake et al. (1989) found that poor oculomotor control rather

than the unsteadiness of the eccentric fixation was the major reason for the slow


41

reading rates for subjects with low vision and central scotoma (Timberlake et al.,

1989).

1.3.3.4.1 Visual span

Legge et al. (1997) proposed the hypothesis of "shrinking visual span" and argued

that the reading rate deficit in low vision was mainly due to the reduction in visual

span. The term “visual span” describes the sensory limitation of a person’s

perception and is defined as the size of the region that can be processed without the

help of any contextual information in one fixation (Rayner, 1979). Legge and

colleagues (1997) defined this in terms of the number of characters that are

recognised on each fixation. Fewer letters are recognised on each fixation by people

with visual impairment, therefore forward eye movements (saccades) are shorter and

more numerous, and the reading rate is reduced (Legge et al., 1997). They proposed

that the size of the visual span has a fundamental correlation with reading rate.

However, no eye movements of the subjects with low vision were recorded in

Legge's study (1997). Thus there was no direct evidence that shrinkage of visual

span and reduced forward saccade lengths resulted in slower reading rates.

1.3.3.5 Age

A few studies have investigated the effect of age on reading rate for people with low

vision (Legge et al., 1992; Ahn and Legge, 1995; Song et al., 1996; Sunness et al.,

1996; Lovie-Kitchin et al., 2000 (a)). Legge et al. (1992) investigated a few clinical

variables and found that age and distance visual acuity accounted for 30% of the

variance in low vision reading rate. They argued that the reduction in reading rate as

a consequence of increasing age was due to the extra attentional capacity required to

read with low vision which people with visual impairment might not have. Ahn and

Legge (1995) similarly reported that age accounted for 43.7% of the variance in

reading rate with magnifiers. In contrast, Lovie-Kitchin et al. (2000 (a)) found that

age was not a significant predictor of reading rate. They attributed the variation to

the way that reading rate was measured across studies as the major reason for the

different results (Lovie-Kitchin et al., 2000 (a)). In Lovie-Kitchin’s study, subjects


42

were instructed to read aloud at normal reading rate to obtain meaning from the text,

whereas previous studies by Legge and colleagues (1992; 1995) instructed the

subjects to read the text aloud as quickly as possible. According to the levels of

reading proposed by Carver (1990), reading for understanding (rauding) is a more

complicated reading process than skimming (reading with no requirement of

understanding) as integration of thoughts from reading is required. Lovie-Kitchin et

al. (2000 (a)) suggested that the effect of age might have a greater impact on the

faster reading rate (which was skimming) rather than on the slower rauding rate used

in their study.

1.3.3.6 Text variables

For people who are visually impaired, any change in the text variables may lead to

an increase in task difficulty, which in turn reduces the reading rate (Legge et al.,

1985 (a); Legge et al., 1985 (b)). In this section, the effects of text size, illuminance,

contrast, field of view and text structure on low vision reading are discussed.

1.3.3.6.1 Text size

It is essential that the visual task is larger than the individual’s threshold print size

(acuity limit) in order to read fluently. Many studies have found that reading rate for

people with normal and low vision increases as the print size increases from

threshold to a critical print size (CPS) at which maximum reading rate is achieved

(Bailey and Lovie, 1980; Legge et al., 1985 (a); Legge et al., 1985 (b); Lovie-Kitchin

and Woo, 1987; Lowe and Drasdo, 1990; Legge, 1991; Rubin and Turano, 1994;

Plass and Yager, 1995). For print size beyond the CPS, reading rate becomes stable

with no significant change with further increases in print size. Only when print size

becomes so large that the field of view is restricted or smooth-pursuit eye movements

can no longer match the speed of text movement does reading rate begin to reduce

(Legge et al., 1985 (b); Lovie-Kitchin and Woo, 1987; Legge et al., 1989 (a); Legge,

1991). Maximum reading rate is the mean of the reading rates at and above the CPS.

The required CPS to achieve maximum reading rate varies across different studies,


43

which may be attributed to the differences in methodology of reading rate measures,

causes of low vision and numbers of subjects recruited.

Legge et al. (1985 (b)) found that text sizes of 3o (1.56 logMAR) and 12o (2.16

logMAR) on a closed circuit television (CCTV) were required to achieve maximum

reading rates for subjects with normal (intact) central fields and central field loss

respectively. This suggests that people with low vision and central field loss require

larger print sizes to achieve maximum reading rate than people with intact central

field. The reading rates in Legge's study were measured by a forced scrolling

technique, in which the text on the screen was scrolled at the maximum rate until the

subjects could no longer read the print. This forced the subjects to read at their

maximum reading rate. Lovie-Kitchin and Woo (1987) used an unconstrained

scrolling technique to measure the reading rate of their subjects with low vision.

They found that maximum reading rate could be achieved at print sizes between 1.73

and 1.83 logMAR for the majority of their subjects with low vision. The reported

print sizes were comparatively smaller than the CPS for subjects with central field

loss in Legge’s study which may be because the subjects recruited by Lovie-Kitchin

and Woo had a mixture of low vision causes. Lowe and Drasdo (1990) used a similar

method to measure reading rates as a function of print size and found that subjects

with low vision required print sizes ranging from 1.5o to 9o (1.26 logMAR to 2.03

logMAR) to achieve maximum reading rates. Their range of print sizes was larger

than the print sizes reported by Lovie-Kitchin and Woo (1987), which may be

because Lowe and Drasdo (1990) measured reading rates with very wide fields of

view (25o to 100o). The effect of field of view on reading rate is discussed in section

1.3.3.6.4. The reported print sizes for maximum low vision reading rate are very

large. This suggests that a physical increase in print size is not realistic for real world

reading. As such, prescribing optical or electronic low vision aids rather than large

print reading materials is commonly used to assist patients with visual impairment

for reading.

Acuity reserve

Whittaker and Lovie-Kitchin (1993) introduced the term “acuity reserve” to explain

that print size had to be larger than the reader’s visual acuity, such that he/she can

read fluently and comfortably. Acuity reserve is defined as the ratio of print size that


44

the person intends to read (target print size) to their threshold print size (Whittaker

and Lovie-Kitchin, 1993). They suggested that visual acuity per se might not be as

important as acuity reserve in affecting reading performance. A range of optimum

acuity reserves to achieve maximum reading rate have been found for subjects with

low vision from different studies (Legge et al., 1985 (a); Legge et al., 1985 (b);

Lovie-Kitchin and Woo, 1987). After reviewing previous studies in the literature,

Whittaker and Lovie-Kitchin (1993) made some general recommendations on the

minimum required acuity reserve for different types of reading. An acuity reserve of

at least 1.3:1 (0.1 log unit) was required for spot reading (≥ 40 wpm) (refer to section

1.3.1). For fluent (≥ 80 wpm) and highly fluent reading rates (≥ 160 wpm), acuity

reserves of at least 1.5:1 (0.3 log units) and 3:1 (0.5 log units) were the minimum

requirements respectively (Whittaker and Lovie-Kitchin, 1993).

In contrast, Legge and colleagues (1992; 1995) suggested that the acuity reserve

required for fluent reading rate should be determined on an individual basis by using

the MNRead acuity chart (Mansfield et al., 1993; Mansfield et al., 1994) rather than

a generalised acuity reserve suggested by Whittaker and Lovie-Kitchin (1993). There

have been no studies comparing the acuity reserve determined by these two

strategies. As such, one of the aims of this current study was to compare the required

acuity reserve determined by these two approaches (Chapter 3).

1.3.3.6.2 Illuminance

There have been numerous research studies investigating the effect of illumination

on visual performance of people with low vision in the past 30 years. There is a

general agreement that the majority of people with low vision show an improvement

in visual acuity (Lie, 1977; Julian, 1984; Cornelissen et al., 1991) and reading rate

(Sloan et al., 1973; LaGrow, 1986; Bullimore and Bailey, 1989; Bullimore and

Bailey, 1995; Bowers et al., 2001 (a)) with increasing illumination. The visual

benefit from increasing illumination is more obvious and significant for low vision

people with AMD than for people with other low vision causes (Brown and Garner,

1983; Lovie-Kitchin and Bowman, 1985; Eldred, 1992; Bowers et al., 2001 (a)).


45

Bullimore and Bailey (1995) suggested that increasing the lighting reduced the size

and change the shape of the central scotoma for subjects with macular disease and so

the reading performance improved. They further pointed out that the significant

improvements in reading rate might be associated with subjects using different retinal

areas for fixation at low and high levels of illumination. Lei and Schuchard (1997)

found that some patients with AMD developed two well-defined PRL, one was used

for fixation at low light levels while the other one was used at high light levels (refer

to section 1.3.3.3). The results showed that people with central field loss preferred

using the retinal area located at or close to the fovea at high illuminance, which was

an area with lesser retinal eccentricity but higher resolution compared with the PRL

for low light levels. Visual acuity achieved with the PRL for high illumination was

significantly better than that for low illumination (Lei and Schuchard, 1997).

Bowers et al. (2001 (a)) reported significant improvements in near visual acuity and

reading performance with an increase in illumination from 50 to 2000 lux for

subjects with AMD. They suggested that using a magnifier with a good internal

source of illumination could reduce the amount of required magnification. Therefore,

illuminated stand magnifiers were prescribed for the subjects with low vision who

participated in this study (Chapters 4 and 5).

1.3.3.6.3 Contrast

The effects of contrast sensitivity and contrast reserve on reading rate have been

discussed in section 1.3.3.2. As a consequence of reduced contrast sensitivity,

reading rate may decline. However, contrast of the reading task also plays an

important role in reading performance. Research has found that a reduction in print

contrast can significantly decrease reading performance for people with low vision

(Brown, 1981; Legge et al., 1987; Rubin and Legge, 1989). The reduction in reading

rate is due to a reduction in contrast reserve (Whittaker and Lovie-Kitchin, 1993).

Contrast reserve

Whittaker and Lovie-Kitchin (1993) systemically reviewed and analysed previous

research studies related to contrast sensitivity and reading performance. In their


46

review (1993; 1994), the concept of “contrast reserve” was introduced. Contrast

reserve is defined as the ratio of the letter contrast of the printed letters to

individual’s contrast threshold. Whittaker and Lovie-Kitchin (1993) suggested that a

minimum contrast reserve required for spot reading rate (> 40 wpm) was 3:1, for

fluent reading rate (> 80 wpm) was 4:1 and for highly fluent reading rate (> 160

wpm) was 10:1. Their suggestions were supported by the results of a recent study by

Mohammed and Dickinson (2000). As contrast sensitivity has been shown to be a

factor affecting reading rates in people with low vision, assessment of contrast

sensitivity is recommended as part of a low vision consultation (Waiss and Cohen,

1991; Whittaker and Lovie-Kitchin, 1993).

Text with good contrast is essential to achieve maximum reading performance.

Cohen (1993) measured the contrast of some commonly used reading tasks and

reported that the contrast of newspaper print was 60 to 70%, while that for paperback

books was 75%. Among all reading materials, text printed by laser printer gives the

best contrast of 94% (Cohen, 1993). Owing to the wide range of contrasts

encountered in different reading tasks, it is important to ensure that contrast of the

print is not a limiting factor affecting reading performance for people with visual

impairment. As such, measuring the contrast sensitivity function of a patient with

low vision and the contrast of the reading materials that the patient intends to read

are necessary to ensure that sufficient contrast reserve is provided for fluent reading.

1.3.3.6.4 Field of View

Field of view (FOV) or window size refers to the number of characters that can be

seen. Numerous studies have investigated the relationship between field of view and

reading performance (Legge et al., 1985 (b); Lovie-Kitchin and Woo, 1987; Lowe

and Drasdo, 1990; Beckmann and Legge, 1996; Fine et al., 1996; Fine and Peli,

1996; Lovie-Kitchin and Whittaker, 1999 (a)) (Table 1.6). In general, reading rate

increases as FOV increases. A range of FOV from 5 to 24 characters has been

suggested from different studies for optimum reading performance. For example,

Legge et al. (1985 (b); 1991) recommended a FOV of 4 or 5 characters for optimum


47

reading rates, whereas Lovie-Kitchin and Woo (1987) suggested a much larger field

of view of 15 characters to achieve optimum reading rates.

Table 1.6 Required field of view (FOV) for optimum reading rate from previous studies

Research study Sample LVA FOV for optimum reading rate Comments

Legge et al. (1985 (b)) 16 LV CCTV 4-5 Forced scrolling

technique

Lovie-Kitchin and Woo (1987) 10 LV CCTV 11- 15

Unconstrained reading (manual scanning) and no retrace

Lowe and Drasdo (1990) 9 LV CCTV 25-35

Scrolled text (at optimum print size) with navigation demand

Fine and Peli (1996) 24 LV TV

screen 6-7 Scrolled text (with reference to 80% of maximum reading rate)

Fine et al. (1996) 20 LV STM 9

Static text (acuity reserve was not controlled)

CCTV 10

Unconstrained text scrolling with navigation demand required (Window size was cited with reference to 85% reading rate)

Beckmann and Legge (1996) 7 LV

Drifting text 5.2

Forced text scrolling without navigation demand required

Lovie-Kitchin and Whittaker (1999 (a))

27 LV Text

Varied with different acuity reserve (Lovie-Kitchin, 1996)

Forced scrolling technique

LV - Low vision LVA - Low vision aids STM – stand magnifiers CCTV – closed circuit television

The wide discrepancy in the FOV recommended for optimal reading is probably due

to differences in text presentation in each study (Whittaker and Lovie-Kitchin, 1993;

Beckmann and Legge, 1996). If the text is presented continuously at a forced rate by

scrolling method, as done by Legge and colleagues (1985 (b); 1996), smaller FOV is

required because the text is drifted without any manual scanning or retrace. By this

method, subjects do not need to read the text along the line (scanning) or to look for

the beginning of next line (retrace). In contrast, for text that is manually controlled by


48

the subjects (Lovie-Kitchin and Woo, 1987; Lowe and Drasdo, 1990), a larger FOV

is needed to achieve optimum reading rates since page navigation is required. Lovie-

Kitchin and Whittaker (1999 (a)) investigated the relationship between field of view

and reading rate for adults with low vision. They demonstrated that there was no

significant interaction between field of view and reading rate provided that acuity

reserve was optimal (> 0.2 log unit for acuity reserve). This suggests that acuity

reserve (magnification) rather than field of view has a more substantial impact on

reading rate.

1.3.3.6.5 Text structure

Context is an important determinant of reading rate. Reading materials using

unconnected words or meaningful text gives different reading rates for people with

normal (Carver, 1990) and low vision (Legge et al., 1989 (a)). Legge et al. (1989 (a))

reported that reading rates for subjects with low vision were 15 to 30% faster when

the test stimuli were sentences rather than unrelated words. Carver (1990) suggested

that the faster reading rate with sentences could be because no load on memory is

needed to prevent the words from being articulated at the maximum possible

articulation rate (Carver, 1990). When reading unrelated words, readers cannot rely

on the context cues provided by sentence structure but rely on their visual abilities to

recognise each unrelated word. However, the reading rates measured with these two

types of reading materials are highly correlated (Legge et al., 1989 (a); Carver, 1990;

Watson et al., 1990).

People usually read meaningful text rather than unrelated words in their habitual

reading. Therefore, measuring the reading rates using coherent text better reflects

people’s normal reading and was used in the study described in this thesis.

1.3.3.6.6 Mode of text presentation

Different modes of presentation have been used in psychophysical studies to measure

reading rate in people with normal vision and low vision. These include static text

(Legge et al., 1989 (a); Rubin et al., 1994; Rubin and Turano, 1994; Bullimore and


49

Bailey, 1995; Bowers, 1998 (a); Fletcher, 1999; Lovie-Kitchin et al., 2000 (a)),

scrolling (or drifting) text (Krischer et al., 1983; Buettner et al., 1985; Legge et al.,

1985 (a); Legge et al., 1985 (b); Lovie-Kitchin and Woo, 1987; Lovie-Kitchin, 1988;

Rubin and Legge, 1989; Lowe and Drasdo, 1990; Legge et al., 1992) and rapid serial

visual presentation (RSVP) (Rubin and Turano, 1992; Rubin and Turano, 1994).

Among the three types of presentation, in these studies text is more commonly

presented as a single line of words moved across a computer monitor in continuous

motion from right to left in front of readers. Legge et al. (1989 (a)) and Fine and Peli

(1995) found that the reading rate measured by forced scrolling text was faster than

the reading rate on static text among subjects with low vision (with and without

central field loss). This suggests that a dynamic text display might be advantageous

for patients with low vision because the requirement for return saccades is not

necessary and there is a "capturing of the eyes" motion by the display which might

be helpful (Legge et al., 1989 (a)). Theoretically, no eye movements are required in

RSVP during reading because each word is presented sequentially to the same retinal

location. Reading rate for people with low vision and central field loss measured by

RSVP is significantly faster than the reading rate measured by static reading

measures (Rubin and Turano, 1994; Fine and Peli, 1998). However, the difference

was not as great as anticipated, nor as great as in people with normal vision.

Despite an increased reading rate measured by scrolling and RSVP techniques, both

presentations do not reflect the typical reading situation for the majority of elderly

people. This is because the layout of information from reading books or sheets of

paper or computer screens, which require manual scanning and retrace during

reading, is ignored (National Research Council, 2002). Therefore, static or stationary

reading on a page of text, which reflects habitual reading, was adopted as the mode

of text presentation in this study.

1.3.4 Predicting reading performance from vision variables

A number of clinical measures affect reading performance. Among these clinical

variables – visual acuity (distance and near), contrast sensitivity, media status, visual


50

field status, size and location of scotoma, age, etc. - one or a combination may be

more important and useful for predicting low vision reading rate. Table 1.7

summarises the results from previous studies that have examined the clinical

variables predicting low vision reading rate. Because of the differences in methods –

in text presentation (scrolled or static), types of reading task (text or unrelated

words), aids used for reading, clinical variables included in the regression analyses

and characteristics of subjects recruited (e.g. causes of low vision) – among different

studies, the variables that significantly predicted reading rate were different across

studies.

For example, Lowe and Drasdo (1990) recruited only nine subjects with low vision

to study the variables of low vision reading rate, a sample size too small for analysis.

Legge and colleagues (1992) had 141 subjects but, in order to ensure print size was

not a limiting factor affecting reading rate, they used reading materials subtending 6°

in static mode to investigate the factors predicting reading rate. Because the size of

the reading text was so large compared with normal print or even large print reading

tasks, their results might not reflect the performance when reading real words. In this

study, predictors of reading rates with and without magnifiers were considered and

investigated (see Chapter 4).

1.4 Reading rehabilitation for people with age-related macular

degeneration (AMD)

Due to the limited number of medical treatments available for patients with AMD,

referral to low vision clinics for rehabilitative purposes is important. The aim of

vision rehabilitation is to improve the functioning and independence of people who

are visually impaired, such that their quality of life is not significantly affected by the

visual impairment (Nilsson and Nilsson, 1986; Nowakowski, 1994; Raasch et al.,

1997).


51

Table 1.7 Summary of previous studies on factors predicting reading rate of people with low vision

Research study Sample Aids Variables included Variables significantly predicting reading rate

Brown (1981) 10 (media opacity) 7 (AMD)

5.74x TPS of unrelated words on CCTV DVA, CS CS at low spatial frequencies accounted for 44 to 59% of

variance

Legge et al. (1985 (b)) 16 Scrolled text (optimal

print size) DVA, NVA, Media, Field Field status and media status accounted for 64% of variance

Rubin (1987) 28 Scrolled text on TV screen (5x TPS)

NVA, CS (peak sensitivity and area under CS function)

NVA accounted for 36% variance NVA and CS accounted for 64% of variance

Rubin and Legge (1989)

19 eyes (17 subjects)

Scrolled text on TV screen (of different print size)

DVA, NVA, Media, Field, CS CS at 0.3 cycles per degree accounted for 37% of variance. NVA and CS at 0.3 cycles per degree accounted for 66%.

Lowe and Drasdo (1990) 9 CCTV (scrolled text at

optimum print size) DVA DVA accounted for 35% of variance.

McMahon and Viana (1991) 19 AMD Static text (1.5x TPS) NVA, Saccadic frequency Saccadic frequency and NVA accounted for 70% of variance

Legge et al. (1992) 141

Static text with 6° characters (sentences and unrelated words)

DVA, Media, Field Age, Cause of low vision

DVA accounted for 10% of variance. Age accounted for 20% of variance Age and DVA accounted for 30% of variance.

Leat and Woodhouse (1993)

30 Unrelated words and letters of 1M or 1.6x TPS with LVA

DVA, CS CS at 0.5 cycles per degree accounted for 38% of variance.

Ahn and Legge (1995) 40 Static text of N12 with

LVAs Media, Field, DVA, Age, LVA type, MNRead score

MNRead score accounted for 80% of variance. Age accounted for 43.7% of variance. LVA type accounted for 42.3% of variance


52

Table 1.7 Summary of previous studies on factors predicting reading rate of people with low vision (continued)

Research study Sample Aids Variables included Variables significantly predicting reading rate

Bullimore and Bailey (1995) 13 (AMD) Static text (optimal

print size) DVA, NVA, CS, Scotoma size

NVA accounted for 48% and 52% of variance of maximum reading rate of optimal print that and 2x threshold print size (0.3 log acuity reserve) respectively

Song et al. (1996) 94 (mixed) Scrolled text (sentence) Field status, DVA, Age DVA accounted for 11.6% of variance

Age accounted for 21.4% of variance

Sunness et al. (1996))

30 eyes (AMD)

Random words (1.3 to 6x TPS) on television monitor

DVA, Size of atrophic area, Age

Size of the atrophic area accounted for 53% (for better seeing eyes) and 76% (for worse-seeing eyes) of variance

Sunness et al. (1997) 74 (AMD)

Unrelated words of 5.74x TPS on TV or cards

DVA, Scotoma DVA and location of scotoma (near fixation) accounted for 54% of variance with the reduction in DVA

Fletcher et al. (1999) 99 (mixed)

Scrolled text on TV screen (of different print size)

DVA, Presence of scotomas, Position of scotoma

DVA accounted for 42% of variance Presence of scotoma accounted for 33% of variance

Lovie-Kitchin et al. (2000 (a)) 22 (MD) Static text (passage) at

CPS DVA, NVA, Age NVA accounted for 73% of variance in sentence reading rate (MNRead)

Sample size included subjects with mixed causes of low vision unless otherwise indicated. Closed circuit television (CCTV) and low vision aids (LVA) were the major materials used for the measurement. AMD - age-related macular degeneration MD - macular degeneration TPS - Threshold print size CPS – critical print size DVA - distance visual acuity NVA - near visual acuity CS - contrast sensitivity


53

The majority of people who are visually impaired report that reading is the major

difficulty they encounter in their daily life, and seeking improvement in reading

performance is the main reason for these people attending low vision clinics

(Krieger, 1967; Faye, 1970; Hall et al., 1987; Farrall, 1991; Elliott et al., 1997b;

Wolffsohn and Cochrane, 1999). Reading rehabilitation for people with visual

impairment is essential given the strong association between reading ability and

quality of life (Hazel et al., 2000; McClure et al., 2000).

The most common therapeutic approach to assist patients with low vision to achieve

their reading goals is to provide magnification. Many research studies have shown

that low vision aids are particularly helpful for patients with AMD for reading

(Nasrallah et al., 1988; Archambault and Colenbrander, 1989; Stelmack et al., 1991;

Virtanen and Laatikainen, 1991). The aids enlarge the retinal image, such that details

of the image can be resolved by more peripheral, unaffected retinal areas to improve

reading performance.

1.4.1 Types of magnification in low vision

There are several types of low vision aids that are commercially available and

commonly used to assist people for reading. This section discusses the definition of

different forms of magnification and the associated low vision aids (Nowakowski,

1994).

1.4.1.1 Relative size magnification (RSM)

RSM is the magnification achieved by increasing the actual size of the object.

Therefore, people with low vision can view the enlarged object at a size that is

greater than their threshold resolution. The magnification is determined by comparing

the angle subtended by the magnified object at the entrance pupil of the eye to the

angle subtended by the original object (Woo, 2001). This implies that the change of

the angular size of the image equals the change in the object size. Therefore the

magnification can be simplified as the ratio of the size of the image to the size of the


54

original object provided that the viewing distance is kept constant. Examples of RSM

include large print books and enlarged numbers on telephone dial pads.

In general, large print refers to reading materials with print sizes between 18 and 24-

point. The use of large print can eliminate the need for optical aids for patients with

mild or moderate low vision and may be the only reading material accessible to those

with severe vision loss, even using strong magnifiers (Williams, 1999). However, for

people with moderate to severe visual impairment, RSM is not applicable because it

is impractical to increase the physical dimension of the reading text indefinitely until

it is resolvable by the people with low vision.

In this study, large print was used to provide magnification in the investigation of

reading practice on reading performance (see Chapter 4).

1.4.1.2 Relative distance magnification (RDM)

RDM is defined as the magnification provided by decreasing the distance between

the object and the eye. When the viewing distance decreases, the retinal image size

of the object increases (or is magnified). However, the person who uses this type of

magnification must either have sufficient accommodative amplitude to focus with

these short distances or have an appropriate positive lens (i.e. near spectacles) to

achieve a clear image for the short working distance. The magnifying effect is

calculated as the original viewing distance divided by the image distance. For

example, people with visual impairment prefer to sit closer to the television or to

hold a book closer in order to enlarge the retinal image size. Optical low vision aids

provide this type of magnification.

There have been many terms describing magnification in regard to low vision aids by

different authors, such as apparent magnification (or angular magnification), relative

magnification (or conventional magnification), iso-accommodation magnification

(Bailey, 1984 (b)). However the definition of each magnification is very ambiguous

and ill defined. Bailey (1984 (a)) introduced the term, equivalent viewing distance

(EVD) explaining the magnifying effect of low vision aids. EVD is defined as the


55

distance at which the original object would subtend an angle that is equal to the angle

subtended by the image at the observer's eye (Bailey, 1984 (a)). For example, if the

near threshold print size of a person with a near addition of + 2.5 D is 16-point print

(N16) at 40 cm, a low vision aid with an EVD of 10 cm can improve the threshold

resolution limit to 4-point print (N4)i.

1.4.1.3 Relative image magnification (RIM) or angular magnification

RIM is a combination of RSM and RDM including various lenses or lens systems.

This is a complicated type of magnification mainly used in complex optical low

vision aids, such as telescopes and binoculars (Zimmerman, 1996). This

magnification makes an object at distance appear closer to the eye and the image is

spread over the retina giving a magnifying effect. By comparing the angular subtense

of the image formed by the optical instrument or optical aid with the angular

subtense of the original object, angular magnification is defined.

1.4.1.4 Projection magnification

This type of magnification refers to an increase of object size by an overhead

projector or a movie camera onto the screen or monitor. Electronic aids such as

closed circuit television (CCTV) and projectors are common examples of projection

magnification (Zimmerman, 1996). The method of calculating the magnification is

similar to the one used for RSM by comparing the projected image size to the

original object size.

1.4.2 Determination of required magnification

In the past, there have been a number of methods to calculate required magnification

for low vision aids by the use of simplified equations (Dickinson, 1998). As some

assumptions have been made in calculating the magnification using these equations,

the magnification determined was usually under-estimated when compared with the

i The equation for the calculation is:

401016 x


56

final magnification required by the patients (Cole, 1993). Recently, there have been

two approaches suggested in the literature to determine the magnification to achieve

fluent reading rate (Legge et al., 1992; Whittaker and Lovie-Kitchin, 1993; Ahn et

al., 1995; Lovie-Kitchin and Whittaker, 2000). A comprehensive literature review of

previous and current methods of calculating magnification and their limitations is

given in Chapter 3. An investigation of the reading performance with magnification

determined by the two currently used methods was conducted in this study (Chapter

3).

After calculating the appropriate magnification, clinicians have to select an optical

low vision aid based on the patient’s need for reading. There are many commercially

available low vision aids, which are discussed below.

1.4.3 Optical and electronic aids

Optical low vision aids improve reading performance of people with visual

impairment by providing magnification to enlarge retinal images. With the assistance

of low vision aids, people with low vision are able to retain some reading ability

such that they can read small print (Dickinson, 1998; Colenbrander et al., 1999).

Some people even reported that they could read longer and faster with their

prescribed low vision aids than without (Herreros et al., 1998). Table 1.8 summarises

commonly used low vision aids and their principles of magnification (Dickinson,

1998; Woo, 2001).

Table 1.8 Overview of low vision aids

Types of low vision aids Examples of low vision aids Types of magnification

Spectacle lenses (high near addition) Relative distance

Hand-held magnifier Relative distance and angular Stand magnifier Relative distance and angular

Optical

Near telescope Relative distance and angular CCTV Projection + relative size

Electronic Head-mounted video magnification Projection


57

1.4.3.1 High near addition

A high near addition is a spectacle-mounted convex lens which enables patients to

use relative distance magnification with minimal limitation of the field of view

(Johnston, 1982). Some patients prefer this type of low vision aid rather than other

magnifiers, such as hand-held or stand magnifiers, because it is a more cosmetically

acceptable. It provides the widest field of view among all optical low vision aids and

both hands are free during use. However, a short working distance is required when

the lens power is high. This results in the major drawback of using a high near

addition, as many patients find the short working distance uncomfortable. Writing is

another difficulty when the lens is stronger than 10 D because the working distance

reduces to less than 10 cm (Dickinson, 1998).

1.4.3.2 Hand-held magnifier

A hand-held magnifier is a convex lens that a subject holds, by means of a handle, at

various distances from the spectacle plane using the principles of relative distance

and angular magnification. It is portable, inexpensive and easy to use especially for

quick spot reading, for example when shopping. The magnifier has to be held at the

focal distance of the magnifying lens to obtain a magnified image focused at infinityj

with accommodation relaxed; this may be uncomfortable for prolonged reading due

to fatigue of the hand and arm. In addition, steady hands and good eye-hand

coordination are essential to maintain a steady and fully magnified image. Distortion

may be increased in the periphery of the lens when the magnifier is held too far away

from the eyes, reducing the clear field of view (Dickinson, 1998).

Provided that the magnifier is placed at the focal point, the magnification of a hand-

held magnifier remains constant regardless of the eye-to-lens distance. However,

patients do not always use hand-held magnifiers in this way. The magnifier may be

positioned such that the object is closer to the magnifier than the focal point.

Divergent light, rather than parallel light, then emerges from the magnifier lens such

that the virtual image is formed between infinity and the patient's near point and

j If the magnifier is held at its focal distance, the image is formed at infinity. The presbyopic patient does not need a reading addition to focus on an image which is at infinity.


58

either accommodation or a near prescription is necessary to obtain a focused retinal

image. In this situation, the magnifying system is no longer a simple single-lens

system but a two-component magnifying system of the magnifying lens and reading

addition (for a presbyopic patient) (Dickinson, 1998). Figure 1.3 gives the ray

diagram of this two-component magnifying system and its magnifying effect. The

magnification produced by this combined system is usually less than the magnifying

effect produced by the simple plus lens locating at the focal point of the lens because

patients usually hold the magnifiers far away from their near additions. The

combined magnification is calculated by the following equations.

Equivalent power (Fe) = FM + FA - zFMFA

Magnification = 4Fe

According to the first equation, the equivalent power (Fe) depends on the magnifying

lens (FM), near addition (FA) and the distance between the near addition and the

magnifiers (z). When the distance between the near addition and the magnifier,

which is termed eye-to-lens distance decreases, magnification increases. Therefore, it

is essential to have an appropriate eye-to-lens distance achieving the required

magnification and a clear image.

l =object distance l’ = image distance z = eye-to-lens distance

Figure 1.3 Ray diagram of a hand-held magnifier where the object was located within the object distance

Magnifying lens

Near addition Eye

Object Focal point of FM

l

l’ z

FM FA


59

In addition to the magnifying effect, field of view of a magnifying lens is another

parameter that has to be considered. Field of view is defined as the angle subtended

by the lens aperture at the image of the eye’s entrance pupil or the number of

magnified characters that can be seen through the magnifying lens. It is limited by

the dioptric power of the lens and physical diameter of the magnifying lens (Jose,

1983). The stronger the magnifying lens, the larger the image size but the lens is

usually smaller to minimise the peripheral aberrations. As a result of these two

factors, the field of view decreases. In addition the field of view is inversely related

to the magnifier-to-eye working distance (z). Therefore, the patient is usually

recommended to hold the magnifier close to the eye (to reduce the magnifier-to-eye

distance) to maximise the field of view (and magnification) with the magnifiers

(Dickinson, 1998).

Field of view = (z) lens theofpower Equivalent

lens theofDiameter

1.4.3.3 Stand magnifier

The design of a stand magnifier is similar to that of a hand-held magnifier, with a

convex lens incorporated into a stand. The distance between the lens and the reading

material is fixed, such that the user is not required to hold the magnifier away from

the page to obtain a magnified image. The stand magnifier is therefore beneficial to

people with poor motor control or shaky hands. Other advantages of stand magnifiers

are the large range of magnifying powers available and the more comfortable

working distance from the reading material compared with the equivalent high near

addition. For young patients, the versatility of the distance between the stand

magnifier and the eye is an advantage. However, the restricted working distance

between the eyes and lens required to ensure sufficient magnifying effect and field of

view may result in posture and fatigue problems after prolonged reading (Bailey,

1981 (a); Bailey, 1981 (b); Bailey and Tuan, 1998).

The position of the lens system places the object plane inside the focal plane of the

lens. As a consequence, the image is not located at infinity. Therefore sufficient

accommodation or near addition are needed to focus this image at different distances.


60

The accommodation demand is determined by the distance between the eye and

image, which is the sum of the distances between the image and the lens and between

the lens and the eye. Most elderly patients need reading glasses while using stand

magnifiers, but the power of the reading addition has a major influence on the eye-

to-lens distance and therefore on the resultant magnification and field of view

(Bailey et al., 1994). The eye-to-lens distance must be taken into account during

stand magnifier prescription to achieve a clear and magnified image. Patients have

some freedom to change their eye-to-lens distance within the bounds of their

tolerance to defocus, but this results in changes in magnification and field of view.

Eye-to-lens distance and magnification are inversely related. Thus, when the eye-to-

lens distance increases, magnification and field of view decrease, as described for

hand-held magnifiers above. Details for the calculation and selection of appropriate

magnifiers used in this study are given in Chapters 3 and 4 and details of the

measurement of the optical parameters of stand magnifiers are given in Appendix 4.

In one low vision clinic, simple optical aids such as high additions, hand-held

magnifiers and stand magnifiers were prescribed to a significant proportion (56.8%)

of the patients with low vision (Leat and Rumney, 1990). This implies that the needs

of people who are visually impaired can be met with simple and low cost magnifiers

to assist reading. Sloan and Habel (1965) reported that about 45% of patients chose

stand magnifiers as their reading devices. Recently, a study in the United Kingdom

reported that 48.2% of prescribed magnifiers were illuminated stand magnifiers

(Doorduyn et al., 1998).

Despite the improved near resolution limit provided by magnifiers, patients with low

vision still have difficulty reading with magnifiers (hand-held and stand magnifiers)

and may not be able to attain a normal reading rate (Spitzberg et al., 1989). Reduced

field of view (Lovie-Kitchin and Bowman, 1985; Whittaker and Lovie-Kitchin,

1993) and difficulty with manipulating magnifiers on reading material (McMahon

and Spigelman, 1989; Spitzberg et al., 1989; Beckmann and Legge, 1996; Fine et al.,

1996; Bowers, 2000 (b)) are the two major problems that have been identified by

previous research studies in the use of magnifiers. A review of previous research on


61

the impact of field of view and manipulation of stand magnifiers in reading is given

in Chapter 5 (section 5.1.1).

The advantage of increasing illuminance for reading has been discussed in section

1.3.3.6.2. Prescribing magnifiers with built-in illumination is therefore believed to be

more beneficial for reading for people with visual impairment. Therefore,

illuminated stand magnifiers were adopted as the low vision aid prescribed to the

subjects who participated in this study (Chapter 4).

1.4.3.4 Near telescope

A near telescope is a complex optical system involving more than one lens. It can be

a spectacle-borne device, which focuses at near either by adding a plus lens cap

(permanent or removable) to a distance telescopic system or by having sufficiently

wide range of distance for focussing. It provides a longer working distance than high

near addition but a smaller field of view and narrower depth of field. Heavy weight,

poor cosmesis and a reduction in contrast due to the multiple lens system are the

major disadvantages of a near telescope. Hence it is not commonly prescribed in low

vision clinics compared to other low vision aids (Nowakowski, 1994).

1.4.3.5 Closed circuit television (CCTV)

Amongst the armoury of low vision aids available for reading, the closed circuit

television (CCTV) is an electronic low vision aid that provides greater magnification

than optical aids and allows the patient to have a more comfortable working distance

for reading. It consists of a video camera that transmits a “live” image with a zoom

lens and aperture giving the details of reading material on the monitor. Adjustable

magnification is available over a wide range with variable reading distances.

Therefore, CCTVs have been successfully used for reading tasks, providing high

magnification, good control over illumination, freedom to vary head position and

control of contrast and contrast polarity (Dickinson, 1998). However, not all of them

are portable and they are quite expensive which may make them unaffordable by

some people with low vision.


62

1.4.4 Non-optical aids

Apart from optical and electronic low vision aids, non-optical aids are also essential

in low vision rehabilitation in particular when the optical aid has some limitations for

the patients (Brown, 1997; Williams, 1999). There are many non-optical aids that

can facilitate low vision reading such as large print reading, filters and so on.

However, in this section, only illumination and reading stands which provide a more

comfortable reading environment without providing any magnifying effect are

discussed.

1.4.4.1 Lighting

Weale (1963) stated that older people receive only 33% of the illumination at the

retina received by young people. Because of neural changes, the yellowing of the

crystalline lens and the reduction in pupil size with age, older people need more

illumination. As a consequence of the reduced retinal illumination, older people need

two to three times more illumination than that required by younger people for

reading (Cole, 1974; Lie, 1977; Julian, 1984; LaGrow, 1986). In addition, research

studies have shown that the majority of people with visual impairment show an

improvement in vision functions, such as visual acuity and reading performance in

better illuminated environments (Sloan and Brown, 1962; Sloan et al., 1973; Lie,

1977; Lovie-Kitchin and Bowman, 1985). Cornelissen and colleagues (1994) noted

substantial improvements in object perception with increased light levels. They

pointed out that it was essential to provide optimum lighting conditions for people

with visual impairment to conduct different activities (van der Wildt et al., 1994).

The benefits of increasing the illuminance (or lighting) have been discussed in

section 1.3.3.6.2. In addition to increasing illuminance, control of glare and

unwanted shadows is essential to provide a comfortable and efficient reading

environment. Therefore patients with low vision are commonly advised to use

reading lamps to provide additional illumination which should be directed over the

shoulder of the better eye and be held close to the reading material for maximum

illumination.


63

1.4.4.2 Reading stand

A reading stand sits on a desk and holds reading material at a height and angle which

is adjustable and comfortable to the user. It provides a comfortable posture for the

very close reading distance required during the use of low vision aids, and reducing

the stress on the neck and back during prolonged reading. Because of the advantages

of the reading stand, reading materials or cards used in the current study for reading

assessment were all held on reading stand (Lovie-Kitchin and Bowman, 1985).

1.4.5 Reading performance with low vision aids

Although low vision aids allow people who are visually impaired to read smaller

print, the introduction of a low vision aid reduces the field of view available for

reading. The field is restricted by the physical aperture of the magnifier which limits

the number of magnified characters that can be seen through the lens. There have

been a number of studies showing that reading rate is significantly reduced when

magnifiers are used (Mancil and Nowakowski, 1986; McMahon and Spigelman,

1989; Neve, 1989 (b); Cohen and Waiss, 1991 (a); Cohen and Waiss, 1991 (b);

Fotinakis and Dickinson, 1994; Bowers, 2000 (a)) (Table 1.9).

Indeed restricted horizontal field of view and difficulty manipulating the magnifiers

during reading have been suggested as the main reasons for a reduction in reading

rate with magnifiers compared with reading rate on large print (Mancil and

Nowakowski, 1986; McMahon and Spigelman, 1989; Neve, 1989 (b); Cohen and

Waiss, 1991 (b)).

The relationship between the field of view and reading rates from previous studies

has been discussed in section 1.3.3.6.4. Lovie-Kitchin and Whittaker (1998 (a); 1999

(a)) suggested that the reduced field of view was not a limiting factor affecting

reading rate provided that sufficient magnification (acuity reserve) was available.


64

Table 1.9 Summary of previous studies on the reading rates with and without low vision aids on adults

Research study Sample LVA Training Reading

tasks Results Comment on LVAs provided

Comparison of magnification with and without LVA

Mancil and Nowakowski (1986)

40 highly educated NV

Spectacles HHM STM Telemicroscope

Minimal instruction N10

RR with LVA was significantly reduced by 27% to 71%

LVAs of different equivalent power were used as working distance was not considered

Retinal image sizes (or level of magnification) with and without LVA were not equivalent. Working distances and other parameters were not controlled.

McMahon and Spigelman (1989)


17.6 D STM Brief training Not reported

Optimal RR was significantly reduced by the use of STM

Retinal image sizes with and without LVA were not confirmed to be equivalent. Working distance (between eye-to-lens) was not controlled which might result in different EVD between visits.

Cohen and Waiss (1991 (a))



Brief training N9

LVAs (except the spectacles with near addition) significantly reduced RR by 6% to 19%

LVAs of the same equivalent power were used (12D)

Retinal image sizes (level of magnification) with and without LVA were not equivalent.

Cohen and Waiss (1991 (b))



Brief training N9


Different equivalent power but same field of view of LVAs



65

Table 1.9 Summary of previous studies on the reading rates with and without low vision aids on adults (continued)




Koenig et al. (1992)

6 LV (students)

Subjects’ own low vision aids

Experienced

N18 (Without LVA) Regular print (with LVA)

No significant difference

Similar retinal image sizes used to measure RR with and without LVA had been considered.

The print sizes of the two reading tasks were not confirmed that they provided sufficient acuity reserve to achieve maximum or fluent RR.

Bowers (1994) 16 NV HHM Spectacle-mounted

Inexperienced N10 (1.2M)


Both LVAs were +16 D

Retinal image sizes (level of magnification) with and without LVA were not equivalent. The distance between the HHM and the text was not controlled, which might result in different EVD for the two LVAs.

Fotinakis and Dickinson (1994)


HHM of 3x and 6x Inexperienced N10

RR with LVA was significantly reduced by 19% (3x) and 34% (6x)

Working distance control (eye-to-lens) was attempted for maximum magnification.


Rumney and Leat (1994)

27 LV 7 NV


Subjects with LV were experienced

0.2 logMAR larger than near acuity (NV) and those with LVA (LV)

RR with LVA was significantly reduced by 55%

RR with LVA in LV people was compared with RR without LVA in NV people (unpaired sample)

Baseline RR for people with NV was faster than those with LV. The reduction of RR due to LVA for subjects with LV appeared to be more significant because it was compared to the RR for subjects with NV with no LVA.


66

Table 1.9 Summary of previous studies on the reading rates with and without low vision aids on adults (continued)




Ahn and Legge (1995) 40 LV


Experienced

Text at N12 with LVA and at CPS without LVA


Retinal image sizes (level of magnification) with and without LVA were equivalent.

Ortiz et al. (1999) 10 LV

Head-Mounted Video Magnifier (low-vision enhancement system)

Inexperienced

Range of different print sizes (sentence)


Maximum RR with and without LVA was compared.

The retinal image sizes with and without LVA were equivalent.

Bowers (2000 (a)) 20 NV

HHM Spectacle-mounted

Inexperienced N10

RR with LVA was significantly reduced by 35%


Dickinson and Fotinakis (2000)

12 NV HHM Inexperienced N10 N18


Same EVD was used for each HHM


Lovie-Kitchin et al. (2000 (a))

22 LV (MD)


Experienced Text at CPS No significant difference

Retinal image sizes (level of magnification) with and without LVA were equivalent.

Subjects with low vision in the sample size mainly had mixed causes of low vision unless otherwise indicated. NV – Normal vision LV – Low vision HHM – hand-held magnifier STM – stand magnifier LVA – low vision aid RR – reading rate EVD – equivalent viewing distance CPS – critical print size


67

In support of this some other research studies have found that reading with a low

vision aid does not significantly reduce reading rate compared to that achieved

without a low vision aid (Ahn and Legge, 1995; Lovie-Kitchin et al., 2000 (a)). The

reasons for the conflicting findings (with respect to whether or not magnifiers reduce

reading rates) are mainly due to the differences between studies in terms of

characteristics of the participants, the amount of experience in reading with

magnifiers and levels of retinal magnification achieved with and without low vision

aids.

Those studies that showed significant reductions in reading rate as a consequence of

magnifier use assessed subjects with normal vision rather than low vision (Mancil

and Nowakowski, 1986; McMahon and Spigelman, 1989; Neve, 1989 (a); Neve,

1989 (b); Cohen and Waiss, 1991 (a); Cohen and Waiss, 1991 (b); Bowers, 2000

(a)). These people were inexperienced in reading under restricted field of view and in

manipulating the magnifiers for reading. Therefore it is reasonable that their reading

rates would reduce when a magnifier was introduced since their habitual reading was

interrupted. However, those studies which showed no significant difference in

reading rate with and without magnifiers assessed subjects with low vision who were

experienced in using low vision aids for reading (Ahn and Legge, 1995; Lovie-

Kitchin et al., 2000 (a)). Thus, reading rate with magnifiers was not significantly

different from reading rate without a magnifier but with large print, provided the

magnification levels were equal for the two conditions.

In addition, the maximum reading rate for people with normal vision is faster than

that for subjects with low vision even when the print size is optimal (Legge et al.,

1985 (a); Legge et al., 1985 (b); Bullimore and Bailey, 1995; Fine and Peli, 1995;

Beckmann and Legge, 1996; Legge et al., 1997). Therefore, any change for people

with faster reading rates will result in a more obvious reduction in reading rate

compared with people with slower reading rates.

Different retinal image sizes provided with and without the low vision aid is another

reason for the significant reduction in reading rate when reading with a low vision

aid. The majority of previous studies did not use the same magnification to compare


68

the reading with and without low vision aid (Mancil and Nowakowski, 1986; Cohen

and Waiss, 1991 (a); Cohen and Waiss, 1991 (b); Bowers and Ackerley, 1994;

Fotinakis and Dickinson, 1994; Dickinson and Fotinakis, 2000; Bowers, 2000 (a))

(Table 1.9). This was because the subjects in these studies were all normally sighted,

who could read small print without magnification. Often, the test materials used to

measure the reading rates with and without low vision aid were the same, suggesting

that the retinal magnification was much greater with the magnifier than without.

Reading rate for adults with normal vision is stable over a wide range of print sizes

from 0.5 to 1.5 logMAR (N10 to N100 at 40 cm), but decreases at print sizes greater

than 1.5 logMAR or smaller than 0.5 logMAR (Legge et al., 1985 (a); Legge et al.,

1988; Legge, 1991). This range of print size varies slightly across a few studies in

which different types of reading materials and different modes of presentation were

used (refer to section 1.3.3.6.1) (Bailey and Lovie, 1980; Lovie-Kitchin and Woo,

1987; Lovie-Kitchin et al., 1994). An increase in print size does not significantly

affect the reading rate for normally sighted people if the enlarged (or magnified) text

is within the suggested range of print size for maximum reading rate. In general,

previous studies used reading tasks of 9- or 10- point print to compare the reading

without a low vision aid and the reading achieved with a low powered aid (e.g. 3 - 4x

magnification) on people with normal vision. The size of the magnified image was

approximately 30-point print with a 3x magnifier or 40-point print with a 4x

magnifier, which were well below the largest character size for maximum reading

rate (N100). For this reason, the effect of different retinal image sizes on reading

rates with and without a low vision aid should perhaps be minimal as the print sizes

of the magnified images are all less than N100. There has been no research

investigating which of these suggested reasons – lack of experience in reading under

restricted field of view, difficulty in manipulating the magnifiers and/or different

levels of retinal magnifications with and without a low vision aid, is the major reason

for the reduction in reading rate with low vision aid for subjects who are normally

sighted. Because of the interaction between these three factors, it is very difficult to

separate them for further investigation.


69

In addition to the change in reading rate, reading with a magnifier induces changes in

eye movements, from typical “staircase” pattern to “optokinetic nystagmus-like”

(OKN-like) eye movements (Cummings et al., 1989; Bowers and Ackerley, 1994;

Fotinakis and Dickinson, 1994). The reduction in reading rates with magnifiers for

people with normal vision has been attributed to the increase in the number of

forward saccades per line (Bowers and Ackerley, 1994; Fotinakis and Dickinson,

1994) and to a lesser extent, to the increase in time required for page navigation

(retrace) (Bowers and Reid, 1997; Bowers, 2000 (a)). For people who are visually

impaired, the reading rate is comparatively slower than for people with normal vision

because more saccades and more frequent regressions are made due to the visual

impairment (Bullimore and Bailey, 1995) (section 1.3.3.4). When a low vision aid is

used to assist their reading, the forward saccade length (and forward saccade

numbers) is not significantly different from reading without the low vision aid

(Bullimore and Bailey, 1995). However, the time for page navigation (retrace) is

significantly increased when the low vision aid is employed (Bullimore and Bailey,

1995; Bowers et al., 2001 (b)). The increase in retrace time due to low vision aid use

for subjects with visual impairment is less than that for subjects with normal vision

(Bowers et al., 2001 (b)). The reading rates with and without low vision aids do not

change significantly provided that the subjects are experienced magnifier users

(Lovie-Kitchin et al., 2000 (a); Bowers et al., 2001 (b)).

In the literature, to the author’s knowledge, there has been no study comparing the

reading performance with and without low vision aids for people with visual

impairment who have no experience in using magnifiers for reading. The participants

recruited in previous studies were either people with normal vision or people with

low vision who were experienced in using magnifiers. Based on the previous results

for people with normal vision, reading rate for people with visual impairment may be

significantly reduced when they are first prescribed a low vision aid. This may be

due to their inexperience in reading under restricted field of view and difficulty in

manipulating the magnifier for reading (presuming that the magnification levels with

and without the low vision aid are the same). This study was designed to compare

reading rates with and without a low vision aid for people who were visually

impaired, for the period when they were adapting to using the device for reading. The


70

duration of magnifier adaptation required for these people to achieve a magnifier-

aided reading rate equivalent to the reading rate without low vision aid was

determined (see Chapter 4).

1.5 Low vision training and rehabilitation for reading

1.5.1 Training

The success of low vision reading rehabilitation depends on whether the goal

determined by the patient with low vision can be achieved. Some patients defined

their reading goal as “being able to read newspaper”. Prescribing low vision aids is

one of the most common forms of low vision rehabilitation.

In the literature, numerous programs have been suggested to train people with AMD

to use eccentric viewing and to develop their visual skills when reading (Table 1.10).

Similarly, a number of training programs have been proposed to alleviate the

problem of reading with magnifiers (Goodrich et al., 1977; Goodrich and Mehr,

1986; Nilsson and Nilsson, 1986; Nilsson, 1988; Nilsson, 1990; Solan et al., 1995;

Stoll et al., 1995; Nilsson et al., 1998; Goodrich et al., 2000).

1.5.1.1 Improvement of visual skills

Patients with established central scotomas develop eccentric viewing at extra-foveal

locations, which means a PRL is adopted to provide better vision function (Guez et

al., 1993; Trauzettel-Klosinski et al., 1995; Sunness et al., 1996). Previously,

extensive eccentric viewing training schemes of 10 to 15 sessions, of 30 to 60

minutes duration were suggested by Goodrich and colleagues (1977; 1986) to aid the

development of subjects’ visual skills when reading. This results in high costs for

low vision clinics and for patients, as low vision practitioners are required to conduct

the training.


71

Table 1.10 Summary of previous studies on training eccentric viewing and/or the use of low vision aids

Research

study Sample Duration of training

Type of training

Total length Results Remark Comments

Goodrich et al. (1977)

24 LV 10 days (50 minutes per day)

Use of LVA (CCTV, optical LVA)

10 days

RR and reading duration significantly increased with CCTV and optical LVAs

The improvement in RR for the CCTV group was more apparent than that for the optical aid group

The 12 optical LVAs prescribed were only near additions, with no HHM and STM. There was no control group to compare people with no training.

Goodrich and Quillman (1977)

25 10 days Eccentric viewing 10 days

Both RR and duration increased with practice and training

Significant improvement in RR was found at days 1 and 2 followed by a gradual increase.

There was no control group to compare people with no training.

Nilsson (1986) 79 (DR)

6.75 sessions (1 hour per session)

Eccentric viewing (23% subjects received this training) Use of LVA

3.6 years

Significant improvement of distance and near VA with LVA, ability to read TV (64.6%) titles and newspaper text (86.1%)

There was no control group to compare people with no training

Nilsson and Nilsson (1986)

120 (AMD)

7.5 visits (with 1 hour each)

Eccentric viewing Use of LVA (distance and near)

5 years

Significant improvement in distance and near VA with LVA, ability to read TV titles (49.2%) and newspaper text (80%)

The near LVAs were near additions and hyperocular lenses, not HHM or STM. There was no control group to compare people with no training

Nilsson (1988) 96

No. of hours varied based on different LV causes

Eccentric viewing Use of LVA

3.6-6 years

Significant improvement in distance and near VA with LVA (>95%)

Due to the varied low vision causes, not many subjects required eccentric viewing training

No assessment regarding the reading on daily activities (e.g. newspaper text) as included in previous studies.


72

Table 1.10 Summary of previous studies on training eccentric viewing and/or the use of low vision aids (continued)

Research

study Sample Duration of training

Type of training

Total length Results Remark Comments

McMahon and Spigelman (1989)


Brief training STM 2 weeks

Significant improvement of STM RR after two weeks of practice.

Working distance (between eye-to-lens) was not controlled which might result in different EVD in between visits. The improvement in RR might be due to the increase in magnification rather than by practice or training. Subjects recruited had normal vision, not low vision.

Nilsson (1990)

40 (AMD)

4.8 sessions (1 hour)

Eccentric viewing (50% subjects) Use of LVA

1 month

Significant improvement in ability to read TV titles and newspaper text and to write letters for the trained group. (100% for the trained group compared to 25% for the control group)

Control group with instruction given was included in the study

Nilsson et al. (1998) 6 (AMD)

4-5 hours (1 hour per week)

Eccentric viewing (for another more favourable eccentric position)

1 month

83% (5 subjects) could use their new eccentric PRL for reading The RR at this new PRL was significantly faster than the old PRL.

The sample size was too small for statistical analysis. No longitudinal assessment on the stability of the new PRL was conducted.


73

Table 1.10 Summary of previous studies on training eccentric viewing and/or the use of low vision aids (continued)

Research study Sample

Duration of

training Type of training Total

length Results Remark Comments

Bowers (2000 (b)) 3 NV 2 days (3

sessions) Brief instruction of STM used

Short-term

Significant improvement in STM RR

Significant improvement of RR in first 2 sessions

Short-term learning effect was assessed in this study. Subjects recruited had normal vision, not low vision.

Goodrich et al. (2000; 2001)

90 AMD 3 training models k

Optical LVA CCTV

Not reported

Optical LVA needed 5 training sessions while training for CCTV required 7 sessions (Empirical model). RR and reading duration increased significantly by training across time.

Control group, which was named as “private sector” model with minimal training, was included and showed the improvement in RR and reading duration was less than that in training groups.

Baseline RR was measured after the 1st session of training rather than just after the provision of LVAs, so RR before any training was unknown. No direct comparison of RR could be made between control and training groups.

Sample size for subjects with low vision (LV) were mainly with mixed causes of low vision unless otherwise indicated. DR – diabetic retinopathy AMD – age related macular degeneration NV – normal vision RR - reading rate VA - visual acuity LVA - low vision aids HHM - hand-held magnifiers STM - stand magnifiers CCTV - closed circuit television TV - television PRL – preferential retinal locus EVD – equivalent viewing distance

k Three training models were used in this study: Traditional model used in the Department of Veteran Affairs (as per the program in 1977: 10 sessions of optical LVA training and 15 hours of CCTV training); Empirical model (five sessions of optical LVA training and five sessions of practice; and then with seven hours of CCTV training and 8 hours of practice); Private sector model (1 session of optical LVA training and 2 CCTV training sessions and three practice sessions).


74

Nilsson and colleagues (1986; 1986; 1988) not only emphasised the importance of

educational training in the use of residual vision (eccentric fixation) but also training

in the use of low vision aids. A few different training programs for patients of

different low vision causes have been suggested (Nilsson, 1986; Nilsson and Nilsson,

1986; Nilsson, 1988), which consisted of fewer sessions than the programs proposed

by Goodrich et al. (1977; 1986). However, the training in Nilsson's programs lasted

for a number of years (3 to 6 years). Because of the involvement of long training

periods, it was not surprising that vision and therefore the reading rate decreased at

the last visit although the majority of subjects showed significant improvement in

threshold resolution (for both distance and near) and in magnifier reading

performance compared with that at their first visit (before any training). In addition,

no control groups were included in Goodrich's and Nilsson's studies to compare the

objective measures and/or subjective reports on the ability to perform daily activities

for the subjects who received training compared to subjects who received no

training. To overcome these drawbacks, a 1-month training program was devised

(Nilsson, 1990). This study compared the ability to read TV titles and newspaper text

and to write letters for subjects in a training group with those in a control group

(Nilsson, 1990). There was a significant improvement in managing these activities

reported by people in the trained group compared with those in the non-trained

(control) group. However, no objective measures such as reading rate or threshold

print size achieved with low vision aids were made.

A number of research studies have reported that people with central field loss prefer

to place their PRL in the left visual field (i.e. scotoma to the right of fixation) (Guez

et al., 1993; Trauzettel-Klosinski et al., 1994; Sunness et al., 1996; Fletcher and

Schuchard, 1997; Fletcher et al., 1999). As discussed in section 1.3.3.3 this location

for the PRL is expected to reduce reading rate. To minimise the effect of this

inappropriate location of the PRL on reading, Nilsson and co-workers (1998)

proposed a training program (4 to 5 hours) for patients with AMD to re-establish the

PRL to a better location for reading. However, the number of subjects (n=6) who

participated in this study was too small for statistical analysis. No longitudinal

assessment on the stability of the new PRL was included, such that the benefit of the

program on training a new PRL was questionable. A few studies have shown that


75

patients with central field loss develop a PRL even though they received no specific

training (Timberlake et al., 1986; Schuchard, 1995; Fletcher and Schuchard, 1997).

Timberlake et al. (1986) found that patients with macular scotoma did not attempt to

use the non-functional fovea. Instead, they used a single, non-damaged retinal area,

immediately adjacent to the scotoma to fixate. Fletcher and Schuchard (1997) came

to a similar conclusion when they performed SLO on patients with macular scotoma.

They pointed out that the majority of patients with low vision, as many as four of

five patients (80%), used an eccentric PRL for visual tasks (Fletcher and Schuchard,

1997). As people with AMD may develop their own PRL, the benefit and

effectiveness of training programs in developing eccentric fixation is arguable.

In addition to training eccentric viewing and training in the use of low vision aids,

reading practice without a low vision aid has been suggested for improving low

vision subjects' reading skills (Watson et al., 1992; Stoll et al., 1995). Watson et al.

(1992) compared reading comprehension in a group of subjects with AMD who were

required to read for 10 minutes daily (practice group) to another group of AMD

subjects who received twice weekly planned reading lessons for one hour each

(training group). Results indicated that daily reading practice could provide similar

improvements in reading comprehension compared to that achieved by the training

group (Watson et al., 1992). This is probably because daily reading practice shifts

subject’s attention from decoding to extracting meaning from the text, which

improves reading comprehension (Samuels, 1979; Layton and Koenig, 1998). For

people using low vision aids, reading practice can allow subjects to develop more

effective reading skills and so increase their reading performance (Corn and Ryser,

1989; Koenig et al., 1992).

1.5.1.2 Training for the use of low vision aids

Apart from eccentric viewing training and reading practice, instructions on the use of

magnifiers, such as shortening the eye-to-magnifier distance for greater field of view

and the use of an optimum focal distance have been recommended for people with

low vision to enhance magnifier use (McIlwaine et al., 1991; Freeman and Jose,


76

1997; Dickinson, 1998). If training and instruction are given in the use of magnifiers,

incorrect use or rejection of magnifiers can be reduced (Langmann et al., 1994).

Goodrich et al. (1998; 2000; 2001) suggested that training and practice in using low

vision aids could improve magnifier reading rate for patients with low vision. They

suggested a training program (5-10 days), which was less intensive than the previous

programs suggested between 1970 and 1990 (15 days) in the use of low vision optical

aids (Goodrich et al., 1977; Goodrich et al., 1980; Goodrich and Mehr, 1986).

Among the proposed training schemes, no specific period of time or training

procedures have been suggested for training people with visual impairment to use the

low vision aids for reading. In addition, most of the research studies investigating the

effect of training on low vision reading with magnifiers have included the

development of their visual skills (training for eccentric viewing). It is difficult to

determine which type of training - training eccentric viewing or training the use of

low vision aids - has the most effect on improving reading performance with low

vision aids. Most of the programs to train the use of low vision aids have focussed on

near additions (Table 1.10). There have been only two training studies which used

stand magnifiers as the low vision aid to study the effect of training on reading rate,

but the subjects recruited in these studies were young adults with normal vision

rather than people with visual impairment (McMahon and Spigelman, 1989; Bowers,

2000 (b)). There have been no general guidelines suggested from previous studies on

training patients with low vision using stand magnifiers for reading, therefore further

research in this area is needed, especially as stand magnifiers are probably the most

commonly prescribed low vision aid (Doorduyn et al., 1998).

The ultimate goal of training patients with low vision to use magnifiers is to provide

an easy and efficient way to improve reading performance with the low vision

devices. Watson and her colleagues (1992) suggested that reading practice could

provide similar improvements in reading performance to that achieved by

comprehensive reading training. Therefore, research aimed at investigating the effect

of daily reading practice on subjects’ reading performance with magnifiers was

conducted in this study. Bowers (2000 (b)) found an improvement in reading after


77

short-term practice of using stand magnifiers, but this was for subjects with normal

vision not low vision. An experiment to investigate the effect of reading practice

(large print) prior to the prescription of stand magnifiers, on reading performance

with stand magnifiers for subjects with visual impairment was conducted in this

study (Chapter 4).

1.6 Objectives of this study

The main aim of this thesis was to investigate the reading performance with stand

magnifiers and the effect of reading practice on magnifier reading performance of

people with AMD.

The specific objectives of this research program were to:

(a) Select an appropriate reading chart to monitor reading performance (Chapter 2).

(b) Derive prescription guidelines for determining the appropriate magnification to

be prescribed for the subjects with AMD (Chapter 3).

(c) Investigate the effect of large print reading practice, with and without a restricted

field of view prior to the prescription of stand magnifiers, on the reading

performance with stand magnifiers for subjects with AMD (Chapter 4).

(d) Investigate the effect of a line guide on the reading performance with stand

magnifiers for subjects with AMD (Chapter 5).

(e) Provide recommendations for training patients with AMD who are prescribed

stand magnifiers (Chapter 6).

The reading chart and the method to determine the appropriate magnification derived

from the results reported in Chapters 2 and 3 respectively were adopted to measure

reading performance and to select the appropriate magnifiers for subjects who

participated in the experiments reported in Chapters 4 and 5.

The overall hypothesis of this study was that reading practice prior to the

prescription of stand magnifiers would improve reading performance with the stand

magnifiers for subjects with AMD.

Chapter 2 Validation of reading chart for measuring reading performance

78

CHAPTER 2

Validation of reading chart for measuring reading

performance

2.1 Introduction................................................................................................. 79 2.1.1 Assessing reading performance............................................................... 79 2.1.2 Relation between reading rate and print size........................................... 81 2.1.3 Reliability of the reading charts .............................................................. 84 2.1.4 Aim of study............................................................................................ 84

2.2 Subjects ........................................................................................................ 85 2.2.1 Sample size.............................................................................................. 85

2.3 Methods........................................................................................................ 86 2.3.1 Reading charts ......................................................................................... 86 2.3.2 Procedures ............................................................................................... 88

2.3.2.1 Distance visual acuity....................................................................... 88 2.3.2.2 Near visual acuity............................................................................. 88 2.3.2.3 Reading rate...................................................................................... 89

2.4 Analysis ........................................................................................................ 90

2.5 Results .......................................................................................................... 91 2.5.1 Repeatability of reading measures .......................................................... 91 2.5.2 Agreement of reading performance among charts .................................. 95 2.5.3 Comparison of reading parameters calculated by two methods – MNRead

Analysis 0.3 and graphical methods ..................................................... 99

2.6 Discussion................................................................................................... 102 2.6.1 Validation of the reading charts ............................................................ 102 2.6.2 Relationship of the reading measures analysed by MNRead Analysis and

graphical methods ............................................................................... 104

2.7 Conclusion ................................................................................................. 105


79

2.1 Introduction

2.1.1 Assessing reading performance

Several methods are available to assess a person’s reading performance such as

evaluation of reading eye movements, cognitive comprehension tests and reading

rate (Legge et al., 1985 (a); Carver, 1990; Bradley-Klug et al., 1998). Carver (1990)

suggested that reading rate was an “inextricable dimension that reflected the

accuracy of comprehension”. Maximum reading rate as an index for measuring

reading performance has been commonly used in previous research studies as it is

easy to apply, highly reproducible and sensitive to stimulus and ocular variables

(Carver, 1990; Legge, 1991; Legge et al., 1992; Lovie-Kitchin et al., 2000 (a)).

Subjects are often instructed to read as fast as possible to determine maximum

reading rate (Legge et al., 1985 (b); Culham et al., 1992; Lovie-Kitchin et al., 1994),

in which case reading rate is more concerned with the mechanics of the reading task

than with comprehension of the text (Dickinson and Rabbitt, 1991). To better reflect

reading performance in the real world reading, reading rate for understanding –

"rauding" – has been recently used to measure individuals’ reading performance

(Lovie-Kitchin et al., 2000 (a)). Rauding, the 3rd “gear” of the five reading gears

proposed by Carver (1990), involves the requirement of comprehension although

comprehension of the complete thoughts contained in sentences of textual material is

not assessed as such. Oral reading rate is relatively constant across a range of levels

of text difficulty (Zumber and Wetzel, 1981; Carver, 1983), however, reading rate

reduces if the reading text is too difficult to comprehend. Therefore, the readability

of the material selected to measure reading rate should be well below readers’ ability,

such that the level of difficulty of the text is not a limiting factor affecting reading

rate.

Reading rate is simply measured by counting the number of words correctly read

over the time spent to complete the task. Carver (1990) recommended that reading

rate should be measured in standard length words which are comprised of six

characters.


80

Aquilante (1998) investigated the repeatability of reading rate measures and found

maximum oral reading rate was a reliable measure of reading performance for people

with normal and low vision. However, the reading text used in her study was

presented on a computer monitor rather than on paper. Measuring reading rate by

computer is not convenient in clinical practice. Reading charts (of static text) are

more commonly used to measure reading rate for clinical purposes. There are many

different reading charts available with different formats and layouts (Dickinson,

1998). The charts commonly used for the assessment of reading performance of

patients with low vision (Dickinson, 1998) are Bailey-Lovie word charts (Bailey and

Lovie, 1980), and Minnesota Low-Vision Reading charts (MNRead) (Legge et al.,

1989 (a); Mansfield et al., 1993; Mansfield et al., 1994; Ahn et al., 1995).

Unlike the Bailey-Lovie word chart, the MNRead chart use meaningful sentences as

the reading task. Reading rate for meaningful sentences of print is faster than reading

rate for unrelated words because both the reading strategy and the level of

comprehension used for the two reading tasks are different (Carver, 1990). Carver

(1990) suggested that for the five levels (or gears) of reading, reading rates varied

from 600 to 138 words per minute (wpm). Reading rate for materials of meaningful

text (passages) is faster than that for unconnected words because of the redundancy

of the coherent text (Carver, 1990). Measuring reading rate using sentences can

reflect people’s daily reading performance as meaningful text, rather than unrelated

words, is commonly read. For this reason, Bailey (2000) designed two text versions

of the Bailey-Lovie charts using sentences which were modified from the MNRead

charts - Bailey-Lovie text reading chart.

The designs of the MNRead and Bailey-Lovie text reading charts are similar except

for the page justification and the layout of the charts (Table 2.1). The sentences from

the Bailey-Lovie text reading chart are all left justified rather than centrally-aligned

which is the justification of the sentences from the MNRead chart. Page justification

of common everyday printed materials (such as newspaper or book) is mostly left

aligned or justified for both the left and right margins. Thus, reading rate measured

by the left-justified sentences from the Bailey-Lovie text reading chart is better

related to real world reading. Moreover, the dimensions of the Bailey-Lovie text

reading chart are similar to the size of an A4 card (21 x 29.7 cm) in landscape


81

orientation compared with the MNRead chart which is portrait orientation with a size

of 27.9 x 35.6 cm (similar to the size of an A3 paper). As such, the Bailey-Lovie text

reading chart is easier to set up for experiments and fit onto a reading stand for

measuring reading performance at a controlled distance. Apart from the different

sentence layout between the charts, the sentence structure, sentence composition and

font used are similar. Thus reading performance obtained with these two reading

charts would be expected to be very similar. However, there has been no previous

study validating this reading chart for assessing people's reading performance or to

compare the reading performance between these two reading charts.

Table 2.1 Comparison of formats and layouts of the reading charts

Bailey-Lovie text reading Chart MNRead Chart

Range of print sizes (logMAR) 1.1 to –0.2 1.3 to -0.5

Format of text Sentences Sentences Total number of words at

each print size 10 10

Number of lines for each print size 2 3

Text alignment Left Central

2.1.2 Relation between reading rate and print size

Print size is one of the many variables affecting reading rate (Legge et al., 1985 (a);

Legge et al., 1985 (b)). People with either normal or low vision cannot read fluently

if the print size of the reading material is at, or close to, their resolution limit (Legge

et al., 1985 (a); Legge et al., 1985 (b); Legge et al., 1992; Whittaker and Lovie-

Kitchin, 1993; Bullimore and Bailey, 1995). By plotting the reading rates versus

different print sizes, a broad peak of maximum reading rate (MRR) at a range of

print sizes is found (Figure 2.1). The maximum oral reading rate is limited by

reader’s ability, difficulty of the reading material and the amount of information to

be extracted (Carver, 1990). However, these parameters would not affect the result of

this study as all participants in this study were highly educated with no difficulty

understanding the reading material which was sixth grade reading level or below.


82

However, when print size decreases beyond a particular print size - critical print size

- reading rate declines rapidly. Critical print size (CPS) is the smallest print size at

which MRR is achieved (Legge et al., 1989 (a); Legge et al., 1992). Previous studies

have shown that for adults with normal vision, CPS is between 0.3 and 0.6 logMAR

(N6 and N12 at 40 cm) (Legge et al., 1985 (a); Lovie-Kitchin and Woo, 1987;

Whittaker and Lovie-Kitchin, 1993; Lovie-Kitchin and Whittaker, 1998 (a)). These

print sizes are the sizes of the reading materials that people commonly read daily. For

example, the print sizes for newspaper and normal print book are 8 point print (N8)

and 10-point print (N10) respectively (Lovie-Kitchin and Whittaker, 1999 (b)).

MRR is usually calculated as the mean of the reading rates at and above the CPS

(Figure 2.1). Mansfield and colleagues (1994) designed a computer program -

MNRead Analysis 0.31 to calculate the MRR and CPS by entering the reading times

at different print sizes into the computer. This program searches for the “plateau” in

the reading rates across print sizes by using pair-wise combinations and comparisons

of all reading rates. By calculating the mean and standard deviation of the reading

rates for the print sizes between the candidate pair of data points, reading rates for

print sizes which are at least 2 standard deviations slower than the mean reading rate

on the plateau are excluded from the plateaua. The MRR is the geometric mean of the

reading rates included in the plateau and the CPS is the smallest print size on the

plateau.

Alternatively, reading rate as a function of print size can be plotted (graphical

method). A smooth curve is fitted ‘by eye’ to the data points. This curve is comprised

of three segments – a plateau line for maximum reading rate, a turning point where

the reading rate begins to reduce (CPS) and another line where reading rate reduces

as print size decreases. CPS can be determined as the smallest print size, to the

nearest whole line, at which the reading rate starts to reduce. The MRR can then be

calculated as the mean of the reading rate at and above CPS. For example in Figure

2.1, the CPS was N12 while the MRR was the mean of the reading rates for print

sizes at N12 and above. Determination of CPS was not difficult for most readers as

there was usually a sharp cut-off point at which reading rate reduces substantially

a http://vision.psycho.umn.edu/www/people/stevem/mnr/MNA03.html (December, 1997)


83

with a decrease in print size. For some readers whose reading rate did not decline

obviously in the graphical display, the print size at which reading rate was 30% faster

than the reading rate at a smaller print size (0.1 log unit less) was considered as the

CPS. As CPS is critical to determine the number of reading rates included in

calculating the MRR, any difference in CPS could result in different MRR.

Generally, it would be expected that there would be no significant difference

between the CPS determined from the Analysis 0.31 and the graphical display as the

concept in determining the CPS and MRR is similar in both methods. However, there

has been no previous study to compare the reading measures obtained from MNRead

Analysis and the graphical method. If the reading measures given by these two

analyses are the same, then the graphical method would be preferred for clinical

purposes because no computer and software would be required, which is more cost-

effective and convenient.

Reading rates versus print sizes

1

10

100

1000

64 48 40 32 24 20 16 12 10 8 6 5 4

Print size (N-point)

Rea

ding

rate

(wpm

)

Figure 2.1 Example of the determination of maximum reading rate (MRR), critical print size (CPS) and text visual acuity (VA).

MRR is the mean of the reading rates for print sizes at and above the CPS. Reading rate reduces significantly for print size smaller than CPS (N12) until the smallest print size that the person can resolve, which is the text VA (N6).

Text VA

CPS

MRR = Mean of these reading rates


84

2.1.3 Reliability of the reading charts

In recent years reading performance has been used as one of the tests of vision

function, in particular for people with visual impairment. The reliability of reading

parameters measured with reading charts is important in clinical low vision

assessments and related research studies for assessing the change in reading

performance across time. It is also particularly important because reading rate has

been found to be one of the variables predicting performance with low vision aids

(Ahn and Legge, 1995; Lovie-Kitchin et al., 2000 (a)). In addition, there is a strong

correlation between reading rate and vision measures (e.g. near visual acuity)

(Bullimore and Bailey, 1995; Bowers, 1998 (a); Lovie-Kitchin et al., 2000 (a)).

Reliability of reading performance parameters can be assessed by comparing

repeated measurements of reading performance on the same group of subjects.

Previous studies have investigated the reliability of MNRead charts using people

with normal (Mansfield et al., 1993) and low vision (Mansfield et al., 1993; Bane et

al., 1996). To the author’s knowledge, there has been no study on the reliability of

the reading measures obtained when using the Bailey-Lovie text reading chart. To

determine whether the reading parameters obtained from the Bailey-Lovie text

reading chart were as reliable as those obtained from the MNRead chart (and hence

validate the use of the Bailey-Lovie text chart), both reading charts were included in

this study. Provided that no significant differences in reading parameters were found

between the two reading charts, the Bailey-Lovie text reading chart would be the

preferred chart for measurement of reading performance in subsequent studies (for

reasons previously listed).

2.1.4 Aim of study

The primary aim of this study was to validate the reliability of the Bailey-Lovie text

reading chart for assessment of reading performance. In addition, a secondary aim

was to compare the MNRead Analysis 0.31 and graphical methods for calculating

MRR and CPS.


85

As a first step in the validation of a measurement tool, it is important to determine

the inherent reliability using a homogeneous sample of subjects who are unlikely to

introduce any additional confounding factors that could affect the measurement of

interest. For this reason, people with normal vision rather than low vision were

selected as the subjects for this validation study. Using people with normal vision

should limit the introduction of confounding factors that might affect reading

performance (e.g. the fluctuation of vision, change in the reading behaviour).

The experimental hypotheses of this study were:

1. There would be no differences in reading parameters (MRR and CPS)

measured from MNRead and Bailey-Lovie text reading charts.

2. There would be no differences in reading parameters (MRR and CPS)

calculated by the MNRead Analysis 0.31 and graphical methods.

2.2 Subjects

Eighteen subjects with normal vision aged between 21 and 57 years (mean age 36.6

± 12.7 years) were recruited from among staff and students of the School of

Optometry, Queensland University of Technology. Subjects were all native English

speakers with high educational backgrounds and no known cognitive or linguistic

problems. Distance visual acuity for all subjects was better than 0.0 logMAR (6/6)

with a range of -0.02 to -0.20 logMAR (6/6+1 to 6/3.8) and no known ocular disease

(Table 2.2).

2.2.1 Sample size

In a previous study, mean reading (rauding) rate for people with normal vision and

young age was reported to be 213.6 ± 38.2 wpm (2.33 ± 0.08 log wpm) (Bowers,

1998 (b)). Based on the standard deviation of reading rate from this study, a sample

size of 17 subjects would be needed to achieve sufficient statistical power to show

differences of 35 wpm (0.07 log wpm) (Appendix 2a). This suggests that the number

of subjects recruited in this study (18) provided a sample of adequate size to

provided sufficient power in statistical analyses.


86

Table 2.2 Subjects' details

ID Gender Age Distance visual acuity (logMAR)

Near visual acuity (logMAR)

1 F 25 -0.02 0.06 2 M 21 -0.12 -0.10 3 M 23 -0.16 -0.06 4 F 24 -0.06 0.00 5 F 28 -0.18 -0.12 6 M 32 -0.10 -0.02 7 M 25 -0.20 -0.04 8 M 30 -0.14 -0.04 9 F 26 -0.10 -0.06

10 M 28 -0.10 -0.02 11 M 44 -0.16 -0.08 12 M 43 -0.16 -0.08 13 M 44 -0.08 0.04 14 M 55 -0.10 0.00 15 M 56 -0.16 -0.06 16 M 54 -0.06 0.10 17 F 48 -0.08 0.10 18 F 53 -0.10 -0.04

2.3 Methods

2.3.1 Reading charts

The Minnesota Low-Vision Reading chart (MNRead) uses short sentences of

different print sizes (Legge et al., 1989 (a); Mansfield et al., 1993; Mansfield et al.,

1994; Ahn et al., 1995). Print sizes range from 1.3 logMAR (N64) to –0.5 logMAR

(N1) in 0.1 logarithmic steps at the 40 cm test distance. Each sentence, which

contains 60 characters (10 standard wordsb) including a space between each word

and at the end of each line, is printed over three lines and is centrally justified. The

text is printed in high contrast (approximately 85%), black-on-white (normal

contrast) or white-on-black (reversed contrast) print. Only two versions of black-on-

white (normal contrast) print MNRead charts with different sentences were available

for use in this study.

b A standard word contains 6 characters (Carver, 1990).


87

The Bailey-Lovie text reading charts are modified MNRead charts (Legge et al.,

1989 (a); Mansfield et al., 1994). Each sentence used in this chart is taken from the

MNRead Corpusc but each sentence is printed over two lines at each print size and is

left justified (Figure 2.2). Print sizes range from 1.1 logMAR (N40) to -0.2 logMAR

(N2) in 0.1 logarithmic steps at 40 cm with one sentence at each print size. Again,

only two versions of the Bailey-Lovie text reading chart with different sentences

were available to be used in this study (Table 2.1). Times Roman is the font used in

this chart, which is commonly used for newspaper or other everyday printed

material. The spacing between words and rows is standardised. One letter (or

character) is as the spacing between words while the height of the words in the

smaller row is as the spacing between rows of different print sizes.

Figure 2.2 Bailey-Lovie text reading chart (in reduced size).

c http://vision.psych.umn.edu/www/people/stevem/mnr/sentences.html (May, 2000)

halla

This figure is not available online. Please consult the hardcopy thesis available from the QUT Library


88

Illumination and working distance were controlled by placing the reading charts on a

reading stand in a constant position. The 40 cm working distance was controlled by a

taut measuring string held by the subject during the experiment. The charts were

evenly illuminated by two overhead fluorescent tubes and the mean illuminance was

325 lux with a range of 300 to 340 lux from bottom to the top of the chart.

2.3.2 Procedures

2.3.2.1 Distance visual acuity

Distance visual acuity was measured binocularly using high contrast Bailey-Lovie

letter charts with the subject’s spectacle prescription which provided the best

correction (Bailey and Lovie, 1976). The Bailey-Lovie letter chart comprises the

same number of letters on each row with letters of approximately equal legibility.

The letters are selected from the series of 10 sanserif letters recommended in British

Standard 4274 (1968). The spacings between letters and between rows are uniform

and there is a logarithmic (geometric) progression of letter sizes. The design of this

chart facilitates the use of non-standard testing distances (Bailey and Lovie, 1976).

The Bailey-Lovie letter chart with a background luminance of 150 cd/m2 was used at

a test distance of 6 m in this study. Visual acuity was scored to the nearest letter and

recorded in logMAR (logarithm of the Minimum Angle of Resolution) with a

weighting of -0.02 log units per letter (Kitchin and Bailey, 1981). Subjects were

asked to start reading letters at a level well above threshold and to continue reading

down the chart as far as they could until a complete line of letters was incorrectly

read. This was to ensure that they had reached their resolution limit. Distance visual

acuity was measured once at the initial visit to select the subjects whose distance

visual acuity was 0.0 logMAR or better.

2.3.2.2 Near visual acuity

Monocular and binocular near word visual acuities were measured using the Bailey-

Lovie word chart at 40 cm at the initial visit. Presbyopic subjects wore their near

spectacle corrections for all measures of near visual acuity and reading performance.


89

The Bailey-Lovie word chart uses unrelated words arranged in a logarithmic

progression of sizes with standardised font, size progression, size range, spacings and

number of words for each print size (Bailey and Lovie, 1980). The print sizes range

from 1.4 logMAR (N80) to –0.2 logMAR (N2) at 40 cm in 0.1 logarithmic steps. To

keep the chart to a manageable size, there are two words for the three uppermost

rows (N80 to N48) and three words for the next three rows (N40 to N24). For print

sizes of N20 and smaller, each line contains six words.

As subjects participated in this study were all normally sighted, they were requested

to start reading the words from 0.8 logMAR (N20), which was a level well above

threshold, and to continue reading down the chart as far as they could. As threshold

print size approached, they were forced to guess until a complete line of words was

incorrectly identified. Near visual acuity was scored to the nearest word and recorded

in logMAR with 0.0167 log units for each word read correctly as there were 6 words

on each line for print sizes from 0.8 to –0.2 logMAR (N20 to N2).

2.3.2.3 Reading rate

Reading rates were measured for each subject using each of the MNRead and Bailey-

Lovie text reading charts in a random block sequence. Reading rates were measured

10 times (on separate occasions) within a period of two weeks to assess the reliability

of each chart. To ensure print size was large enough for MRR, subjects were asked to

start reading the sentence at the print size which was 0.8 log unit larger than the

threshold print size determined from the Bailey-Lovie word chart (see 2.3.2.2 above).

Subjects continued to read sentences at decreasing print sizes to the smallest print

size they could resolve. They were instructed to read each sentence aloud at their

normal reading rate and to read for understanding (Carver, 1990). At the end of each

sentence (one print size), the subjects were asked to look away to prevent them from

previewing the next line while reading time was recorded. The two available versions

of each type of chart were alternated on successive visits.

The time to read each print size (for each of the reading charts) was recorded with a

Micronta LCD Stopwatch to ± 0.5 second and any words or part of the words missed

or incorrectly read were recorded (reading errors). Errors, in number of words, were


90

taken into account in calculating the oral reading rate. Reading rate, in terms of

standard words per minute (wpm), was calculated for each print size by dividing the

number of standard wordsd read correctly by the reading time (in minutes).

(secs) timereading60 x errors) - wordsstandard ofnumber (=rate Reading

By measuring reading rates over a range of print sizes, CPS could be determined.

MRR was taken as the mean of the reading rates for print sizes at and above the CPS.

Text visual acuity was the smallest print size that subjects could correctly read 50%

or more of words.

2.4 Analysis

As the Bailey-Lovie text reading charts consist of the same number of words for each

print size (10 standard words) as the MNRead chart, the computer program –

MNRead Analysis 0.3 (Mansfield et al., 1993; Mansfield et al., 1994) was used to

determine the reading parameters for all the reading charts in this study. In addition

to using the computer analysis, reading rate for each sentence was plotted against

print sizes (in N-notation). The data points were joined and a smooth curve was fitted

by eye (graphical method). CPS was selected by direct observation of the print size at

which the reading rate started to decline (at N12 in Figure 2.1). MRR was then

calculated as the mean of the reading rates at and above the CPS. The MRR and CPS

determined by these two methods were compared for each of the reading charts.

Data were analysed using the Statistical Package for the Social Sciences (SPSS) -

version 10. Log transformation was used to ensure that MRR data were not

significantly different from normal distributions (Kolmogorov-Smirnov Goodness of

Fit test, p>0.4). Parametric statistics were used: repeated measures of analysis of

variance (ANOVA), Pearson correlations and paired t-tests. Bland and Altman

analysis (Bland and Altman, 1986) was used to examine the relationships between

reading measures (MRR and CPS) from different charts. Pearson and Spearman

d Number of words read was calculated as the total number of characters correctly read divided by 6.


91

correlations were used to analyse the relationship between the first trial of reading

measures (MRR and CPS) determined by the MNRead Analysis 0.3 program output

and the graphical method.

As the repeated measures of CPS (in terms of logMAR) across 10 trials among the

three different charts were significantly different from a normal distribution

(Kolmogorov-Smirnov Goodness of Fit test, p<0.05), the Friedman test was used to

analyse the repeatability of CPS for each chart. The statistical analysis supported by

the non-parametric repeated measure (Friedman test) was less comprehensive than

the analysis by repeated measures ANOVA. The interaction effect of the charts in the

repeated measures of CPS across trials could not be analysed by the nonparametric

analysis. Therefore, the analysis of CPS was conducted for each chart individually to

assess the repeatability of the CPS across trials. Bland and Altman analysis was used

to examine the levels of agreement between the reading parameters (MRR and CPS

at the first trial) determined by the two methods – MNRead Analysis 0.3 and

graphical method. A probability of less than 0.05 was taken to indicate statistical

significance for all analyses.

2.5 Results

2.5.1 Repeatability of reading measures

MRR and CPS were the main measures of reading performance for this study. Table

2.3 gives the means of each of these reading parameters measured from each chart

for each subject. Across the ten trials, MRR improved significantly with practice for

both charts (repeated measures ANOVA, Charts (2) x Times (10); F9,9=6.7,

p=0.005). The post hoc analysis showed that the significant increase in MRR was

found from the 4th trial (p<0.02). The MRR improved gradually from the 4th to the 8th

trial (p<0.02); however, it became stable at the last three measures (p=0.4), at which

the MRR was reached. Between the two charts, there was no significant difference in

the change in MRR across times (repeated measures ANOVA, Charts (2) x Times

(10); F9,9=2.19, p=0.13, Figure 2.3). This indicates that the improvement in the MRR

across trials was similar between the two charts. The reason for the gradual increase


92

in reading rate is mainly due to a learning effect since the same reading charts were

used repeatedly in the study.

Table 2.3 The mean of maximum reading rate (MRR) and critical print size (CPS) of 10 repeated measures for each chart

Mean maximum reading rate (log wpm) Mean critical print size (logMAR)

ID MNRead chart

Bailey-Lovie text reading chart MNRead chart Bailey-Lovie text

reading chart 1 2.30 2.33 0.05 0.14 2 2.35 2.33 -0.09 -0.02 3 2.36 2.36 -0.01 0.01 4 2.45 2.46 -0.08 0.00 5 2.42 2.43 -0.11 -0.06 6 2.38 2.40 -0.06 0.01 7 2.44 2.44 -0.03 0.04 8 2.58 2.59 -0.02 0.03 9 2.47 2.48 -0.08 -0.03

10 2.50 2.50 -0.02 0.05 11 2.43 2.43 -0.10 0.01 12 2.37 2.39 -0.08 0.00 13 2.36 2.38 0.09 0.15 14 2.40 2.41 -0.01 0.16 15 2.36 2.33 -0.03 0.00 16 2.41 2.44 0.26 0.36 17 2.38 2.39 0.09 0.16 18 2.31 2.30 0.00 0.03

-0.013 0.058 Mean 2.40 2.41 (N3+1 at 40 cm) (N4+3 at 40 cm)


93

2.2

2.3

2.4

2.5

1 2 3 4 5 6 7 8 9 10

Trial

Log

read

ing

rate

(wpm

)

MNReadchart

Bailey-Lovie textreadingchart

Figure 2.3 Mean and standard errors of maximum reading rate (MRR) across 10 trials for each chart for the 18 subjects. Error bars show one standard error of the meane.

MRR improved significantly across trials for both reading charts (p=0.005) with no significant difference in the change of MRR between charts (p=0.13).

To assess the repeatability of the reading measures from the two versions of each

chart, the MRR at the first trial was compared to the MRR at the second trial. There

were no significant differences in MRR measured with the two versions of each chart

(repeated measures ANOVA, Charts (2) x Trials (2); F1,17=2.60, p=0.13) or between

the two reading charts (F1,17=0.64, p=0.43). To assess the level of agreement between

the MRRs measured by each reading chart, Bland and Altman analyses were

conducted by plotting the differences in the reading rates at the first two trials as a

function of their means (Bland and Altman, 1986) (Figure 2.4). The mean differences

in MRR between trials were 0.0092 and 0.0088 log wpm and 95% confidence limits

in the mean difference were 0.035 and 0.034 log wpm for the MNRead and Bailey-

Lovie text reading charts respectively. The upper and lower limits of agreement (i.e.

95% confidence interval) were calculated as the mean difference plus and minus 1.96

times the standard deviation respectively. The limits of agreement for the MNRead

chart were 0.078 and –0.056 log wpm while that for the Bailey-Lovie text reading

chart were 0.075 and –0.058 log wpm. The 95% confidence interval of the mean

e For Figures 2.3 and 2.5, curves for each chart were translated horizontally for easier discrimination between curves.

p=0.005


94

difference were narrow compared with the limits of agreement indicating that the

difference in MRR measured across the two trials was small for both reading charts.

In addition, the correlation coefficients between the difference in MRR and the mean

of the MRR across two trials were not significant (r=-0.097, p=0.70 for MNRead

chart; r=0.12, p=0.64 for Bailey-Lovie text reading chart). These results indicate that

the MRRs measured by different versions of the MNRead and Bailey-Lovie text

reading charts are repeatable.

There were no significant differences in the CPS from both the Bailey-Lovie text

reading chart across the 10 repeated measures (Friedman, Fr9,17=11.33, p=0.25)

(Figure 2.5). However, the CPS from the MNRead chart at the last trial was

significantly smaller than the CPS at the first and third trials (Friedman, Fr9,17=42.72,

p<0.001). In comparison, CPS measured by the MNRead chart seemed to fluctuate

more than the CPS obtained from the Bailey-Lovie text reading chart (Figure 2.5).

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

2.2 2.3 2.4 2.5 2.6

Mean of maximum oral reading rate

Diff

eren

ce in

max

imum

ora

l rea

ding

rate

(lo

g w

pm)

MNReadchart

Bailey-Lovietext readingchart

Figure 2.4 Difference in maximum oral reading rate (MRR) between the first and second trials (2 versions for each chart) for the MNRead and Bailey-Lovie text reading charts for the 18 subjects. A positive difference indicates that reading rate at the second trial was faster than reading rate at the first trial.

Bland and Altman analysis showed that the MRR measured by the two trials were in good agreement because of the small 95% confidence interval of the mean difference and the small range of limits of agreement for both reading charts.


95

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

1 2 3 4 5 6 7 8 9 10

Trials

Crit

ical

prin

t siz

e (lo

gMA

R)

MNReadchart

Bailey-Lovie textreadingchart

Figure 2.5 Comparison of the mean and standard errors of critical print size (CPS) across 10 trials for each chart for the 18 subjects. Error bars show one standard error of the mean.

CPS measured by the Bailey-Lovie text reading chart was not significantly different (p=0.25) across the 10 repeated measures. However, the CPS from the MNRead chart was less repeatable as the CPS at the last trial was significantly smaller (p<0.001) than CPS at the initial trials.

2.5.2 Agreement of reading performance between charts

The measures of both MRR and CPS at the first and second trials were repeatable for

both reading charts. Therefore, reading measures (both MRR and CPS) from the first

trial of each was used to assess the level of agreement between the reading measures

from the different charts using correlation and Bland and Altman analyses (Bland

and Altman, 1986).

The reading parameters (MRR and CPS) measured by MNRead chart were highly

correlated with those measures obtained from Bailey-Lovie text reading chart

(p<0.001, Figure 2.6).

p = 0.25

p<0.001


96

2.2

2.25

2.3

2.35

2.4

2.45

2.5

2.2 2.25 2.3 2.35 2.4 2.45 2.5 2.55

Maximum reading rate by MNRead chart (wpm)

Max

imum

read

ing

rate

by

Bai

ley-

Lovi

e te

xt re

adin

g ch

art (

log

wpm

)

Figure 2.6a Correlation of maximum reading rates (MRR) measured by the two reading charts. The MRR obtained using the Bailey-Lovie text reading chart was highly correlated with that from the MNRead chart.

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

-0.2 -0.1 0 0.1 0.2 0.3 0.4

Critical print size by MNRead chart (logMAR)

Crit

ical

prin

t siz

e by

Bai

ley-

Lovi

e te

xt

read

ing

char

t (lo

gMA

R)

Figure 2.6b Correlation of critical print sizes (CPS) measured by the two reading charts. The CPS obtained using the Bailey-Lovie text reading chart was highly correlated with that from the MNRead chart.

rp =0.80, p<0.001

rs =0.71, p=0.001


97

Figure 2.7 examines MRR obtained from the Bailey-Lovie text reading and the

MNRead charts. The mean difference was less than 0.01 log wpm (5.85 wpm) with

95% confidence limits of the mean difference of 0.014 to –0.014 log wpm. The limits

of agreement were 0.055 and –0.054 log wpm. The 95% confidence intervals of the

mean difference were narrow compared with the limits of agreement indicating that

the differences in MRR measured by both charts were small. The limits of agreement

in MRR between the two charts covered a small range which would be acceptable for

clinical or research measures. In addition, the difference in MRR between charts was

not significantly correlated with the mean MRR (r=0.19, p=0.55). As both the mean

differences of MRR and the 95% confidence intervals of the mean difference using

these two reading charts are small, it is reasonable to suggest that the Bailey-Lovie

text reading or MNRead charts are interchangeable for the measurement of MRR.

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

2.1 2.2 2.3 2.4 2.5 2.6

Mean of maximum oral reading rate

Diff

eren

ce in

max

imum

ora

l rea

ding

rate

(lo

g w

pm)

Figure 2.7 Difference in maximum reading rate (MRR) between Bailey-Lovie text reading and MNRead charts measured at the first trial. A positive difference indicates that reading rate from the Bailey-Lovie text reading chart was faster than reading rate from the MNRead chart.

Bland and Altman analysis showed that the MRR measured by the reading charts were in good agreement because of the small 95% confidence interval of the mean difference (0.014 to -0.014 log wpm) and the small range of limits of agreement (0.055 to –0.054 log wpm).

Mean

Mean + 1.96SD

Mean - 1.96SD

SD = standard deviation

Upper limit of agreement

Lower limit of agreement


98

The Bland and Altman analyses in Figure 2.8 examine the differences in CPS

measured by Bailey-Lovie text reading and MNRead charts. The mean CPS from the

Bailey-Lovie text reading chart was 0.06 logMAR larger than that from MNRead

chart. This is a fairly large difference, equivalent to about one line on the charts. The

non-significant correlation between the difference and mean CPS obtained from the

two reading charts showed that the difference was independent of the mean (rs=-

0.021, p=0.94), In addition, the range of the limits of agreement was approximately 3

lines. This range in CPS between the two charts would not be acceptable for clinical

or research measures since magnification would be altered.

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

0.25

0.3

-0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5

Mean of critical print size (logMAR)

Diff

eren

ce in

crit

ical

prin

t siz

e (lo

gMA

R)

Figure 2.8 Differences in critical print size (CPS) measured using the Bailey-Lovie text reading and MNRead charts at the first trial for the 18 subjects. A positive difference indicates that CPS from the Bailey-Lovie text chart was greater than CPS from the MNRead chart. A number of data points overlapped, so that only 9 data points appear.

Bland and Altman analysis showed that the CPS determined by the Bailey-Lovie text reading and MNRead charts were in poor agreement because of a large mean difference (0.06 logMAR). Also, the wide 95% confidence interval (0.10 to 0.02 logMAR) and of limits of agreement (0.23 to -0.11 logMAR) indicated that the CPS measured by these two charts were not interchangeable.

Mean

Mean + 1.96SD

Mean - 1.96SD




99

2.5.3 Comparison of reading parameters calculated by two analysis methods

The reading parameters (MRR and CPS) were calculated by both the MNRead

Analysis 0.3 and the graphical method by plotting reading rates versus print sizes.

Results showed that both MRR and CPS at the first trial determined by the MNRead

Analysis were not significantly different from those calculated by the graphical

method for both reading charts (paired t-test t=0.99, df=35, p=0.33 for MRR and

Wilcoxon matched test, z=1.32, p=0.19 for CPS).

Both the MRR and CPS calculated by the MNRead Analysis were highly correlated

with those measures determined by the graphical method (p<0.01, Figure 2.9 and

Table 2.4). Bland and Altman analyses indicated that MRR and CPS determined by

MNRead Analysis 0.3 agreed well with the values determined by the graphical

method of direct observation (Figure 2.10).

2

2.1

2.2

2.3

2.4

2.5

2.6

2.2 2.3 2.4 2.5 2.6

Maximum reading rate by MNRead analysis (log wpm)

Max

imum

read

ing

rate

by

grap

hica

l m

etho

d (lo

g w

pm)

M NRead chart

Bailey-Lovietext chart

Figure 2.9a Correlation of maximum reading rates (MRR) determined by the two analysis methods at the first trial for the 18 subjects.

MRR determined by the MNRead Analysis 0.3 and graphical methods were highly correlated.

rp =0.88, p<0.001 for MNRead chart

rp =0.87, p<0.001 for Bailey-Lovie text reading


100

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

-0.2 0 0.2 0.4 0.6

Critical print size by MNRead analysis (logMAR)

Crit

ical

prin

t siz

e by

gra

phic

al m

etho

d(lo

gMA

R)

M NRead chart

Bailey-Lovietext chart

Figure 2.9b Correlation of critical print sizes (CPS) determined by the two different methods at the first trial for the 18 subjects. Fewer than 18 points are shown for both reading charts as some data points overlapped.

CPS determined by the MNRead Analysis 0.3 and graphical methods were highly correlated.

Table 2.4 Correlation of two methods from different charts

Charts MNRead chart Bailey-Lovie text reading chart

MRR CPS MRR CPS

Correlation * 0.88 0.69 0.87 0.92

Significance level <0.001 0.002 <0.001 <0.001 *Pearson correlation (rp) was conducted to assess the relationship of MRR determined by the two analytical methods. However non-parametric Spearman’s correlation (rs) was used, as the first trial of CPS obtained in both charts was significantly different from a normal distribution (p<0.05).

rs =0.69, p=0.002 for MNRead chart

rs =0.92, p<0.001 for Bailey-Lovie text reading


101

-0.1

-0.08

-0.06

-0.04

-0.02

0

0.02

0.04

0.06

0.08

2.2 2.3 2.4 2.5 2.6

Mean of maximum reading rate (log wpm)

Diff

eren

ce in

max

imum

read

ing

rate

(log

wpm

)

Figure 2.10a Comparison of difference in maximum reading rates (MRR) determined by the MNRead Analysis 0.3 and graphical method at the first trial for Bailey-Lovie text reading chart. A positive difference indicates that reading rate from the MNRead Analysis was greater than reading rate from graphical method.

MRR determined by the two methods were in good agreement with zero mean difference in MRR and small 95% confidence interval in the mean difference (0.019 to –0.017 log wpm). The limits of agreement (mean ± 1.96SD) were 0.07 to –0.069 log wpm.

Mean

Mean + 1.96SD

Mean - 1.96SD




102

-0.1

-0.05

0

0.05

0.1

0.15

-0.2 0 0.2 0.4 0.6

Mean critical print size (logMAR)

Diff

eren

ce in

crit

ical

prin

t siz

e (lo

gMA

R)

Figure 2.10b Comparison of difference in critical print size (CPS) determined by the MNRead Analysis 0.3 and graphical method at the first trial for Bailey-Lovie text reading chart. A positive difference indicates that CPS from the MNRead Analysis was greater than CPS from graphical method. Fewer than 18 points are shown for both reading charts as some data points overlapped.

CPS determined by the two methods were in good agreement, with a small mean difference of 0.017 logMAR and small 95% confidence interval of the mean difference (0.036 to –0.002 logMAR). The limits of agreement ranged from 0.092 to –0.06 logMAR.

2.6 Discussion

2.6.1 Validation of the reading charts

In order to validate whether the Bailey-Lovie text reading chart gave reliable reading

measures of reading performance, repeatability of each measure was assessed using

this chart and agreement with the reading measures from MNRead chart was

examined. MRR and CPS were the two important measures obtained from measuring

reading rates over a range of print sizes with each chart. MRR was repeatable for the

first two trials with each chart (Figure 2.4). As only two versions of each chart were

available, MRR increased significantly from the 4th trial onwards. This was probably

Mean

Mean + 1.96SD

Mean - 1.96SD




103

due to the familiarity with the charts used in this study. Layton and Koenig (1998)

found that reading rate, fluency and accuracy increased as the same reading materials

were used repeatedly. For subsequent studies involving repeated measures of reading

performance, more versions of the reading chart with different sentences should be

produced in order to minimise familiarity effects. In contrast, the CPS did not vary

significantly (except at the last trial) although the same charts were used repeatedly

to assess reading performance. This is mainly because CPS reflects the print size in

which fluent or maximum reading rate can be achieved. Therefore, familiarity with

the charts or memorisation of the sentences has little impact on the change in CPS.

In low vision rehabilitation, Legge and colleagues (1992; Ahn et al., 1995) suggested

that individual CPS for fluent reading should be determined by measuring reading

rates at a range of print sizes for every patient (Mansfield et al., 1993; Mansfield et

al., 1994; Ahn et al., 1995). Based on the CPS and the patient’s target print size (goal

reading material), magnification and therefore an appropriate low vision aid can be

determined. Thus a consistent finding for CPS is important in magnification

calculation. In this study, the CPS obtained from the Bailey-Lovie text reading and

MNRead charts was highly repeatable with no significant difference across two

repeated measures. This suggests that one measure of CPS is sufficient for clinical

purposes.

The mean differences and the limits of agreement in MRR measured by the Bailey-

Lovie text reading and MNRead charts were small, which was acceptable for clinical

measures. However, the mean difference in CPS between the two charts was nearly 1

line (0.1 log unit) and the range of limits of agreement in CPS was nearly three and a

half lines (0.35 log unit). The wide mean difference and large range of limits of

agreement might result in a different level of magnification being calculated from the

CPS determined from the two charts. For this reason, the Bailey-Lovie text reading

and MNRead charts are not interchangeable determining CPS for clinical purposes.

This implies that the same reading chart (with different versions) should be used for

repeated measures of reading performance across different sessions. The designs of

these two charts are very similar and the repeatability in MRR was not significantly

different from each chart (Figure 2.4). Reading parameters (MRR and CPS)


104

measured with both reading charts were highly repeatable, thus either one could be

used for clinical or research purposes.

Left justified sentences from the Bailey-Lovie text reading chart can better reflect

real world reading, as the page justification of common everyday printed materials

(such as newspaper or book) is mostly left aligned or justified for both the left and

right margins. In addition, the manageable size of the Bailey-Lovie text reading

chart, which can fit onto a reading stand for measuring reading performance at a

controlled distance, allows an easier set up for the experimental measures. For these

reasons, the Bailey-Lovie text reading chart was selected as the reading chart used in

the subsequent studies to measure subjects’ reading performance. Because of the

significant improvement in maximum reading rate after the 4th trial, ten different

versions of the Bailey-Lovie text reading chart with different sentences were

produced for subsequent studies (Chapters 3-5).

2.6.2 Relationship of the reading measures analysed by MNRead Analysis and

graphical methods

In determining MRR and CPS, researchers can use either a computer program -

MNRead Analysis 0.3 designed by Mansfield et al. (1994; 1996) - or the graphical

method. In this study, a high correlation and good level of agreement were found in

determining the MRR and CPS between these two methods. The graphical method is

preferred for clinical purposes because no computer and software is required, which

is more cost-effective and more convenient. As such, direct observation by graphical

method was selected to determine the CPS and calculate the MRR for the subsequent

studies.

While the results of this study are for people with normal vision, previous research

studies have shown that the reading performance assessed by MNRead chart were

reliable for testing people with normal (Mansfield et al., 1993) and low vision

(Mansfield et al., 1993; Bane et al., 1996). Therefore, there is no reason to believe

that the reliability of the Bailey-Lovie text reading chart would be any different for

people with visual impairment as the design of the two reading charts were quite

similar.


105

2.7 Conclusion

Measuring reading performance is important for providing information about real-

world reading for people with normal and low vision. As reading rate is sensitive to

visual factors, it is commonly included in clinical assessment (Ahn et al., 1995).

Therefore, a chart that can produce reliable measures of reading performance is most

useful for clinical and research purposes. This study compared reading performance

measured using Bailey-Lovie text reading and MNRead charts and found that there

were no significant differences in the repeated measures of the reading parameters

using each of these charts. The Bailey-Lovie text reading chart (which has not

previously been described in the literature) was as reliable and repeatable as the

MNRead chart. Clinically, the result from this study indicates that one measure of

MRR and CPS is sufficient to obtain an accurate measure of people’s reading

performance. The Bailey-Lovie text reading chart was adopted as the reading chart

for the subsequent experiments (Chapter 3 to 5) measuring participants’ reading

performance on sentences because of its repeatability and better reflection of

people’s reading performance in the real world. To minimise memorisation of the

sentences, ten different versions of the Bailey-Lovie text reading chart were

produced with the same chart design but different sentences.

With regards to the two methods of analysing the reading measures, the results

obtained from the MNRead Analysis 0.3 were not significantly different from those

determined by direct observation of the graphical method. The graphical method,

which was easier to use during experimentation, was selected for calculating the

MRR and determining the CPS in the subsequent studies.

Chapter 3 Validation of the method of calculating magnification for reading with low vision

106

CHAPTER 3

Validation of the method of calculating magnification for

reading with low vision

3.1 Introduction ................................................................................................ 107 3.1.1 Magnification ......................................................................................... 107 3.1.2 Previous methods to calculate required magnification .......................... 108 3.1.3 Importance of acuity reserve .................................................................. 110 3.1.4 Current methods to calculate required magnification ............................ 110

3.2 Subjects........................................................................................................ 112 3.2.1 Inclusion criteria..................................................................................... 114 3.2.2 Exclusion criteria.................................................................................... 114 3.2.3 Sample size............................................................................................. 115

3.3 Methods ....................................................................................................... 115 3.3.1 Preliminary assessment .......................................................................... 115 3.3.2 Distance visual acuity............................................................................. 115 3.3.3 Near visual acuities ................................................................................ 116 3.3.4 Reading assessment................................................................................ 116 3.3.5 Procedures .............................................................................................. 118

3.3.5.1 Reading rate without low vision aids .............................................. 118 3.3.5.2 Magnification calculation................................................................ 118 3.3.5.3 First visit.......................................................................................... 121 3.3.5.4 Second visit ..................................................................................... 121

3.4 Analysis........................................................................................................ 124

3.5 Results and Discussion ............................................................................... 125

3.5.1 Methods of determining magnification .................................................. 125 3.5.2 Repeatability of measurements .............................................................. 131 3.5.3 Comparison of reading performance with and without low vision aids 133

3.6 Conclusion................................................................................................... 136


107

3.1 Introduction

Reading is one of the most highly valued activities of daily living within today’s

society. Any ocular disorder which deprives people of the ability to read causes

severe restriction of daily activities. Providing magnification to enlarge the retinal

image is a way to overcome a person’s reading disability. Magnification can be

provided by various low vision aids (LVAs) such as hand-held, stand, and spectacle-

mounted magnifiers, telescopes or electronic reading aids (Dickinson, 1998;

Brilliant, 1999). In order to prescribe an appropriate low vision aid (LVA) to assist

people with low vision to retain their reading ability, accurate and efficient

calculation of the required magnification is necessary as part of low vision

rehabilitation. The purpose of this study was to determine the most appropriate (and

efficient) method for calculating the required magnification for reading in order to

standardise experimental procedures for the prescription of magnifiers in subsequent

studies in this thesis.

3.1.1 Magnification

The magnification provided by an optical magnifier in situ can be thought of as

either the increase in retinal image size when the magnifier is used (i.e. ratio of

retinal image size with magnifier to retinal image size without magnifier) or the

increase in the visual angle subtended at the user’s eye when the magnifier is used.

As retinal image size is directly related to visual angle subtended, magnification

expressed in these two ways is the same.

Bailey has proposed that the equivalent power of the optical system be used to

represent the magnifying effect of an optical LVA (Bailey, 1980 (b); Bailey, 1984

(a); Bailey et al., 1994). The reciprocal of equivalent power is the equivalent focal

length of the lens system which Bailey termed equivalent viewing distance (EVD)

(Bailey, 1980 (b); Bailey, 1984 (a); Bailey et al., 1994). EVD is the distance at which

the original object would subtend the same angular size as that subtended by the

image formed by the LVA. Therefore, by knowing the resolution limit of the patient

(near visual acuity), a new resolution capability (i.e. near visual acuity achieved with

magnifier) can be determined if the EVD of the magnifying system is known. For


108

example, if the threshold print size of a patient is 12-point print (N12) at 40 cm,

his/her resolution limit (new threshold print size) can be improved to N4 (one third

of the original threshold print size) if a plus lens with EVD of 13.33 cm (one third of

the original working distance) is used.

3.1.2 Previous methods to calculate required magnification

In the past, low vision rehabilitation practitioners have measured patients’ distance

and/or near visual acuity to calculate the required magnification for LVAs by the use

of simplified equations. These equations were Kestenbaum’s rule and the Lighthouse

Method (Dickinson, 1998). Kestenbaum's rule assumes that distance visual acuity

and near reading can be equated, the reference addition is +2.50 D and the desired

acuity level is 6/15 Snellen equivalent (0.4 logMAR) at near (Kestenbaum and

Sturman, 1956). Kestenbaum’s formula for calculating magnification is the

reciprocal of distance visual acuity divided by four. The Lighthouse Method however

considers near visual acuity rather than distance visual acuity. The desired acuity is

assumed to be 1M (6/15 Snellen equivalent at 40 cm) or 8 point print (N8) with a

reference addition of +2.50 D. The equation for calculating magnification by the

Lighthouse Method is two and a half times near visual acuity (M unit) divided by

four (Table 3.1).

Cole (1993) found that the magnification levels determined by these methods were

usually under-estimated when compared with the final prescription. These equations

consider only the patients' visual acuity and ignore their desired reading material,

assuming it is always 1M (N8) print size. By taking the patients' target reading

materials into account, for example, newspapers, magazines or large print books,

appropriate magnification can be determined to meet the patients' reading

requirements. To improve the accuracy of the calculation of magnification, Cole

(1993) introduced another equation: Reciprocal of vision method. This method

predicts the magnification based on both the patient's best-corrected distance visual

acuity and the required near visual acuity which can be predicted from the patient's

target reading materials. It assumes that the patient’s distance and near visual acuities

are equivalent and that the reference distance is 40 cm. Cole’s Reciprocal of vision


109

for calculating magnification is 2.5 times distance visual acuity, divided by 4 times

the required near visual acuity (Table 3.1).

Table 3.1 Previous formulae for calculating magnification Example: Distance visual acuity= 6/60

Near visual acuity = 6/60 (with +2.50 D add at 40 cm) or 4M

Required near visual acuity = 6/15 (or 1M)

Rule Equation for calculated magnification

Kestenbaum’s Rule 4

acuity visualdistance of Reciprocal

Lighthouse Method 4

2.5 x unit) (Macuity lNear visua

Reciprocal of vision 4

5.2×notation)(Snellen acuity lnear visua Required

acuity visualDistance

All these equations assume that unit magnification occurs for a working distance of 25 cm. Each formula gives a magnification of 2.5x, which expressed in dioptres is equivalent to the magnification provided by a +10 D lens (based on F/4), or an equivalent viewing distance of 10 cm.

However, magnification was still found to be under-estimated by this method when

compared with the final magnification prescribed (Cole, 1993; Flom et al., 1993;

Elam, 1997). This underestimation may be due to a number of factors which affect

reading with magnification. The introduction of magnifiers can result in a restricted

field of view, reduced illumination, aberrations and poor technical skills in

manipulating magnifiers, all of which can cause a reduction in reading rate with

magnification (Legge et al., 1985 (b); Lovie-Kitchin and Woo, 1987; Lowe and

Drasdo, 1990; Beckmann and Legge, 1996; Ortiz et al., 1999). In addition, the

calculation of required magnification needs to include some acuity reserve to help

offset these difficulties in using magnifiers and provide useful reading rates

(Whittaker and Lovie-Kitchin, 1993).


110

3.1.3 Importance of acuity reserve

A number of studies have demonstrated that the reading rates of subjects with normal

and low vision increase as print size increases from threshold (Legge et al., 1985 (a);

Legge et al., 1985 (b); Lovie-Kitchin and Woo, 1987). People with either normal or

low vision cannot read fluently if the print size of the reading material is at or close

to threshold size (Legge et al., 1985 (b); Legge et al., 1992; Whittaker and Lovie-

Kitchin, 1993; Bullimore and Bailey, 1995). This implies that the print size of a

reading task must be larger than threshold to achieve fluent reading rates. For this

reason, Whittaker and Lovie-Kitchin (1993) introduced the term “acuity reserve”.

Acuity reserve is defined as the ratio of print size that the patient intends to read to

their threshold print size. Watson (Watson et al., 1997) suggested that one of the

reasons that closed-circuit television or video magnifiers gave better reading

performance than optical magnifiers was the greater acuity reserve provided.

In the past, guidelines for prescribing magnification for reading with low vision

suggested that minimum magnification should be prescribed in order to maximise the

field of view available (Mehr and Freid, 1975; Lovie-Kitchin and Bowman, 1985).

However, Whittaker and Lovie-Kitchin (1993) showed that inadequate acuity reserve

was the greater impediment to reading with low vision. They have shown that field

of view is not a limiting factor affecting reading rate if sufficient acuity reserve is

given (Lovie-Kitchin and Whittaker, 1999 (a)).

3.1.4 Current methods to calculate required magnification

Two different approaches are suggested in the literature for the provision of acuity

reserve in determining required magnification to achieve fluent reading rate (Legge

et al., 1992; Whittaker and Lovie-Kitchin, 1993; Ahn et al., 1995; Lovie-Kitchin and

Whittaker, 2000). Whittaker and Lovie-Kitchin (1993) suggested that print size that

was 2 times (0.3 log unit) larger than threshold was sufficient to achieve fluent

reading rate for subjects with both normal and low vision. They recommended using

this fixed acuity reserve in the calculation of magnification for reading (Lovie-

Kitchin and Whittaker, 1999 (b)). For example, if a person's near visual acuity is N40

at 40 cm and his/her goal is to read the newspaper (N8 or 1M print) fluently, a fixed

acuity reserve of 0.3 log unit means that the required threshold size will be N4


111

(0.5M) so that the person can read N8 (1M) fluently. Therefore a magnifier with

equivalent viewing distance of 4 cm is required.

The magnification determined by fixed acuity reserve (two times as small as the

target print size), as suggested by Whittaker and Lovie-Kitchin (1993), was aimed at

achieving fluent reading rate rather than maximum reading rate. For people with low

vision, measuring the maximum reading rate is not as crucial as measuring the fluent

reading rate, as the latter has greater functional relevance in everyday life (Legge et

al., 1985 (b); Bullimore and Bailey, 1995). In addition, the reading rate for people

with AMD is substantially slower than the reading rate achieved by people with

normal vision or people with other causes of low vision (Legge et al., 1985 (b);

Bullimore and Bailey, 1995). For this reason, the difference in reading rate for

maximum and fluent reading may not be significant.

The fixed acuity reserve suggested by Whittaker and Lovie-Kitchin (1993) was

generalised from data across groups of subjects with low vision. Some individuals

might need more or less acuity reserve for fluent reading. Legge and colleagues

(1992; 1995) suggested that the acuity reserve (and therefore magnification) required

for fluent reading should be determined on an individual basis by using the MNRead

acuity chart (Mansfield et al., 1993; Mansfield et al., 1994). From the measurement

of reading rates at different print sizes, an individual’s required acuity reserve was

calculated as the ratio between critical print size (CPS) - the smallest print size that

gave maximum reading rate – and the person’s goal print size. For example, if a

person's CPS is N64 at 40 cm, he/she needs a magnifier with an equivalent viewing

distance of 5 cm to fluently read N8 print size material.

The fixed and individual acuity reserve methods each have their advantages and

disadvantages. The fixed method can simplify the clinical procedures of calculating

the required magnification, as only near visual acuity and the target print size of the

person’s reading materials are required. However, this method may over- or under-

estimate the required acuity reserve for different individuals. The individual

assessment overcomes this disadvantage of the fixed acuity reserve method but it

requires measurement of the patient’s reading rates at a number of different print

sizes, which makes the clinical assessment more complicated. Lovie-Kitchin and


112

Whittaker (2000) discussed the two methods of using acuity reserve to prescribe

magnification for reading and suggested a combination of the two methods to used.

However, there have been no studies comparing the two prescribing methods to

support this suggestion. The aim of this study was to validate the ‘fixed’ acuity

reserve method for selecting magnification by comparing it with the ‘individual’

acuity reserve method. If the magnification and reading performances with

magnifiers selected by these two methods were similar, the fixed acuity reserve

method would be chosen as the first step in the selection of magnifiers for the

subsequent studies in this thesis because it is a simpler method.


1. There would be no difference in the reading rates achieved with magnifiers

selected by the fixed acuity reserve method and by the individual acuity

reserve method compared with the reading rate with subjects’ own

magnifiers.

2. There would be no difference in the equivalent viewing distance (EVD)

calculated by the fixed acuity reserve method and the individual acuity

reserve method compared with the EVD of subjects’ own magnifiers.

3. Reading performance measured with and without magnifiers would not

change across two visits, provided vision was stable.

3.2 Subjects

Nine subjects with low vision due to AMD between 68 and 89 years (mean age 79.4

± 5.6 years) were selected from the Queensland University of Technology (QUT)

Vision Rehabilitation Centre (VRC). Distance visual acuity in the better eye ranged

from 0.38 logMAR (6/15+1) to 1.26 logMAR (6/120+2) (Table 3.2). Subjects with

significant levels of hearing impairment were excluded. All subjects were fluent

English speakers.


113

Table 3.2 Subjects’ details

ID Age (years)

Years of AMD

Years of visual

impairment *

Low vision aids +

Years of LVA use

Target print size (N-point)

Reading frequency

‡

Distance visual acuity

(logMAR)

Near visual acuity (word)

with +4 D (logMAR)

1 79 3 1 3 1 8 4 1.10 1.20 2 89 3 1 2 2 8 1 0.38 0.46 3 83 0.8 0.6 2 0.33 8 1 1.26 1.10 4 68 2 2 1 2 8 1 0.74 0.88 5 80 4 1 2 1 8 1 1.04 0.98 6 80 3 3 3 3 8 1 1.08 1.08 7 76 2 2 3 1 16 4 1.26 1.30 8 81 8 6 3 4 8 2 1.02 1.10 9 79 1 1 3 0.33 8 1 0.60 0.84

* Years of visual impairment as reported by the subjects

+ Code for low vision aids (LVA) used at near: 1. High add spectacles (+4 D addition or greater); 2. Hand-held magnifiers; 3. Stand magnifiers.

‡ Reading frequency: 1. Read daily; 2. 2-3 times per week; 3. Once per week; 4. Rarely.


114

3.2.1 Inclusion criteria

All subjects were diagnosed by ophthalmologists to have age-related macular

degeneration (AMD). Seven subjects had non-exudative (dry) AMD and two subjects

had the exudative (wet) form of the disease. The recruited subjects had received a

comprehensive vision examination or follow-up assessment of their optical LVAs

(high additions, hand-held magnifiers or stand magnifiers) during the 12 months

preceding their recruitment (Table 3.2). They were selected to have less than 1 line

(0.1 log unit) change in near visual acuity since their last assessment at the VRC to

confirm that the magnification provided by the LVAs was still appropriate for

reading their target print size.

3.2.2 Exclusion criteria

The major exclusion criterion for subjects’ selection related to their ocular health. As

the overall study focussed on people with low vision due to AMD (i.e. central field

loss), subjects with other significant ocular pathology (e.g. cataract of more than

NC2 sclerosis; mild or advanced glaucoma) were excluded from the study.

The other exclusion criterion related to patients’ reading ability. Subjects who

reported or whose clinic records indicated any cognitive problem or poor general

health were not included in this experiment. Previous studies have stated that reading

rate could be reduced due to difficulties with comprehension among people with

normal (Shankweiler and Liberman, 1972) and low vision (Baldasare et al., 1986).

Therefore, it was important to ensure that subjects would have no difficulties

understanding the reading materials. A simplified Neale Analysis of reading ability

was used to confirm that all recruited subjects had Grade 6 or above reading

comprehension (Neale et al., 1999). A short Grade 6 story was read to each subject.

Any subject who could not correctly answer 3 out of 4 questions about the content of

the passage was excluded from the study.

In addition, subjects were requested to participate in the study for two visits. Only

those subjects with stable vision over the period of study (up to two months) were


115

recruited. Originally, nineteen subjects were recruited, but this requirement resulted

in ten subjects being excluded, leading to the small sample size (n=9).

3.2.3 Sample size

The goal of this study was to compare magnification requirements determined by

different methods and the magnifications required are referenced to each individual’s

acuity. While there is a wide diversity of visual acuities and reading rates in low

vision population (Sunness et al., 1996; Lovie-Kitchin et al., 2000 (a); Bowers et al.,

2001 (b); Martin et al., 2002), a large sample size would be needed to give

statistically significant results, which was beyond the scope of a PhD study. A small

pilot study examining the methods of calculating magnification was still considered

to be relevant.

3.3 Methods

3.3.1 Preliminary assessment

A full optometric assessment including objective and subjective refraction and

ophthalmoscopy was conducted for each subject before the experiment to ensure that

the best possible vision had been achieved. As there was no significant difference in

the subjective refractions and subjects’ current spectacle prescriptions, all the

assessments were conducted with subjects’ current spectacles.

3.3.2 Distance visual acuity

Distance visual acuity was measured using high contrast (93%) Bailey-Lovie

(logMAR) letter charts monocularly with the subject’s current spectacle prescription

(Bailey and Lovie, 1976). The design of the Bailey-Lovie letter chart has been

discussed in section 2.3.2.1.

Three versions of the Bailey-Lovie letter chart with a background luminance of 150

cd/m2 were used. The chart was randomly selected for each subject with the test


116

distance selected from the standard logarithmic progression such that the subject

could read at least the top two rows on the chart (Bailey and Lovie, 1976). Visual

acuity was scored to the nearest letter and recorded in logMAR (logarithm of the

minimum angle of resolution) by deducting 0.02 log unit per letter read accurately

(Kitchin and Bailey, 1981). Subjects were asked to start reading letters at a level well

above threshold and to continue reading down the chart as far as they could until a

complete line of letters was incorrectly read. This was to ensure that they had

reached their threshold visual acuity.

3.3.3 Near visual acuities

Near visual acuity was measured using the Bailey-Lovie word chart (Bailey and

Lovie, 1980). Details of this chart have been described in section 2.3.1. Six versions

of this chart were used in this experiment and were randomly presented to prevent

chart familiarity. The charts were presented on a reading stand at a working distance

determined by the subject’s near spectacle addition. In addition, near acuity was

measured at a working distance of 25 cm with a +4.0 D addition. Illumination and

working distance were controlled by placing the reading charts on a reading stand at

a constant position. The working distances were controlled by a measuring string

held by the subject during the experiment and monitored by the experimenter. The

charts were evenly illuminated by overhead fluorescent tubes and the illumination

ranged from 320 to 380 lux (range of illumination from the top to the bottom of the

chart). The lighting and experimental set up used for the assessment of near visual

acuity were identical to those used for the assessment of reading performance.

3.3.4 Reading assessment

To measure reading rates for sentences and passages, two different types of reading

materials were selected - Bailey-Lovie text reading charts and short passages of text.

From the results of Chapter 2, Bailey-Lovie text reading charts were selected for the

assessment of reading rate on sentences. The design of the Bailey-Lovie text reading

chart has been described in section 2.3.1. As only two versions of this chart were


117

availablea, ten more versions were produced using sentences selected from the

MNRead Corpusb.

In addition to sentences, nine short reading passages, derived from other reading tests

(Sloan Reading Cards for low vision patients; University of Waterloo near vision test

card; Maclure bar reading book for children - Clement Clarke International Ltd.),

were used to assess reading performance. Each passage contained 263 ± 5.56

characters (43.85 ± 0.93 standard words). Passages were produced in different print

sizes from N20 to N8 and were printed across A4 cards with landscape orientation

with print contrast of 85% black on white (Figure 3.1). The text was left justified

with single-spacing between lines. The reading level of each passage was analysed

by Flesch-Kincaid Grading Level System (Microsoft word, 2000). Based on the

average number of syllables per word and words per sentence, all passages had sixth

grade reading level or below. Thus the readability of passages was well below the

reading ability of subjects, ensuring that reading performance was not limited by text

difficulty (Carver, 1990).

He moved forward a few steps. The house was so dark behind him; the world was so dim

and uncertain in front of him, that for a moment his heart failed him. He might have to search

the whole garden for the dog. Then he heard a sniff, felt something wet against his leg.

Print size: N8 (Number of characters: 269)

Figure 3.1 Example of one of the nine short reading passages.

a The first two versions of Bailey-Lovie text reading charts were supplied by Professor Ian Bailey from University of Berkeley, California (2000). b http://vision.psych.umn.edu/www/people/stevem/mnr/sentences.html (updated May, 2000)


118

3.3.5 Procedures

3.3.5.1 Reading rate without low vision aids

The eye with better near visual acuity was selected as the eye for monocular reading

assessment. Subjects were requested to hold the string at a fixed working distance

which depended on their own near additions and read the text aloud (Figure 3.2).

Reading rates without LVAs were measured by using the Bailey-Lovie text reading

charts for a range of print sizes with the subject’s habitual near addition. Procedures

for reading rate measures and calculation of reading rate for each print size have

been described in section 2.3.2.3. In brief, subjects were required to read aloud

sentences of decreasing print sizes until they reached their threshold visual acuity.

Reading rate for each sentence on the Bailey-Lovie text reading chart was plotted

against print size (refer to Figure 2.1). A smooth curve was fitted to the data points

and CPS was selected by direct observation. Maximum oral reading rate was taken as

the mean of the reading rates for print sizes at and above CPS. Reading performance

without LVAs was determined three times with different versions of Bailey-Lovie

text reading chart. The mean of the three measures of maximum reading rate was

used for analyses while mean CPS was estimated to the nearest 0.1 log unit.

3.3.5.2 Magnification calculation

Magnification in terms of equivalent viewing distance (Bailey et al., 1994; Lovie-

Kitchin and Whittaker, 1999 (a)) was calculated from the reading measures without

LVA using the two methods described in section 3.1.4 – the first, using a fixed 2:1

(0.3 log unit) acuity reserve for all subjects (Whittaker and Lovie-Kitchin, 1993;

Lovie-Kitchin and Whittaker, 2000) and the second, using the acuity reserve

determined individually for each subject from their CPS (Legge et al., 1992; Ahn et

al., 1995).


119

Figure 3.2 Measurement of reading rate without low vision aids using Bailey-Lovie text reading chart.

(i) Fixed acuity reserve

Using the fixed acuity reserve method (Lovie-Kitchin and Whittaker, 1999 (b);

Lovie-Kitchin and Whittaker, 2000), required EVD was calculated as follows:

VA)near s(subject' sizeprint resholdCurrent thdistance ewingCurrent vi

2sizeprint Target

(cm) EVD Required ×=

For example, if a person could read N20 at 25 cm with a +4 D addition but would

like to read 8-point print (N8) fluently, magnification allowing him/her to read N4

was needed. This patient needed an EVD of 5 cm.

2025

28 (cm) EVD Required ×=

Measuring string for working distance Bailey-Lovie text reading chart


120

(ii) Individual acuity reserve

Using the individual acuity reserve method (Legge et al., 1992; Ahn et al., 1995),

required EVD was calculated as follows:

distance ewingCurrent vi CPS

sizeprint Target (cm) EVD Required ×=

For a person whose CPS was N48 at 25 cm to read 8-point print (N8) fluently, an

EVD of 4.17 cm was required.

25 488 (cm) EVD Required ×=

Based on the EVD calculated by the 2 different methods, optical LVAs of the same

type as the subject’s own magnifier (i.e. high addition, hand-held or stand

magnifiers) were chosen. In selecting an appropriate hand-held magnifier of required

EVD, subjects were requested to wear their distance prescriptions and to place the

magnifier at the appropriate lens-to-page distance, which was the focal length of the

magnifying lens, to ensure the calculated EVD was achieved. For subjects who used

stand magnifiers, the eye-lens distance, image distance and subjects’ near spectacle

addition were taken into consideration in selecting the stand magnifier with the

appropriate EVD (Bailey et al., 1994; Lovie-Kitchin and Whittaker, 1999 (b)). Based

on the calculated EVD and the near addition (i.e. clear image distance obtained from

reading addition), an appropriate stand magnifier with a suggested eye-to-lens

distance was selected. For example, if a subject with a near addition of +3 D required

an EVD of 6.67 cm, a 20 D STM was prescribed as it gave an EVD of 6.66 cm with

an eye to image distance of 35.1 cm. This magnifier allowed the subject to read

clearly with sufficient magnification provided that the eye-to-lens distance was

approximately 10 cm (refer to Table 4.5) and the subject was instructed accordingly.

However, if the image distance provided by the near addition was different from the

calculated image distance, for the sake of achieving required EVD, a tolerance of

defocus of 5 cm was considered acceptable.


121

3.3.5.3 First visit

Each subject’s target print size was selected according to his/her preferred reading

material. Reading rates with the two LVAs determined by the two methods (fixed

and individual acuity reserve) were measured by asking the subjects to read passages

of their target print size three times (Table 3.3). For example, if the subject indicated

that they wanted to read newspaper print, the target print size was N8 (1M), but if

they wanted to read magazines, the target print size was N12 (1.6M). In addition,

reading rate with the subjects’ own optical aid was measured. The three magnifiers

were used in random block of sequence and the mean of the three reading rates were

taken for each magnifier. The mean reading rate with each LVAs was compared with

the mean maximum reading rate without LVAs.

When measuring reading performance with the magnifiers selected by the fixed and

individual acuity reserve methods, the subjects were instructed on the appropriate

eye-lens distance to give the required EVD. However, they were allowed to adopt

their own eye-lens distance when reading performance was measured with their own

magnifiers; this eye-lens distance was measured in order to calculate the EVD of the

subject’s own magnifier (Bailey et al., 1994).

3.3.5.4 Second visit

In order to assess the repeatability of reading rate measures over a period of time,

subjects were requested to return for further reading assessments approximately two

months after their first visit. These subjects reported that their vision had not changed

during this time and this was confirmed from repeated vision measurements. There

were no statistically significant differences in visual acuity for distance (paired t-

tests; t=1.69, df=8, p=0.13) or near visual acuities (paired t-test; t=0.54, df=8,

p=0.61) between the two visits. At the second visit reading rates for sentences

(Bailey-Lovie text reading charts) and passages were measured with and without

LVAs (one trial each) as described in Table 3.3. All reading measures were the mean

of three repeated measurements at the first visit. However, given the good within-

session repeatability of reading rates reported in Chapter 2, only one trial for each

reading assessment was taken at the second visit to avoid subject fatigue.


122

Table 3.3 Reading measures for first and second visits

Visit Reading Measures Instrument used Working distance Results

Without LVAs Reading on sentences Bailey-Lovie text

reading chart Working distance of near addition Mean maximum oral reading rate (log wpm) Mean CPS (N-notation)

First visit

With LVAs

Reading on passages with LVAs • Fixed method • Individual method • Own LVAs

Passages of texts with LVAs used in random order across subjects

• Fixed working distance for the LVAs selected by the two methods.

• Habitual working distance with own LVAs and EVD was calculated.

Mean reading rate with LVAs for passages at target print size

Reading on sentences Bailey-Lovie text reading chart

Maximum oral reading rate (log wpm) CPS (N-notation) Without

LVAs Reading on passages Passages of text at CPS

Working distance of near addition Oral reading rate (wpm) for passage

Reading on sentences with LVAs • Fixed method • Individual method • Own LVAs

Bailey-Lovie text reading chart

Maximum oral reading rate with LVAs (log wpm) CPS (N-notation)

Second visit

With LVAsReading on passages with LVAs

• Fixed method • Individual method • Own LVAs

Passages of text at CPSPassages of text at

target print size

• Fixed working distance for the LVAs selected by the two methods.

• Habitual working distance with own LVAs and EVD was calculated.

Oral reading rate with LVAs (wpm) for passage at CPS Oral reading rate with LVAs for passage at target print size

LVA – Low vision aid CPS – Critical print size EVD – equivalent viewing distance


123

Maximum reading rate was again calculated as the mean of the reading rates for print

sizes at and above CPS, as described in section 3.3.5.1. However, reading rate with

LVAs in some cases was reduced for large print sizes, because few characters were

visible in the field of view (Figure 3.3). Therefore, when reading rates on large print

sizes were less than 90% of the reading rate at CPS, they were excluded from the

calculation. For example in Figure 3.3, maximum reading rate was the mean of

reading rates at print sizes from N20 to N8.

Reading rates with and without LVA for Subject 3

0

20

40

60

80

100

120

140

64 48 40 32 24 20 16 12 10 8 6 5

Print size (N-point)

Rea

ding

Rat

e (w

pm)

Large printw ithout lowvision aid

Reading ratew ith lowvision aid

Figure 3.3 Example of reduced reading rate with low vision aids when reading large print sentences. Reading rate with LVA was reduced for print sizes larger than N20 (unfilled arrow).

For calculation of maximum reading rate, these reading rates which were below 90% of the reading rate at CPS (N8 - filled arrow) were excluded.


124

3.4 Analysis


version 10. As vision measures were not significantly different from normal

distributions (Kolmogorov-Smirnov Goodness of Fit test, p>0.1), parametric

statistics were used. Paired t-tests were used to compare subjects' distance and near

visual acuities to ensure that there were no significant changes in vision between

experimental visits.

Reading rates measured in all conditions for both visits were not significantly

different from normal distributions (Kolmogorov-Smirnov Goodness of Fit test,

p>0.1), however the standard deviations of the reading rates between subjects were

large. Reading rates were log-transformed as commonly used in previous studies

(Yager et al., 1998; Aquilante et al., 2000). Log reading rates were not significantly

different from a normal distribution (Kolmogorov-Smirnov Goodness of Fit test,

p>0.16), therefore parametric statistics were used – analysis of variance (ANOVA),

repeated measures ANOVA, paired t-tests and Pearson correlation. To compare the

reading rates with the three LVAs and the maximum reading rate without LVAs,

one-way ANOVA and paired t-tests were performed. Repeated measures ANOVA

was used to test the repeatability of reading rates measured at visits 1 and 2. Pearson

correlation was used to analyse the relationship between reading rates with and

without LVAs. A probability of less than 0.05 was taken to indicate a statistical

significance for all analyses.

The probability of finding a significant difference by chance alone (Type I error)

increases rapidly with the number of statistical tests. As the number of tests

measured at the second visit equalled the degrees of freedom, a Bonferroni correction

was made to the probability associated with each test by dividing it by the number of

tests executed (Green et al., 2000; Pallant, 2001). The adjusted probability was 0.006

as there were eight repeated analyses of reading measures at the second visit.


125

3.5 Results and Discussion

3.5.1 Methods of determining magnification

There were no significant differences in the required EVD calculated by the two

different methods or the EVD of the subjects’ own magnifier (one-way ANOVA,

F2,24=0.91, p=0.42). The magnification of the LVAs prescribed for QUT VRC

patients are usually determined initially on the basis of a fixed acuity reserve of 0.3

log unit. Reading performance is then assessed and the magnification may be

modified. For this reason, it was not surprising to find no significant differences in

EVD between that of the subjects’ own LVA and that calculated by the two different

methods.

The difference in the calculating magnifications by the fixed and individual acuity

reserve methods is the consideration of acuity reserve (in terms of log unit) in the

equation. Table 3.4 gives acuity reserves calculated as the ratio of critical print size

to near visual acuity for each subject, expressed as log differences. In this study,

subjects required acuity reserves of 0.1 to 0.4 log units (1.3:1 to 2.5:1) to achieve

fluent reading rate without magnifiers on large print (sentences). The majority of

subjects (seven of the nine) required an acuity reserve of at least 1.5:1 (0.2 log units),

which was similar to the minimum acuity reserve suggested by Whittaker and Lovie-

Kitchin (1993). As the individual acuity reserves were similar to the fixed acuity

reserve suggested by Whittaker and Lovie-Kitchin, it was not surprising that the

magnification determined by the two methods was not significantly different.

Although the difference in EVD of the three magnifiers was not statistically

significant, reading performance with each of the magnifiers may have varied with

differences in magnification for some individual subjects. Therefore, reading

performance achieved with each magnifier was also measured. There were no

significant differences between the log reading rates on passages with LVAs

determined by the two methods (the fixed or individual acuity reserve method) or

with the subjects’ own-magnifiers (Figure 3.4 and Table 3.5). This was true both at

the initial visit (one way ANOVA, F2,24=0.14, p=0.87) and at the repeat visit

(F2,24=0.01, p=0.99) (Figure 3.4).


126

Table 3.4 Acuity reserves to achieve maximum reading rates for individual subjects

ID Near visual acuity

Critical print size

Acuity reserve

Maximum reading rate

(logMAR) (logMAR) (log unit) (log wpm) 1 1.1 1.3 0.2 1.6261 2 0.4 0.7 0.3 2.2008 3 1.1 1.2 0.1 2.1472 4 0.8 1.0 0.2 2.1493 5 0.9 1.2 0.3 1.9974 6 0.9 1.3 0.4 1.9272 7 1.3 1.4 0.1 1.5066 8 1.0 1.5 0.4 1.7286 9 0.7 1.1 0.4 2.2463

1

1.2

1.4

1.6

1.8

2

1st visit 2nd visit

Log

read

ing

rate

(wpm

)

RR w ith lowvision aid byfixed acuityreserve

RR w ith lowvision aid byindividual acuityreserve

RR w ithsubjects' ow nlow vision aid

Figure 3.4 Log reading rates (RR) with low vision aids on passages at target print size for both visits.

Log reading rates on passages (at target print size) with LVAs determined by the two methods (the fixed or individual acuity reserve method) or with the subjects’ own-magnifier were not significantly different. Error bars show one standard error of the mean.

p = 0.87 p = 0.99


127

Table 3.5 Comparison of the reading performance of subjects at two visits

First visit Second visit

Reading measures

Reading rate (wpm)

Log reading rate (log wpm)

Reading rate (wpm)

Log reading rate (log wpm)

Without LVA

Maximum reading rate 106.55 ± 55.45 1.96 ± 0.27 110.31 ± 53.19 1.98 ± 0.26

Fixed acuity reserve * * 98.33 ± 47.50 1.93 ± 0.26

Individual acuity reserve * * 103.19 ± 48.84 1.95 ± 0.27

Sentence With

LVA Subjects’ own

LVAs * * 102.07 ± 50.81 1.94 ± 0.28

Without LVA

Maximum reading rate * * 67.37 ± 34.25 1.76 ± 0.27

Fixed acuity reserve * * 60.40 ± 27.78 1.73 ± 0.24

Individual acuity reserve * * 61.53 ± 27.72 1.74 ± 0.24

Passage at critical print size With

LVA Subjects’ own

LVAs * * 63.03 ± 26.82 1.75 ± 0.23

Fixed acuity reserve 58.42 ± 30.20 1.67 ± 0.29 58.84 ± 31.81 1.70 ± 0.29

Individual acuity reserve 50.13 ± 31.30 1.63 ± 0.32 57.14 ± 30.90 1.68 ± 0.29 Passage at target

print size With LVA

Subjects’ own LVAs 53.2 ± 24.67 1.71 ± 0.26 55.99 ± 28.10 1.68 ± 0.29

* The measurements were made at the second visit only.


128

Individual results of the EVD determined by the two methods are not apparently

different from the EVD achieved with subjects’ own magnifiers although there is a

wide variation among subjects (Figure 3.5). For example, the calculated EVD (by

both methods) for subject 4 was less than the EVD achieved with his own magnifier.

However, the reading rates achieved with magnifiers from each method was not

substantially different.

With the exception of subject 8, individual results illustrate that the reading rates

with the magnifiers and the EVD determined by the two methods were not

apparently different from the reading rates with subjects’ own magnifiers (Figure

3.6). Reading rate with subject 8’s own stand magnifier was substantially faster than

the reading rates with magnifiers determined by fixed or individual acuity reserve

method. This was because this subject was well adapted to his own magnifier which

provided an acuity reserve of 0.6 log unit, compared with an acuity reserve of 0.4 log

unit (determined by individual acuity reserve method) and with 0.3 log acuity. In this

case, both acuity reserve methods under-estimated the acuity reserve giving best

reading rate. In comparing the vision measures (such as distance and near visual

acuity) and reading rates without LVA for subject 8 with those measures for other

subjects, there was no obvious explanation for the stronger acuity reserve required by

this subject.

Visual field assessment and contrast sensitivity, which were not included in this

study, may have affected the acuity reserve required for fluent reading rate.

Clinically, if reading rate with the new magnifier is not satisfactory to read the target

reading materials (small print) fluently, other vision measures, such as visual field,

contrast sensitivity or modification of stronger magnification should be further

investigated to determine the most suitable magnifiers required for fluent reading

rate.


129

0

5

10

15

20

25

LVA by fixed Acuityreserve

LVA by individiualacuity reserve

Subjects' own LVA

Equi

vale

nt v

iew

ing

dist

ance

(cm

)

1

2

3

4

5

6

7

8

9

Figure 3.5 Equivalent viewing distance (cm) determined by each method for each subject.

1

1.2

1.4

1.6

1.8

2

2.2

LVA by fixed Acuityreserve

LVA by individiualacuity reserve

Subjects' own LVA

Log

read

ing

rate

with

LVA

(log

wpm

)

1

2

3

4

5

6

7

8

9

Figure 3.6 Log reading rates (RR) with low vision aids on passages at target print size for each subject at the first visit.

By the use of Bland and Altman analysis (Bland and Altman, 1986), the mean

difference of the log reading rate was 0.04 log wpm with a 95% confidence limits of


130

the mean differencec of 0.106 to –0.02 log wpm (Figure 3.7). The upper and lower

limits of agreement (i.e. 95% confidence interval) were 0.20 and –0.12 log wpm, a

range of about 0.3 log unit. Although the mean difference of the log reading rate was

small, the range of the limits of agreement was wide suggesting that the

measurements of the log reading rate using magnifications determined by the two

different acuity reserve methods are not interchangeable for clinical purpose.

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5 1 1.5 2 2.5

Mean of the reading rates (log wpm)

Diff

eren

ce in

read

ing

rate

s (lo

g w

pm)

Figure 3.7 Difference in log reading rates calculated by the fixed and individual acuity reserve methods versus mean of the log reading rates. A positive difference indicates that reading rate achieved with magnifiers determined from the fixed acuity reserve method was faster than that from the individual acuity reserve method.

Although the log reading rate determined by the fixed acuity reserve was slightly faster than that by the individual acuity reserve method, the mean difference of the log reading rate between the two methods was small. However, the limits of agreement and the 95% confidence limits of mean difference between the two methods were wide.

c The 95% confidence intervals were the confidence intervals of the differences in reading rates determined by fixed and individual acuity reserve methods.

Mean + 1.96SD

Mean – 1.96SD

Mean = 0.04

SD= standard deviation




131

3.5.2 Repeatability of measurements

Log reading rates were quite repeatable when measured in one session. At the first

assessment, no significant difference was found between the log maximum oral

reading rate across the three trials (repeated measures ANOVA, Groups (1) x time

(3), F2,36=1.13, p=0.35). Ahn et al. (1995) similarly found log reading rates without

LVAs to be very repeatable across six measurements. This suggests that one

measurement of reading rate on large print would usually be sufficient to estimate the

maximum reading rate without LVA for subjects with low vision.

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

1.2 1.4 1.6 1.8 2 2.2 2.4


Diff

eren

ce in

read

ing

rate

s (lo

g w

pm)

Figure 3.8 Difference in log reading rates without LVAs at the first and second visits as a function of mean reading rates without LVAs. A positive difference indicates that reading rate at the second visit was faster than that at the first visit.

Bland and Altman analysis showed that the reading rates without LVAs at the first and second visits were in good agreement because of the small 95% confidence interval of the mean difference (0.065 to 0.008 log wpm) and the small range of limits of agreement (0.109 to –0.036 log wpm).

Log reading rates between visits were also repeatable (Table 3.5). The mean log

maximum reading rate (1.98 ± 0.26 log wpm) without LVAs on large print at the

second visit (one trial only) was not significantly different from the mean of the log

maximum reading rate at the first visit (1.96 ± 0.27 log wpm) (paired t-test, t=1.77,

Mean + 1.96SD

Mean – 1.96SD

Mean = 0.037


132

df=8, p=0.11). Bland and Altman analysis (Figure 3.8) shows that the mean

differences in log maximum reading rate between the first and second visits was

0.037 log wpm. The 95% confidence interval of the mean difference were between

0.0082 and 0.065 log wpm. The limits of agreement for the test and retest measures

were 0.109 and –0.036 log wpm. There was no significant correlation between the

difference in log reading rate at the first and second visits (r=-0.023, p=0.95).

Because of the narrow range of 95% confidence limits compared with the limits of

agreement, log reading rate was repeatable across the two visits.

Log reading rates with LVAs for passages at the subjects’ target print sizes also were

repeatable between the two visits for the magnifiers determined by the fixed or

individual acuity reserve methods or using subjects' own LVAs. Bland and Altman

analysis showed that the 95% confidence interval of the mean difference and the

limits of agreement of reading rates with LVAs between the first and second visits

were small (Figures 3.9a and b).

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

1.1 1.3 1.5 1.7 1.9 2.1 2.3


Diff

eren

ce in

read

ing

rate

(log

wpm

)

Figure 3.9a Difference in log reading rates with LVAs determined by fixed acuity reserve method at the first and second visits as a function of mean reading rates with LVAs. A positive difference indicates that reading rate at the second visit was faster than that at the first visit.

Bland and Altman analysis showed that the reading rates with the LVAs were repeatable between visits because of the small 95% confidence interval of the mean difference (-0.027 to 0.067 log wpm) and the small range of limits of agreement (-0.10 to 0.14 log wpm).

Mean + 1.96SD

Mean – 1.96SD

Mean = 0.02


133

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

1.1 1.3 1.5 1.7 1.9 2.1 2.3


Diff

eren

ce in

read

ing

rate

(log

wpm

)

Figure 3.9b Difference in log reading rates with LVAs determined by the individual acuity reserve method at the first and second visits as a function of mean reading rates with LVAs. A positive difference indicates that reading rate at the second visit was faster than that at the first visit.

Bland and Altman analysis showed that the reading rates with the three LVAs were repeatable between visits because of the small 95% confidence interval of the mean difference (-0.0039 to 0.07 log wpm) and the small range of limits of agreement (-0.049 to 0.124 log wpm).

3.5.3 Comparison of reading performance with and without low vision aids

At the first visit, log maximum reading rate without LVA was significantly faster

than the log reading rate with LVA at subjects’ target print size (repeated measures

ANOVA, F3,6=40.67, p<0.01). This did not agree with findings of Lovie-Kitchin et

al. (2000 (a)) and Ahn and Legge (1995) comparing reading rates with and without

LVAs. The discrepancy between the results obtained from this study compared with

that from previous studies was probably because single sentences were used to

measure log maximum reading rate without LVA, but longer passages were used to

measure log reading rate with LVA. Carver (1990) has suggested that reading rate

for longer passages is slower than reading rate for short sentences. To confirm that

the significant reduction in log reading rate with magnifiers compared with the log

reading rate without was due to the different reading tasks, reading rate measures

Mean + 1.96SD

Mean – 1.96SD

Mean = 0.038


134

were made for similar tasks, both sentences and passages, with and without LVA at

the second visit.

At the second visit, there were no significant differences between log maximum

reading rate with each of the three LVAs or without LVA for sentences (repeated

measures ANOVA, F3,6=2.03, p=0.21) and passages at CPS (repeated measures

ANOVA, F3,6=1.45, p=0.32) (Table 3.5). This result agrees with the findings of

Lovie-Kitchin et al. (2000 (a)) who also reported no significant differences in

reading rates with and without LVA for paragraph reading at CPS. Figure 3.10 is an

example of reading rates with and without LVA across a range of print sizes on

sentence reading. In addition these results confirm that log reading rate without

magnifier is a good predictor of log reading rate with magnifier (r=0.96; p<0.0001)

as previously reported by Lovie-Kitchin et al. (2000 (a)) and Ahn and Legge (1995).

These results and those of previous studies (Ahn and Legge, 1995; Bowers et al.,

2001 (b)) indicate that, provided that there is ample acuity reserve, magnifiers do not

reduce reading rate for experienced users. From the measurement of reading rate

without LVA on large print, an estimate of the potential reading rate with LVA can

be made before optical aids are prescribed (Ahn and Legge, 1995).

A number of previous studies have found that reading rate was reduced by the use of

magnifiers (Mancil and Nowakowski, 1986; McMahon and Spigelman, 1989; Cohen

and Waiss, 1991 (a); Dickinson and Fotinakis, 2000; Bowers, 2000 (a)). These

investigators recruited subjects who had normal vision, were highly educated and

who had higher maximum reading rates without LVAs than those achieved by people

with low vision. These subjects were not experienced in using LVAs and in some

studies (Mancil and Nowakowski, 1986; McMahon and Spigelman, 1989; Cohen and

Waiss, 1991 (a)) the image sizes with and without LVAs were different for the two

conditions. Passages of the same print size were used to measure the reading rates

with and without LVAs, giving different image sizes for the two reading conditions.


135

Log reading rates with and without LVA for subject 2

0.5

0.7

0.9

1.1

1.3

1.5

1.7

1.9

2.1

2.3

2.5

64 48 40 32 24 20 16 12 10 8 6 5 4 3 2

Print size (N-notation)

Log

read

ing

rate

(wpm

)RR withoutLVA

RR with LVAby fixed acuityreservemethod

RR with LVAby individualacuity reservemethod

RR withsubjects' ownLVA

Figure 3.10 Example of log reading rates (RR) as a function of print size with and without low vision aids (LVA) for sentences using the Bailey-Lovie text reading chart.

Log maximum reading rates with LVAs were not significantly different from log reading rate without LVA.

The results of this study showed that there was no significant difference between log

reading rate with and without LVAs for experienced subjects when performing

similar reading tasks. However these findings were only for experienced magnifier

users. For inexperienced users, based on previous studies (Mancil and Nowakowski,

1986; McMahon and Spigelman, 1989; Cohen and Waiss, 1991 (a); Dickinson and

Fotinakis, 2000; Bowers, 2000 (a)) it is reasonable to expect a reduction in reading

rate when a LVA is first prescribed. There are at least two possible reasons for this -

people lack the experience in using the LVAs or they have given up reading for a

long time because of their poor vision or both. Perhaps giving these patients with

visual impairment large print reading practice before prescribing the LVAs would

enhance their reading performance with magnifiers. The results of this study suggest

that reading rates with LVAs can reach reading rates on large print, once the patients

adapt to their new LVAs. Previous studies have mainly concentrated on providing

extensive training programs for the development of eccentric viewing (Goodrich and

Mehr, 1986; Nilsson, 1990; Graessley and Kirby, 1996; Nilsson et al., 1998) and the


136

use of optical LVAs (Watson et al., 1992; Goodrich et al., 1998). However, there has

been no study to evaluate subjects’ adaptation to LVAs by monitoring their reading

performance over time. Therefore, in the next study (Chapter 4), the effects of

reading practice and/or training on reading rate with magnifiers were investigated.

3.6 Conclusion

Determining the appropriate magnification of LVAs to assist people for reading is an

important task in low vision rehabilitation. For clinical purposes, a method which is

easy, efficient and accurate, would be most useful. For the AMD subjects in this

study using the fixed acuity reserve method to determine magnification gave similar

results to the individual acuity reserve method. The fixed acuity reserve method, as

described by Lovie-Kitchin and Whittaker (Lovie-Kitchin and Whittaker, 1999 (b);

Lovie-Kitchin and Whittaker, 2000) requires only the identification of the patient’s

target reading material and the assessment of near visual acuity; an acuity reserve of

2:1 (0.3 log-unit) is then used to determine magnification. This study suggested that

for most people with AMD, this method gave magnification which met their reading

goals and was not significantly different from that which had been prescribed.

Therefore, this method of magnification calculation (fixed acuity reserve method)

was applied to the subsequent study reported in Chapter 4, in which the majority of

recruited subjects did not have a magnifier to assist with daily reading. However,

occasionally the reading rate achieved with the magnification calculated in this way

might not be satisfactory if the reading rate with the new LVA is markedly reduced

compared to the reading rate without a LVA. In such cases, individual assessment of

reading rates for different print sizes would be needed to determine the acuity reserve

required for fluent reading rate and calculate appropriate magnification.

Chapter 4 Effect of practice on reading rate with stand magnifiers

137

CHAPTER 4

Effect of practice on reading rate with stand magnifiers

4.1 Introduction.................................................................................................. 139 4.1.1 Previous training program ....................................................................... 140 4.1.2 Aim of the study ...................................................................................... 142

4.2 Subjects ......................................................................................................... 143 4.2.1 Inclusion criteria ...................................................................................... 143 4.2.2 Exclusion Criteria .................................................................................... 145 4.2.3 Sample size .............................................................................................. 146 4.2.4 Subject groups ......................................................................................... 146 4.2.5 Multi-centre assessment .......................................................................... 147 4.2.6 Home assessment..................................................................................... 147

4.3 Methods......................................................................................................... 148 4.3.1 Experimental interventions for each group ............................................. 148 4.3.2 Preliminary assessment ........................................................................... 150

4.3.2.1 Visual acuities .................................................................................. 151 4.3.2.2 Visual field ....................................................................................... 151 4.3.2.3 Contrast sensitivity........................................................................... 152

4.3.3 Reading assessment ................................................................................. 154 4.3.4 Practice stand for reduced field of view .................................................. 158 4.3.5 Stand magnifiers ...................................................................................... 160 4.3.6 Procedures ............................................................................................... 162

4.3.6.1 Reading assessment without stand magnifier .................................. 162 4.3.6.2 Prescription of stand magnifiers ...................................................... 162 4.3.6.3 Reading assessment with stand magnifiers ...................................... 165 4.3.6.4 Reading assessment with practice stand .......................................... 165 4.3.6.5 Questionnaires.................................................................................. 166

4.3.7 Large print books for reading practice .................................................... 166 4.3.8 Change of vision over time...................................................................... 168

4.4 Analysis ......................................................................................................... 168 4.4.1 Reading assessments ............................................................................... 169

4.4.1.1 Transformation of reading rate to log reading rate .......................... 169 4.4.1.2 Repeated measures of log reading rate ............................................ 169

4.4.2 Vision measures....................................................................................... 170 4.4.3 Questionnaire........................................................................................... 171 4.4.4 Correlation between variables ................................................................. 171

4.5 Results ........................................................................................................... 172 4.5.1 Effect of reading practice on log reading rate with STM over time........ 172

4.5.1.1 Comparison of baseline measures among groups ............................ 172 4.5.1.2 Comparison of log reading rate with stand magnifier as a function of

time for experimental groups ..................................................... 174


138

4.5.1.2.1 Change in vision measures for experimental groups ................. 179 4.5.1.2.2 Change in log reading rate with stand magnifier controlled for

change in visual acuity ............................................................... 181 4.5.1.3 Comparison of log reading rate with STM and vision measures as a

function of time for clinical group ............................................. 182 4.5.1.4 Comparison of log reading rate with stand magnifier and vision

measures as a function of time for all groups............................. 184 4.5.1.4.1 Change of vision measures over time for all subjects................ 184 4.5.1.4.2 Change of log reading rate with stand magnifier over time for all

subjects ....................................................................................... 184 4.5.2 Change in log reading rate without stand magnifier across time............. 185 4.5.3 Comparing log reading rate with and without stand magnifier ............... 187 4.5.4 Relationship between near visual acuity and EVD (magnification)

prescribed................................................................................................. 190 4.5.5 Log reading rate with a restricted field of view (P2)............................... 191 4.5.6 Log reading rate with and without a restricted field of view compared with

log reading rate with STM....................................................................... 191 4.5.7 Reading performance with difference reading materials......................... 192 4.5.8 Results of questionnaire and subjective report ........................................ 197

4.5.8.1 Frequency and duration of reading................................................... 197 4.5.8.2 Reading materials............................................................................. 200 4.5.8.3 Subjective response on the effectiveness of STM............................ 202 4.5.8.4 Correlation between reading variables and the reported use of STM

.................................................................................................... 204 4.5.9 Factors affecting log reading rate with STM........................................... 205

4.5.9.1 Correlation between vision and reading variables ........................... 205 4.5.9.2 Factors predicting log reading rate with and without STM ............. 208 4.5.9.3 Factors predicting the change in log reading rate with STM ........... 208

4.6 Discussion...................................................................................................... 210 4.6.1 Effect of reading practice on log reading rate with STM ........................ 210 4.6.2 Comparison of log reading rate without STM with log magnifier reading

rate 215 4.6.2.1 Maximum reading rate without STM............................................... 215 4.6.2.2 Reading rate with STM .................................................................... 216

4.6.3 Comparison of log reading rates under different fields of views ............ 218 4.6.4 Reading performance with different reading materials ........................... 220 4.6.5 Use of stand magnifiers ........................................................................... 220 4.6.6 Satisfaction with stand magnifiers........................................................... 222 4.6.7 Predictors of reading rate with and without magnifiers........................... 223

4.7 Conclusions and recommendations ............................................................ 226


139

4.1 Introduction

Although magnifiers allow people who are visually impaired to read smaller print,

the introduction of an optical low vision aid reduces the field of view available for

reading. Due to the physical aperture of the magnifier, the number of magnified

characters (horizontal field of view) that can be seen through the lens is restricted. In

addition, difficulty manipulating magnifiers during reading has been suggested as

one of the major problems that the magnifier users encounter when magnifiers are

prescribed. Because of these two reasons, it is not surprising that the reading rate

with magnifiers has sometimes be found to be significantly reduced compared with

reading rate on large print (Mancil and Nowakowski, 1986; McMahon and

Spigelman, 1989; Neve, 1989 (b); Cohen and Waiss, 1991 (a); Cohen and Waiss,

1991 (b); Fotinakis and Dickinson, 1994; Bowers, 2000 (a)) (refer to Table 1.9).

Other research studies, however, have reported that reading with a low vision aid

does not significantly reduce reading rate compared to that achieved without a low

vision aid (Ahn and Legge, 1995; Lovie-Kitchin et al., 2000 (a)). The results

reported in Chapter 3 confirmed this. The conflicting findings (with respect to

whether or not magnifiers reduce reading rates) is mainly due to the differences

between studies in terms of characteristics of the participants, the amount of

experience in reading with magnifiers and levels of retinal magnification provided

with and without low vision aids.

Those studies that showed significant reductions in reading rate as a consequence of

magnifier use assessed subjects with normal vision rather than low vision (Mancil

and Nowakowski, 1986; McMahon and Spigelman, 1989; Neve, 1989 (a); Neve,

1989 (b); Cohen and Waiss, 1991 (a); Cohen and Waiss, 1991 (b); Bowers, 2000

(a)). These people were inexperienced in reading under a restricted field of view and

in manipulating the magnifiers for reading. Therefore it is reasonable that their

reading rates would decrease when a magnifier was introduced since their habitual

reading was interrupted. Additionally, they may have had faster reading rates which

were more likely to be reduced with a magnifier. However, those studies which

showed no significant difference in reading rate with and without magnifiers

assessed subjects with low vision who were experienced in using low vision aids for


140

reading (Ahn and Legge, 1995; Lovie-Kitchin et al., 2000 (a)). Thus, reading rate

with magnifiers was not significantly different from reading rate without a magnifier,

provided the magnification levels were equal for the two conditions. This suggests

that training people in reading under a restricted field of view and/or training people

in the skills of manipulating the magnifiers are essential for people with no

experience in using the magnifiers for reading. It is suggested that once people

overcome the difficulties in reading with magnifiers and become experienced

magnifier users, they can achieve their reading rate with magnifiers equivalent to the

reading rate without magnifiers on large print.

4.1.1 Previous training program

In low vision rehabilitation, numerous studies have investigated training programs to

improve the reading ability of people with low vision (Goodrich et al., 1977; Jose,

1983; Nilsson and Nilsson, 1986; Nilsson, 1988; Nilsson, 1990; Langmann et al.,

1994; Nilsson and Nilsson, 1994; Stoll et al., 1995; Freeman and Jose, 1997; Nilsson

et al., 1998; Goodrich et al., 2000; Bowers, 2000 (b); Goodrich and Kirby, 2001).

Previous training programs have been discussed in section 1.5.1. Generally, many of

these programs focus on training eccentric viewing in people who have central visual

field loss (Goodrich and Mehr, 1986; Nilsson, 1990; Graessley and Kirby, 1996;

Nilsson et al., 1998). Other programs emphasise that the training should include not

only eccentric viewing but also the techniques of manipulating magnifiers (Nilsson,

1986; Nilsson and Nilsson, 1986; Nilsson, 1990; Nilsson and Nilsson, 1994).

However, most of these training programs are time-consuming and labour intensive

(Chapter 1, Table 1.10).

Although there have been a number of previous studies which assess the

effectiveness of training programs, many did not include a control group (Goodrich

et al., 1977; Goodrich and Quillman, 1977; Nilsson, 1986; Nilsson and Nilsson,

1986) or did not measure baseline reading performance before training (Goodrich et

al., 1977; Goodrich and Mehr, 1986; Goodrich et al., 1998; Goodrich et al., 2000;

Goodrich and Kirby, 2001). Therefore, no comparison of reading performance could

be made between subjects who received training and those who did not.


141

Training that can provide an easy, less time-consuming way to improve reading

performance for people with low vision would and in effective vision rehabilitation.

Bowers (2000 (b)) compared the reading rate with stand magnifiers (STMs) before

and after practice in using the STM for a period of two days. The results showed that

the magnifier reading rate improved after short-term practice using the STM.

However, subjects recruited in this study were young people with normal vision. The

conclusion derived from this study might not be applicable to people with low vision.

In the experiment described in this thesis, the effect of practice on reading

performance with magnifiers for subjects with visual impairment was investigated. In

addition, in Bower’s (2000 (b)) study, there ware no control subjects who received

no practice using STMs for comparison. As such, it is arguable whether the

improvement in reading rate after short-term practice using the STMs was solely due

to practice or to the repeated measures of reading performance or placebo effect.

Apart from comprehensive training on eccentric viewing and manipulation of

magnifiers, Watson et al. (1992) suggested that training or practice of reading was

important to improve reading performance in low vision people. Her study recruited

subjects with AMD in three groups: control, practice and training groups. The

control group received no reading training or follow up visits. The practice group

received reading practice for 10 minutes daily and four biweekly follow up visits.

The training group subjects received five biweekly in-office reading training sessions

for 1 hour each. Results indicated that daily reading practice could provide similar

improvements in reading comprehension to that achieved by the training group

(Watson et al., 1992), suggesting that practice in reading is an effective rehabilitation

technique. However, this study only focused on reading comprehension of large print

without magnifiers. Even though reading rate was not measured in Watson’s study,

other studies have indicated that daily reading increases reading rate since attention

is shifted from decoding to extracting meaning from the text, thus improving

comprehension (Samuels, 1979; Layton and Koenig, 1998). However, these studies

did not investigate the effect of reading practice on the reading performance with low

vision aids.


142

4.1.2 Aim of the study

To the author’s knowledge, there have been no published studies that have

investigated whether reading practice, prior to the prescription of low vision aids, can

improve reading performance with low vision aids. Therefore, the aim of this study

was to investigate the impact of reading practice on the reading performance with

illuminated STMs for subjects with AMD who were newly prescribed these low

vision aids. Inexperience in reading under a restricted field of view and difficulty

manipulating magnifiers have been suggested as the major limitations that new

magnifier users encounter. In addition to large print reading practice, the impact of

reading practice with a “practice stand”, which simulated the reduced field of view of

a STM with no magnifying lens on reading performance with illuminated STM was

also investigated. Reading rate was expected to decrease (compared with reading rate

on large print) when a STM was first introduced. Therefore, the time required for

reading rate with the magnifier to improve such that it was not significantly different

from reading rate on large print without a magnifier was also assessed.


1. Maximum reading rate without STMs on large print (for sentences or

passages) would improve with large print reading practice.

2. Reading rate (for sentences or passages) would reduce when STMs were

introduced.

3. Reading rate with STMs (for sentences or passages) would improve over time

with practice.

4. After reading practice on large print, reading rate with STMs (for sentences

or passages) would improve.

5. After reading practice on large print and under a restricted field of view (with

the simulated practice stand), reading rate with STMs (for sentences or

passages) would improve over time.


143

4.2 Subjects

Forty-six subjects aged between 71 and 95 years (mean age 81 ± 4.73 years) with

low vision due to AMD were selected from the clinic database of the QUT Vision

Rehabilitation Centre (VRC), Vision Australia Low Vision Clinic (LVC) and by

referral from local ophthalmologists. Distance visual acuity in the better eye ranged

from 0.22 logMAR (6/9-2) to 1.3 logMAR (6/120) (Table 4.1).

Of the forty-six subjects, three (6.5%) withdrew during the experimental period

(details of these subjects are included in Table 4.1). Reasons for their dropout were

general health problem (subject 8) and sudden deterioration of vision (subject 15)

that no longer fulfilled the recruitment criteria. In addition, one subject (subject 38)

was excluded because he changed from the illuminated STM to a hand-held

magnifier and magnifying lamp as his primary reading aids at the first follow up

visit.

4.2.1 Inclusion criteria

Subjects were recruited if they had been diagnosed by ophthalmologists as having

AMD. The majority of subjects (78%) had non-exudative (dry) AMD with only 10

subjects (22%) having the exudative (wet) form of the disease, representing a similar

proportion of patients with wet and dry AMD to that reported in a large population

study by Chisholm (1996). Subjects were required to have not used a STM

previously and to have limited experience in using hand-held magnifiers for daily

reading, for example only spot reading for shopping or reading telephone books. All

recruited subjects were fluent English speakers and had received a preliminary vision

examination to confirm that there were no other major causes of low vision.


144

Table 4.1 Subjects’ details

ID Group + Age

Visual impairment (months) ‡

Type of AMD

Distance VA

(logMAR)

Near word VA

(logMAR)

Text VA(logMAR)

CPS (logMAR)

Word print size with STM (N-point)

Text print size with

STM (N-point)

1 P2 82 17 0 0.30 0.40 0.4 0.9 5 4 2 P2 86 7 0 0.78 1.20 1.2 1.4 8 8 3 P2 76 18 0 0.24 0.40 0.2 0.4 3 3 4 P2 81 48 0 0.70 0.90 0.8 1.1 6 5 5 P2 73 24 0 0.54 0.70 0.8 1.1 6 5 6 P2 71 12 0 0.24 0.48 0.4 0.6 6 5 7 P2 78 16 1 0.62 0.83 0.7 1.0 5 4 8 * P2 75 16 0 0.8 1.33 1.1 1.3 6 6 9 P2 84 16 1 0.32 0.50 0.4 0.6 4 4 10 P2 78 12 0 0.82 1.17 0.9 1.3 5 4 11 P2 84 12 0 0.92 1.20 1.1 1.4 8 5 12 P2 83 18 0 0.72 0.90 0.8 1.1 5 4 13 N 79 36 0 0.58 0.70 0.6 0.9 4 3 14 N 82 6 0 0.78 1.07 1.0 1.3 8 6 15 * N 83 36 0 1.18 1.30 1.2 1.5 10 6 16 N 83 12 0 0.46 0.50 0.4 0.6 4 3 17 N 82 12 0 0.38 0.50 0.4 0.8 6 4 18 N 80 12 0 0.64 0.70 0.7 1.1 4 3 19 N 71 36 0 0.86 1.10 1.0 1.4 8 6 20 N 82 12 1 0.64 0.68 0.6 0.9 4 3 21 N 79 1 1 0.56 0.52 0.5 0.8 4 3 22 N 86 18 1 0.32 0.42 0.4 0.7 5 4 23 N 84 4 1 0.52 0.82 0.6 0.9 3 2.5 24 P1 86 12 0 0.22 0.50 0.4 0.6 4 3 25 P1 78 12 0 0.58 0.70 0.7 0.9 4 2.5 26 P1 84 6 0 1.08 1.10 0.9 1.2 8 6 27 P1 82 12 0 0.60 0.90 0.8 1.3 6 5 28 P1 75 72 0 0.84 1.12 1.0 1.4 5 4 29 P1 79 12 0 0.50 0.70 0.4 0.9 5 4 30 P1 72 7 1 0.90 1.00 0.9 1.1 8 6 31 P1 81 11 0 0.72 0.72 0.6 0.9 6 5 32 P1 85 24 0 0.46 0.50 0.5 0.7 3 2.5 33 P1 78 4 1 0.60 0.56 0.4 0.6 4 3 34 P1 85 40 0 0.90 1.20 1.1 1.4 8 5 35 C 79 12 0 0.60 0.86 0.9 1.3 6 5 36 C 82 24 0 0.86 0.60 0.5 0.6 5 4 37 C 87 24 0 1.10 1.13 1.1 1.3 4 3 38 * C 85 3 0 0.42 0.64 0.5 0.7 5 3 39 C 95 36 1 0.90 1.13 1.0 1.5 6 5 40 C 81 12 0 1.00 1.00 1.0 1.3 5 5 41 C 80 12 0 0.56 0.64 0.6 1.0 6 5 42 C 85 6 0 1.30 1.23 1.0 1.3 8 6 43 C 84 6 0 0.96 1.00 0.9 1.2 5 5 44 C 73 18 0 0.86 0.80 0.7 1.0 10 6 45 C 84 24 0 0.74 1.10 1.1 1.3 5 5 46 C 83 7 1 0.90 1.10 1.2 1.4 4 3 * Subjects who withdrew from the study. + Groups C - Clinical (control) group N - Control group P1 - Large print practice group P2 – large print with reduced field of view practice group ‡ The length of visual impairment was either reported by the subjects or recorded in the clinic database VA = visual acuity CPS = critical print size STM = Stand magnifier Type of AMD 0 = dry AMD 1 = wet AMD


145

4.2.2 Exclusion Criteria

The major exclusion criteria to subjects’ selection related to their ocular health and

near visual acuity. Subjects with any significant ocular pathology other than AMD,

for example nuclear cataract of more than NC2 sclerosis or mild or advanced

glaucoma, were excluded from the study. In addition, subjects whose monocular near

visual acuity in the better eye was worse than N40 at 25 cm (1.4 logMAR) were

excluded from this study. This was because the study was mainly focused on AMD

subjects with moderate visual impairment rather than those who had severe vision

problems.

Other exclusion criteria related to subjects’ general health. Anyone whose clinical

records indicated they had poor general health (e.g. heart disease), likely unstable

vision (e.g. due to diabetes) or cognitive problems reported (either reported in the

record card or self-reported by the subjects) that might affect their reading

performance, were not recruited. Previous studies have indicated that reading rate

could be reduced due to difficulties with comprehension among people with normal

(Shankweiler and Liberman, 1972) or low vision (Baldasare et al., 1986). Therefore,

a simplified Neale Analysis of reading ability (refer to section 3.2.2) was used to

confirm that all recruited subjects had Grade 6 or above reading ability (Neale et al.,

1999).

In summary, subjects recruited underwent a preliminary assessment before inclusion

in the study to ensure that they fulfilled the following participation criteria:

1. Primary diagnosis of AMD (confirmed by ophthalmologist)

2. Monocular near visual acuity of 1.4 logMAR or better

3. Minimal lens opacities (≤NC2)

4. No other ocular disease

5. No history of diabetes

6. Minimum reading ability of grade 6

7. No previous experience in using STM


146

4.2.3 Sample size

Previous studies have shown that reading rate for people with low vision varies

according to different visual acuities and low vision causes (Legge et al., 1985 (b);

Legge, 1991; Legge et al., 1992; Whittaker and Lovie-Kitchin, 1993; Song et al.,

1996; Patel et al., 2001). Reading rate varies greatly, especially among people with

AMD (Sunness et al., 1996; Lovie-Kitchin et al., 2000 (a); Bowers et al., 2001 (b);

Martin et al., 2002). The reading rates measured in this study were no exception. A

large standard deviation was found, which was almost half the mean of each

measure. Because of this wide variation in reading rate between subjects, a large

sample size would be needed to give statistically significant results. Based on the

standard deviation of reading rate from Chapter 3, a sample size of approximately 60

subjects would be needed for each group (a total of 248 subjects) to achieve

sufficient statistical power to show the suggested differences between groups

(Appendix 2). Because of the difficulty in recruiting AMD subjects who were all

older people, a small sample size was obtained in the present study. To achieve the

required number of participants, a longer recruitment period and involvement of

multiple clinical or research centres would be required, which was beyond the

feasibility of the present study.

4.2.4 Subject groups

Subjects were divided into three groups (control group – N; large print practice group

– P1; large print with reduced field of view practice group – P2) but they were not

randomly allocated to the groups. They were assigned to groups such that the

distributions of age and near visual acuities were similar in the three groups. Subjects

were categorised into good (less than 0.7 logMAR), moderate (0.7 to 1.0 logMAR)

and poor vision (worse than 1.0 logMAR) according to their near visual acuities. As

near as possible, equal number of subjects from each category were assigned to each

of the experimental groups. Experimental interventions used for each group are

described in section 4.3.1. An additional control group – the clinical group (C) - was

recruited. Participants in the clinical group were patients who attended the QUT

VRC or Vision Australia LVC and who had a STM prescribed as their primary low

vision reading aid by either student optometrists or registered optometrists. Their


147

recruitment criteria were the same as those used for the three experimental groups.

Details of subjects in each group are summarised in Table 4.1.

4.2.5 Multi-centre assessment

The initial assessments (first visit) of both vision and reading were conducted at three

different sites (laboratories): the Centre for Eye Research at QUT (n=28), the Vision

Australia LVC (n=3) and The Eye Centre (n=12). Procedures for all assessments

were the same at each site. This meant that there was no difference in lighting and

the use of equipment for all measurements (e.g., vision charts) between sites. The

author collected all data at the three different sites. One way ANOVA indicated that

the baseline vision and reading measuresa were not significantly different for subjects

assessed at the three different sites (p>0.05).

4.2.6 Home assessment

This experiment was a longitudinal study, therefore subjects were requested to attend

one of the laboratories (mentioned above in section 4.2.5) several times to monitor

their reading performance across time. All of them were visually impaired and many

had no or limited access to transportation. In order to increase the chance of

recruitment for this research study, home visits under controlled conditions were

offered as an alternative to laboratory visits. For subjects who preferred home

assessment, they attended the laboratory for their first visit while all other visits were

conducted at their home. The experimental set up and measuring conditions such as

the test distances and reading illumination at home were made the same as those used

in the laboratory. For example, an additional table lamp or extra lighting was added

to ensure the illumination on reading material was the same as that used in the

laboratory (range of 320 to 380 lux according to different reading distance).

a The vision and reading measures included distance and near visual acuities, visual field loss (with a 5 mm target) and threshold print size achieved with STM (see section 4.3.2).


148

4.3 Methods

4.3.1 Experimental interventions for each group

The design and experimental procedures for each group are summarised as follows

(Figure 4.1):

1. Control group (N) – no reading practice was given at home but repeated

reading measures with and without STMs were taken for two weeks (weeks

0, 1 and 2) before the STMs were supplied by the experimenter for home use.

The STMs were supplied at week 2 such that subjects could use them for

reading at home (refer to section 4.3.6.2). Repeated reading measures with

STMs were taken at weeks 4, 8 and 20. The control group was used to

compare the reading performance with and without STM for subjects who

received no reading practice.

2. Large print practice group (P1) – large print reading (refer to section 4.3.7)

with formal instructions to read for 10 minutes per day was given for subjects

to practise at home during the first two weeks. Repeated reading measures

with and without STMs were taken for two weeks (at weeks 0, 1 and 2)

before the STMs were supplied at week 2 by the experimenter for home use.

Repeated reading measures were taken after the STMs were prescribed for

daily reading at weeks 4, 8 and 20. The effect of large print reading practice

was investigated by comparing reading performance with and without STMs

for P1 with the control group.

3. Large print with reduced field of view practice group (P2) – subjects were

requested to practise reading large print at home for 2 weeks with formal

instructions to read for 10 minutes per day but with a restricted field of view

(refer to section 4.3.4) during the first two weeks. Repeated reading measures

with and without STMs were taken for two weeks (at weeks 0, 1 and 2)

before the STMs were supplied by the experimenter for home use at week 2.

Afterwards, subjects were asked to practise using the STMs for reading, and

repeated reading assessments with STMs were taken at weeks 4, 8 and 20.

The effect of large print reading practice under a restricted field of view was

investigated by comparing reading performance with and without STMs with

that of P1 and with the control group.


149

Figure 4.1

Flow chart of experimental interventions for each group

Clinicians or student optometrists in clinics prescribed stand magnifiers (at week 0 but

equivalent to week 2 for experimental groups) (see section 4.3.1).

Control group (N)

Sources for subjects for experimental groups • QUT VRC • Referral from local ophthalmologists

Sources for subjects for clinical group: • QUT VRC • Vision Australia LVA

Large print practice group 1

(P1)

Large print with reduced field of view practice

group 2 (P2)

Clinical group (C)

No reading practice was

given

Large print reading practice

was given

Large print reading practice under restricted field of view was given

Experimenter prescribed stand magnifiers (at week 2)

Reviewed the reading performance with stand magnifiers at weeks 4, 8 and 20.


150

4. Clinical group (C) - subjects were recruited directly from the QUT VRC or

Vision Australia LVC. This group was included to investigate whether the

experimental interventions received by the other groups (N, P1 and P2) (such

as the repeated measures of reading rate prior to the provision of a STM) had

any impact on the reading performance with STMs across time.

None of the subjects in the clinical group received any experimental

intervention before the STMs were supplied. Thus no repeated reading

measures with and without STMs were made between weeks 0 and 2. Hence,

the number of visits for clinical group subjects (4 visits) was less than that for

the experimental groups (6 visits). The STMs were prescribed by the

optometrists from the low vision clinics for home use at their initial visit,

which was regarded to be equal to the week 2 visit for the other experimental

groups.

Therefore, the first visit for the clinical group subjects was coded as “week 2”

to avoid any confusion of the “visit number” among different groups.

Repeated reading measures with STMs were then taken at the same time

intervals as those for the other experimental groups after the STMs were

supplied: weeks 4, 8 and 20.

Commonly, patients in low vision clinics are reviewed after a two-week loan of the

prescribed magnifiers to assess their reading performance with the new low vision

aids. A similar regime was adopted for this study. Subjects’ reading performance

with their STMs was reviewed after two weeks home practice (week 4 of study)

using the STMs. In order to assess any further improvement in reading rate with the

magnifiers over a longer period, another two follow up visits, which were scheduled

six weeks (week 8) and eighteen weeks (week 20) after the STM prescription were

included in this study.

4.3.2 Preliminary assessment

A full optometric examination was conducted for each subject prior to the

experiment to ensure that subjects’ refraction and spectacle prescription gave the best


151

vision. This included objective and subjective refractions, ophthalmoscopy, visual

acuities, visual field and contrast sensitivity. All subjects were screened prior to their

participation to ensure that they met the inclusion criteria (section 4.2.1). Results of

vision measures for each subject are given in Table 4.1.

4.3.2.1 Visual acuities

Distance visual acuity was measured monocularly using a high contrast (93%)

Bailey-Lovie (logMAR) letter chart with subject’s current spectacle prescription and

best refraction (Bailey and Lovie, 1976). Details of the Bailey-Lovie letter chart and

methodology of distance visual acuity measurement have been described in section

3.3.2.

Near word visual acuity was measured monocularly with Bailey-Lovie word charts

(Bailey and Lovie, 1980). The design of this chart has been described in section

2.3.2.1. Six versions of the Bailey-Lovie near word reading charts were used and

were randomly presented to prevent chart familiarity. Procedures of the near visual

acuity measurement have been described in section 3.3.3.

4.3.2.2 Visual field

The central 25° visual field was measured monocularly using a Tangent (Bjerrum)

screen placed at 1 m. Twelve meridians, separated by 30°, were assessed with two

target sizes. White Traquair 10/1000 and 5/1000 targets were moved from non-

seeing to seeing at a speed of 5° per second (Henson, 1993). In order to maintain

steady fixation, a letter “E” of 1.3 logMAR for 1 metre test distance in Times New

Roman font was used as the fixation point at the centre of the screen. Subjects wore

their distance correction and were instructed to direct their gaze at the letter "E"

using their habitual fixation (central or eccentric) such that they could see the letter

"E" steadily (Lovie-Kitchin and Whittaker, 1998 (b)). The average background

luminance of the screen was 62 cd/m2. Locations and sizes of the area where the

subjects did not see the Traquair targets were recorded. The presence of any

eccentric fixation and its eccentricity was determined by the location of the blind

spot compared to the physiological blind spot position on a standard visual field


152

recording sheet. Except for the blind spot, all other scotomas were categorised into

four quadrants according to their location relative to fixation (central or eccentric

fixation). If the scotoma was located in the quadrant bounded by 45° and 135°, it was

classified as a superior visual field loss. Similarly, scotomas located between 135°

and 225°, 225° and 315° and 315° and 45° were categorised as left, inferior or right

visual field defect respectively (Lovie-Kitchin et al., 2000 (a)). For subjects who had

scotoma around the fixation target in all quadrants, the scotoma was expressed as

central or ring depending on whether the central visual field remained intact or not.

In addition to the qualitative categorization of the scotoma location, the visual fields

measured by both target sizes were also scored as the solid angles subtended by the

scotoma, expressed as a percentage of a sphere (Weleber and Tobler, 1986; Lovie-

Kitchin et al., 1990). By calculating the difference between the solid angle subtended

by the subject’s central visual field extent and that subtended by the scotoma, the

percentage of the central field remaining was determined (Appendix 3).

Monocular visual field assessment of the better eye was only conducted if there was

a significant difference in near visual acuities between two eyes. If both eyes had

similar near visual acuity, the visual field of each eye was measured. The visual field

of the eye selected for reading measures was used in the analysis. Table 4.2

summarises the position of the scotoma and the percentage of visual field loss for

each individual subjectb. Visual field losses measured by the two targets are given in

this table; however only the field measured with the 5-mm target was used in the

analysis.


There are many tests commercially available to assess contrast sensitivity. Leat and

Woo (1997) investigated several different types of contrast sensitivity charts and

suggested that Pelli-Robson chart was the best predictor of reading rate in low vision

patients due to its high repeatability and wide range of spatial frequency being

measured.

b Subject 45 did not have visual field assessment, so his case was considered as missing data in the analysis.


153

Table 4.2 Subjects’ contrast sensitivity and visual field results

Visual field 10 mm Traquair target 5 mm Traquair target ID

Group Contrast

sensitivity (log CS) Qualitative

(Quadrant) + Ster -adians

% of sphere

Qualitative (Quadrant) +

Ster -adians

% of sphere

1 P2 1.60 R 0.0039 0.031 R, L, I, S (ring) 0.029 0.23 2 P2 1.05 R, L, I, S (ring) 0.0017 0.014 0 0 0 3 P2 1.65 0 0 0 R, I 0.005 0.039 4 P2 1.10 I 0.0047 0.038 R, L, I, S (central) 0.003 0.24 5 P2 1.05 R, L, I, S (central) 0.012 0.095 R 0.0015 0.012 6 P2 1.30 0 0 0 L, S 0.008 0.065 7 P2 1.15 L 0.0016 0.013 R, L, I, S (ring) 0.029 0.23 8 * P2 0.95 R, L, I, S (central) 0.035 0.28 R, L, I, S (central) 0.038 0.30 9 P2 1.15 0 0 0 0 0 0 10 P2 1.45 0 0 0 I 0.002 0.0017 11 P2 1.00 R 0.114 0.91 R 0.13 1.037 12 P2 1.10 S 0.004 0.034 S 0.012 0.098 13 N 1.00 R, L, I, S (ring) 0.009 0.072 R, L, I, S (central) 0.085 0.68 14 N 1.10 R, L, I, S (central) 0.003 0.26 R, L, I, S (central) 0.067 0.53 15 * N 0.75 R, L, I, S (central) 0.05 0.041 R, L, I, S (central) 0.081 0.65 16 N 1.30 I 0.0023 0.018 0 0 0 17 N 1.05 0 0 0 S, I 0.0053 0.043 18 N 1.30 S 0.0068 0.054 L, S, I 0.051 0.4 19 N 1.15 L, S, I 0.034 0.27 S, I, L 0.012 0.094 20 N 1.15 I, L 0.0035 0.0028 0 0 0 21 N 1.30 0 0 0 0 0 0 22 N 1.05 0 0 0 0 0 0 23 N 1.05 0 0 0 0 0 0 24 P1 1.40 0 0 0 R, L, I, S (central) 0.017 0.14 25 P1 1.30 R, L, I, S (central) 0.0082 0.0065 R, L, I, S (central) 0.034 0.27 26 P1 1.15 R, L, I, S (central) 0.02 0.157 R, L, I, S (central) 0.041 0.057 27 P1 1.20 L, S, I 0.0068 0.054 R, I, S 0.013 0.1 28 P1 1.00 I, R 0.014 0.11 0 0 0 29 P1 1.00 0 0 0 S, I, R 0.062 0.49 30 P1 0.90 S, L, R 0.036 0.28 R 0.0079 0.063 31 P1 1.20 R 0.0042 0.033 0 0 0 32 P1 1.35 0 0 0 S, L 0.0021 0.017 33 P1 1.20 S, L 0.0021 0.017 R, S 0.014 0.11 34 P1 1.15 R, S 0.018 0.146 R, L, I, S (ring) 0.012 0.094 35 C 1.05 0.019 0.15 R, L, I, S (ring) 0.12 0.99 36 C 0.55 R, L, I, S (ring) 0.097 0.78 L, R 0 0 37 C 1.20 L, R 0.0071 0.056 0 0 0 38 * C 39 C 1.15 I 0.035 0.28 I 0.047 0.37 40 C 1.00 L, I 0.038 0.3 L, I 0.053 0.42 41 C 1.00 R, L, I, S (central) 0.086 0.68 R, L, I, S (central) 0.13 1.06 42 C 1.10 R, L, I, S (central) 0.071 0.57 R, L, I, S (central) 0.11 0.89 43 C 1.15 Ring (more at I, R) 0.029 0.23 I, L, R 0.049 0.39 44 C 1.10 R, L, I, S (central) 0.066 0.53 R, L, I, S (central) 0.081 0.65 45 C 0.95 46 C 0.85 S 0.038 0.3 S 0.008 0.065

* Subjects who withdrew from the study. + Qualitative categorisation of visual field loss: Superior (S); Inferior (I); Right (R); Left (L). Ring scotoma - the scotoma surrounding the central fixation target with very small area of central vision. Subject 45 did not have visual field assessment, so his case was considered as missing data in the analysis.


154

The Pelli-Robson chart consists of sixteen triplets of letters and each letter subtends 3

degrees at the recommended viewing distance of 1 metre (Pelli et al., 1988). This

would be expected to measure contrast sensitivity at a spatial frequency of

approximately one cycle per degree. Within each triplet the contrast of letters is the

same. For each successive triplet, the contrast decreases by a factor of 0.15 log units.

The Pelli-Robson chart was placed at a testing distance of 1 m with an average

background luminance of 100.6 cd/m2. This illumination was well within the

recommended range of 60 to 120 cd/m2 for this test (Woods and Woods, 1995).

Subjects, with their worse eye occluded, were instructed to read the letters, starting

with the letters of high contrast until no letters in a given triplet were read correctly.

They were encouraged to look at a triplet for at least 20 seconds, to give them

sufficient time to recognise the letters near their contrast threshold. Credit of 0.05 log

unit was given for each letter read correctly (Elliott et al., 1990 (a); Elliott et al.,

1990 (b)). If the letter ‘C’ was read incorrectly as ‘O’, this was scored as correct

(Elliott et al., 1990 (a)). Log contrast sensitivity score was calculated by recording

the total number of letters read correctly, subtracting three and multiplying by 0.05

(Table 4.2). Contrast sensitivity was measured once only at each subject’s first visit.

4.3.3 Reading assessment

Based on the results of the study in Chapter 2, Bailey-Lovie text reading charts were

used to measure reading rate on sentences for a range of different print sizes.

Different versions of this chart with different sentences were used in this study. The

design of the Bailey-Lovie text chart has been described in section 2.3.1.

In addition to assessing reading rate on sentences, reading measures were also

conducted using longer passages. Text for these passages were selected from

different sources: Sloan Reading Cards for low vision patients, Lighthouse Reading

Cards, near vision text card from the University of Waterloo, and Maclure bar

reading book for children (Murray and Scheme). The reading level of each passage

had sixth grade reading level or below, analysed by the Flesch-Kincaid Grading

Level System (Microsoft Word 2000). Passages were reproduced in proportionally

spaced Times New Roman font for each print size (N64 to N8) in 0.1 logarithmic


155

steps. Each passage contained 263 ± 1.41 characters (43.8 standard words) and was

divided into 8 lines (Figure 4.2). In Bailey-Lovie text reading charts, each sentence is

printed over two lines at each print size and is left justified such that the length of the

second line is slightly shorter than the first one. For consistency, a similar layout for

the passages, with progressively shorter rows and left justification was designed.

Twelve different passages for each print size were printed across A4 cards with

landscape orientation on a laser printer in black on white print. Sufficient versions of

the passages were available in this study such that three measures of reading rates

without magnifiers could be taken with different passages. The contrast of the print

was 90% and the text was left justified with single-spacing between lines. The

passages were randomly selected at each visit to compensate for any differences in

text difficulty; and so reduce the likelihood of learning effects.

Navigation requirements for each STM depend on the length of the passages and the

magnification of the magnifiers. The larger lens diameter of lower powered

magnifiers results in less or minimal magnifier manipulation across the page

compared to that required for stronger magnifiers if the width of the passage is not

long enough. In order to provide similar navigation requirements for low vision

subjects with different powers of STMs, longer passages (> 600 words) were also

needed (Figure 4.3). An unavoidable choice was made between keeping length of

rows the same, either in terms of numbers of words (or characters), or in terms of

distances. In this study, the length of rows in terms of distance were maintained the

same. Print size of these passages ranged from N16 to N4 in 0.1 logarithmic steps

(Figure 4.3). When the print size became smaller, more words per line were needed

to keep the width of each line and the total number of lines constant such that the

page navigation requirement was similar for each print size.

Therefore, the number of words (or characters) increased when the print size

decreased from N16 to N8 (Table 4.3). However, for print sizes from N6 to N4, the

number of words (or characters) was the same so that the length of the passages was

not a limiting factor reducing the reading rate due to fatigue. Twelve different

passages were printed from a laser printer across A4 cards with landscape orientation

for each print size. These cards were then laminated.


156

The children decide to explore a nearby mountain. They pack their equipment and then set off. They all walk quite quickly at first, but slow down as the road becomes steeper. After some time they leave the road and take to a mountain path. They go along the bank. Print size: N24 (Number of characters: 264) The children decide to explore a nearby mountain. They pack their equipment and then set off. They all walk quite quickly at first, but slow down as the road becomes steeper. After some time they leave the road and take to a mountain path. They go along the bank Print size: N10 (Number of characters: 264)

Figure 4.2

Example of a reading passage for large print reading


157

Now they walk to the woods. They walk for some way in the woods and find a nice place they know. They all sit by the trees. The sun is out but it is not hot in the woods. They think there is a road in the woods which will take them to the farm. After some time Peter says to Jane, "I do not know the way to go. We have never been here before. Do you know the way?" "No," says Jane. "There are too many trees for me to see the way. I don't know where we are." But Daddy can see the road so they all go on to find it. They go along the road and then come out of the wood. As they walk home along the road, the rain comes on. They all get very wet in the rain, but the children think it is fun. A van comes along the road. In it is a man they know very well. He says he will take them home. Print size: N12 (Number of characters: 788) Now they walk to the woods. They walk for some way in the woods and find a nice place they know. They all sit by the trees. The sun is out but it is not hot in the woods. They think there is a road in the woods which will take them to the farm. After some time Peter says to Jane, "I do not know the way to go. We have never been here before. Do you know the way?" "No," says Jane. "There are too many trees for me to see the way. I don't know where we are." But Daddy can see the road so they all go on to find it. They go along the road and then come out of the wood. As they walk home along the road, the rain comes on. They all get very wet in the rain, but the children think it is fun. A van comes along the road. In it is a man they know very well. He says he will take them home. The van gets to the road where the children live, and they all get out and go into the house. Then it is time for the children to go to bed. Next day the children get up, and Jane says, "It is hot today and we can go and play in the woods." Peter and Jane call for their friends Tom and Mary, and they all set off. As they walk along the street, many cars go by, but soon they come to the end of the road, and they go along a path with trees at the side.

Print size: N8 (Number of characters: 1243)

Figure 4.3

Example of a reading passage for reading with stand magnifiers


158

Table 4.3 Mean number of characters and standard words for passages of different print sizes used for measuring reading rate with stand magnifiers

Print size (N-notation) N16 N12 N10 N8 N6-N4

Mean number of characters 617.7 839 996 1241 1662.3 Mean number of standard

words 102.9 139.8 166 206.8 277

In summary, reading rates were measured with both the sentences (Bailey-Lovie text

reading charts) and passages of text (at CPS) for all conditions. Reading rates with

and without magnifiers across time were the major outcome variables investigated in

this study. Additionally, vision measures such as distance and near visual acuities

and threshold print size achieved with magnifiers were important variables to

monitor vision stability across time.

4.3.4 Practice stand for reduced field of view

A device which simulated the reduced field of view of a STM was introduced to

subjects in the large print with reduced field of view practice group (P2) for their

reading practice at home (Figure 4.4a). The device was made of a small plastic

transparent rectangular stand open at the top and bottom. A rigid (grey) card of

approximately 14 cm width and 6 cm length with a central rectangular aperture was

attached to the bottom of the plastic stand. The aperture was used to narrow the field

of view to approximately 6 characters horizontally and 3 lines vertically (Figure

4.4b). Thus the size of the aperture varied according to print size of the reading

material. Table 4.4 gives the horizontal dimensions of each aperture for different

print sizes. The extended section of the card from the stand was used to prevent the

subject reading beyond the limit of the stand. As the circumferences of the

rectangular stand were transparent, it was covered by white paper with a central

rectangular aperture on each side (Figure 4.4). This prevented subjects from reading

through the transparent sides but allowed sufficient illumination on the reading

materials.


159

Table 4.4 Aperture width of practice stand required for each print size to give approximately 6 characters of horizontal field of view

Print size

(N-notation) Aperture width

(mm) Print size

(N-notation) Aperture width

(mm) 64 67 20 21 48 50.5 16 16.5 40 42 12 12.5 32 33.5 10 10.3 24 25

Figure 4.4a

Practice stand to simulate reduced field of view

The outside of the rectangular stand was covered by white paper with a central rectangular aperture on each side. This prevented subjects from reading through the transparent sides but allowed sufficient illumination on the reading materials.

The extended section of the card from the stand was used to prevent the subject reading beyond the limit of the stand.

A rigid (grey) card with a central rectangular aperture was attached to the bottom of the stand. The aperture was used to narrow the field of view to 6 characters horizontally and 3 lines vertically.

The practice stand was made of a small plastic transparent rectangular stand open at the top and bottom


160

Figure 4.4b

Practice stand to simulate reduced field of view (continued).

The aperture of the practice stand narrowed the field of view to approximately 6 characters horizontally and 3 lines vertically. The size of the aperture varied according to print size of the reading material.

4.3.5 Stand magnifiers

The 1550 series of Eschenbach illuminated STMsc of different equivalent powers

(5.7–46.04 D) were used as the low vision aids in this study (Table 4.5) The optical

parameters of each STM were measured (refer to Appendix 4 for methods). They

were internally illuminated with two C-sized 1.5 volt batteries or by mains power

with plug-in handles according to subjects’ preferencesd. Illuminance of the reading

materials with the STMs and two new alkaline batteries (Energizer size C 1.5 volts)

ranged from 600 to 900 lux depending on the size of the magnifiers. For those whose

STMs were connected to main power, the illuminance of reading materials ranged

from 1200 to 2000 lux.

c Eschenbach Catalogue 2001/2002 d As magnifiers for the clinical group were prescribed in low vision clinics, different types of STMs would have been tried before loan. Among the subjects in this group, two subjects insisted on STMs with main powered illumination.


161

Table 4.5 Eschenbach illuminated stand magnifiers - Optical parameters measured

Descriptions Measured Z=2.5 (cm) Z=5 (cm) Z=10 (cm) Z=15 (cm) Z=20 (cm) Z=25 (cm)

Code number

Labelled dioptric Power

Fe (D)

l' (cm)

F* (D)

EVD (cm) ey/im F*

(D) EVD (cm)

ey/im (cm)

F* (D)

EVD (cm)

ey/im (cm)

F* (D)

EVD (cm)

ey/im (cm)

F* (D)

EVD (cm)

ey/im (cm)

F* (D)

EVD (cm)

ey/im (cm)

15878 5 5.71 10 12.57 7.96 12.5 10.48 9.55 15 7.86 12.73 20 6.29 15.91 25 5.24 19.09 30 4.49 22.27 35 15788 10 10.14 8.5 16.93 5.91 11.0 13.79 7.25 13.5 10.07 9.94 18.5 7.92 12.62 23.5 6.53 15.31 28.5 5.56 17.99 33.5 15849 7 5.85 11.56 11.92 8.39 14.06 10.12 9.88 16.56 7.78 12.86 21.56 6.31 15.84 26.56 5.31 18.82 31.56 4.59 21.81 36.56 15589 12 11.18 13.96 15.55 6.43 16.46 13.5 7.41 18.96 10.68 9.36 23.96 8.84 11.31 28.96 7.54 13.27 33.96 6.57 15.22 38.96 15549 16 14.7 19.9 17.53 5.71 22.4 15.77 6.34 24.9 13.13 7.62 29.9 11.25 8.89 34.9 9.84 10.16 39.9 8.74 11.44 44.9 15539 20 17.02 25.14 19.1 5.24 27.64 17.51 5.71 30.14 15.02 6.66 35.14 13.15 7.6 40.14 11.69 8.55 45.14 10.53 9.5 50.14 15527 23 22.53 33.76 23.73 4.21 36.26 22.2 4.5 38.76 19.67 5.09 43.76 17.65 5.67 48.76 16.01 6.25 53.76 14.65 6.83 58.76 15517 28 26.46 31.94 27.45 3.64 34.44 25.59 3.91 36.94 22.54 4.44 41.94 20.14 4.97 46.94 18.20 5.49 51.94 16.60 6.02 56.94 15507 38 37.55 35.26 37.71 2.65 37.76 35.37 2.83 40.26 31.46 3.18 45.26 28.33 3.53 50.26 25.77 3.88 55.26 23.63 4.23 60.26 15577 50 46.04 41.56 45.70 2.19 44.06 43.25 2.31 46.56 39.05 2.56 51.56 35.60 2.81 56.56 32.71 3.06 61.56 30.25 3.31 66.56

Fe = Equivalent power of the STM (D) F* = Equivalent Viewing Power (EVP) (D) EVD = Equivalent Viewing Distance (cm) Z = Distance between eye and lens surface (cm) l' = image distance (cm) Ey/im =Eye-to-image distance (cm) Details of the optical parameters measurements of STMs are given in Appendix 4.


162

4.3.6 Procedures

All experimental measurements were taken under monocular viewing conditions.

Details of the reading measures are summarized in Table 4.6. The eye with better

near visual acuity was defined as the eye for monocular reading assessment with and

without STMs. However, if near visual acuity was the same for each eye, the eye

with the more extensive right visual field or the eye with the smaller scotoma was

defined as the ‘reading’ eye.

4.3.6.1 Reading assessment without stand magnifier

Reading rates for both sentences and passages (at CPS) were assessed at each visit

(weeks 0 to 2) before the STM was supplied and also at the last visit (week 20). The

procedures for reading assessment and the analyses of reading measures have been

described in section 2.3.2. The methods for determining critical print size (CPS), text

visual acuity and maximum reading rate for sentence reading have been described in

section 3.3.5.1.

The print size of passages for measuring passage reading rate without STMs and the

print size of large print books for home practice were selected according to

individual subject's CPS (see below in section 4.3.7). All reading measures were the

mean of three repeated measurements.

4.3.6.2 Prescription of stand magnifiers

The required magnification for reading depends on the print size of the reading task

that the subject wants to achieve (Cole, 1993; Lovie-Kitchin and Whittaker, 1999

(b)). Hence, it is essential to know the goal reading material and therefore the target

print size. For example, if the goal reading materials are newspaper and magazine,

then the target print sizes will be N8 and N12 respectively (Lovie-Kitchin and

Whittaker, 1999 (b)).


163

Table 4.6 Summarised assessment of different groups at each visit

Time Control (N) (n=10)

Large print practice (P1) (n=11)

Large print with reduced field of view practice (P2) (n=11)

Clinical Control (C) (n=11)

Before stand magnifier prescription

Vision assessment (Distance and near visual acuities) Reading rate with large print sentences and determining critical print size (CPS) on large print Reading rate using large print with passages at CPS Determination of magnification (EVD) of stand magnifier Reading rate with magnifiers with sentences and determining CPS achieved with stand magnifier, modification was made if the goal print was smaller than CPS by 0.1 log unit or more Reading rate with magnifiers with passages at CPS

Reading rate with large print and practice stand using sentences at CPS and passages at CPS

Weeks 0 to 2

Reading practice on large print at home

Reading practice on large print with practice stand at home

Week 2 Stand magnifiers measured at week 0 were prescribed by the experimenter for home use Stand magnifiers were prescribed by optometrists in low vision clinics

After stand magnifier prescription

Weeks 4, 8 and 20 Vision assessment (Distance and near visual acuities) Reading rate with magnifiers with sentences and determining CPS achieved with stand magnifier Reading rate with magnifiers with passages at CPS

Week 20 Reading rate with large print sentences and determining CPS on large print Reading rate with large print using passages at CPS Reading rate with large print and practice stand using sentences (at CPS) and passages (at CPS)


164

Results from Chapter 3 showed that there was no significant difference in reading

rates with magnifiers selected either by the fixed or individual acuity reserve

methods suggested by Whittaker and Lovie-Kitchin (1993) and Legge and colleagues

(Legge et al., 1992; Ahn et al., 1995) respectively. A fixed acuity reserve of 0.3 log

unit was used in this study to calculate the appropriate magnification of STM.

However if CPS with the magnifier was larger than the target print size by more than

0.1 log unit, this suggested that the calculated magnification was insufficient to allow

subjects to read the target print size fluently. Therefore, magnification was re-

calculated by using the individual acuity reserve method (see Chapter 3). Details of

the calculation of magnification (in terms of equivalent viewing distance) have been

described in section 3.3.5.2. Based on the calculated EVD and the near addition of

the subject, an appropriate STM with recommended eye-to-lens was selected (Table

4.5). For subjects participating in the experimental groups, the reading performance

achieved with the magnification calculated by this fixed acuity reserve method was

as expected for 30 of the 32 subjects. For the two subjects (subject 1 and subject 25)

whose reading performance with the magnifier did not reach the predicted level, the

individual acuity reserve method was used to calculate the magnification; 0.5 log

units of acuity reserve was required in each case. There was no obvious reason for

the different results for them.

Nilsson (1990) suggested that detailed instructions regarding the use of optical aids

were very valuable in improving visual acuity and reading ability for low vision

patients. In this study, systematic instructions in the use of the STM were given. All

subjects (except those in the clinical group) were given the following instructions by

verbal communication, visual and/or tactile demonstration and hands-on

performance:

1. The STM must rest flat on the page (not raised or tilted).

2. The recommended working distance between the eye and lens (z) was given to

ensure appropriate magnification and satisfactory focus.

3. Handling technique for moving the magnifier along a line (forward movement)

and back for a new line (retrace) was demonstrated.

4. The method for changing batteries was demonstrated.

5. Subjects were strongly encouraged to change batteries once the illumination

started to dim.


165

6. Appropriate posture for reading with the STM was strongly recommended. For

example, by using a reading stand or raising the reading materials by piles of

books or other supports underneath.

4.3.6.3 Reading assessment with stand magnifiers

Reading rates with STMs for both sentences and passages (at CPS) were measured at

each visit (Table 4.6). Subjects were given sufficient time to become familiar with

the use of STMs before any measurements were taken. Similar to the reading

assessment without magnifiers, reading rates, CPS and text visual acuity were

measured three times on sentences from Bailey-Lovie text reading charts. CPS

achieved with STM was determined to the nearest log unit and reading rates on

passages at the CPS were measured. Passages used in this assessment were the

longer passages (Figure 4.3). To ensure that illuminance provided by the STM did

not affect reading performance, two new alkaline batteries were used at each visit for

the reading assessments.

As the object distance is less than the focal length of the STM, any change in the

working distance between the eye and lens results in a change in magnification.

Therefore, it was important to record the working distance (eye-to-lens distance) at

each visit after the STM was supplied (at weeks 4 to 20) to identify any changes in

the magnification that subjects adopted between visits. The eye-to-lens distance was

measured twice when the subject was reading with the STM; however, this distance

was not kept constant for measurement.

4.3.6.4 Reading assessment with practice stand

In addition to the reading measures with and without STM, repeated measurement of

reading rates under reduced field of view on large print for sentences and passages at

CPS were included at week 0 to week 2 for subjects in P2.

Reading rates on both sentences and passages at CPS were measured under a

restricted field of view simulated by a practice stand at week 20 for all subjects. This

was used to compare any differences between groups in reading rates under full and


166

restricted field of view on large print. In addition, reading rates under the restricted

field of view could be compared with the reading rates with STMs on passages at

CPS in which the retinal image sizes at the two reading conditions were the same

(same level of magnification).

4.3.6.5 Questionnaires

In addition to the objective measures of reading performance, subjects’ subjective

responses on reading with STMs were investigated. A questionnaire, derived from

the Manchester low vision questionnaire (MLVQ) (Harper et al., 1999) was

administered verbally to subjects at weeks 4 to 20 after the provision of a STM for

home use as well as before the provision of the STM at week 2. This questionnaire

(Appendix 5) was not a validated questionnaire but modified from the MLVQ that

addressed the frequency and duration of reading with the STM and the tasks that the

subjects used the magnifier to read, which were most relevant to this study. Most

questions used 5-point Likert scales for classification of the responses (Oppenheim,

1992).

4.3.7 Large print books for reading practice

Large print reading material selected from a storybook of 6th grade level was

prepared for a range of print sizes from N64 to N10. These large print materials were

printed on A4 papers with landscape orientation on a laser printer and bound into a

book (Figure 4.5). Many studies have shown that reading rate increases as print size

increases from threshold (Legge et al., 1985 (b); Lovie-Kitchin and Woo, 1987;

Rubin and Turano, 1994; Plass and Yager, 1995), so it was important to ensure the

print size of reading material was larger than the threshold visual acuity for fluent

reading. Legge and colleagues suggested that fluent reading could be achieved at

CPS, the smallest print size for maximum reading rate (Legge et al., 1989 (a); Legge

et al., 1992; Mansfield et al., 1993; Mansfield et al., 1994). Therefore, individual

CPS for each subject from the two practice groups (P1 and P2) was assessed at week

0 to select the appropriate print size of the large print book for home reading

practice.


167

Figure 4.5

Example of a large print reading book (N24 print).

Large print reading book was prepared for a range of print sizes from N64 to N10 and was given to subjects from both practice groups (P1 and P2). They were requested to practise reading daily for at least 10 minutes for 2 weeks and to record the number of pages they read each day.

Subjects from both practice groups were instructed to read at least 10 minutes per

day for 2 weeks. The number of pages they read each day was recorded in an attempt

to verify the compliance with reading practice. Subjects in P2 were required to do the

large print reading practice under a restricted field of view using the practice stand.

No objective measures were made to assess the subjects’ compliance on home

reading practice. Instead subjects’ self reports of their compliance were used.

Previous research studies have shown a low compliance rate for patients in taking

medications (eye drops) for treatment of their eye disease (Kanski, 1999) or for

contact lens maintenance (Collins and Carney, 1987). However this might not apply

to the subjects recruited in this study. All participants in this study were volunteers

with relatively high motivation to participate in research with the incentive of the

supply of a magnifier on completion rather than regular patients in an out-patient

department. In addition, subjects reported to the experimenter the reasons if they had

not complied with the reading practice (which did not happen often), suggesting their

self-reports could be accepted.


168

4.3.8 Change of vision over time

Since all the recruited subjects had AMD, which is a progressive eye disease, it was

not surprising that visual acuity would deteriorate if the eye disease progressed. If

near visual acuity reduced significantly within the experimental duration (e.g., at

weeks 4 or 8) after the provision of the STM, a new magnifier with stronger

magnification was needed to achieve subjects’ target reading material. This was only

necessary for one subject (subject 30) whose distance visual acuity declined from

0.90 logMAR at week 0 to 1.14 logMAR at week 8. Woods and Lovie-Kitchin

(1995) compared the repeatability of distance visual acuity of subjects with low

vision. Their results suggested that a change of distance visual acuity (by high

contrast Bailey-Lovie visual acuity chart) which was less than 0.17 logMAR was

considered a repeatable measure. Therefore any reduction in distance visual acuity of

0.2 logMAR or more was selected as the criterion to determine subjects who had

significant vision deterioration. Because of the significant reduction in vision for

subject 30, the prescribed STM no longer enabled him to read his target print size

(N8). Therefore, a new STM with shorter EVD (stronger magnification) was

prescribed at this visit (week 8). Reading rates with both the old and new magnifiers

were measured. The reading rates across time were very similar.

4.4 Analysis


version 10 and GraphPad InStat (version 3). A probability of less than 0.05 was

taken to indicate statistical significance for all analyses. Although multiple

comparisons of log reading rates across time were conducted, Bonferroni

adjustments to the probability level for significance to reduce the chance of Type I

errors were not considered necessary as this was a designed experimental study and

the statistical analysis was planned to reflect this, not to develop a predictive model

(Rothman, 1990).


169

4.4.1 Reading assessments

In the following results, only reading rates for passages are reported because the aim

of this study was to investigate the effect of large print reading practice on reading

performance with stand magnifiers to read daily tasks. Since reading newspapers was

the major reading task for which subjects intended to use their low vision aids,

reading performance measured with passages was considered to be more relevant

than reading short sentences. Results for reading rate on sentences are given in

section 4.5.7.

4.4.1.1 Transformation of reading rate to log reading rate

There was a wide range of reading rates, from 25 wpm to 220 wpm (reading rate

without STM on sentences) between subjects. Therefore the reading rates were

transformed to log reading rates yielding frequency distributions which were not

significantly different from the normal distribution (Kolmogorov-Smirnov Goodness

of Fit test, p>0.1). The reading rates with or without STMs reported in this study are

all log reading rates. To compare log reading rates with and without STMs across

time, repeated measures analysis of variance (ANOVA) was performed. In

comparing log reading rates without STMs under full and restricted fields of view

(using the practice stand) together with log reading rate with STMs, one-way

ANOVA and post-hoc analyses were used.

4.4.1.2 Repeated measures of log reading rate

In this study, log reading rates for sentences and passages were analysed by repeated

measures ANOVA with subjects divided into four groups (control, large print

practice, stand and large print practice and clinical groups) due to different

interventions. The analysis of both between-subjects and within-subjects (repeated

measures) results were combined in one analysis provided that an additional

assumption - homogeneity of inter-correlations - was not violated. According to this

assumption, “for each level of the between-subjects variable, the pattern of inter-

correlations among the levels of the within-subjects variable is assumed to be the

same” (Pallant, 2001), p. 211). This implies that the change of the variables across

time (repeated measures) should not be different for subjects in different groups. If


170

the result of the analysis showed that the hypothesis of “the equality of covariance

matrices” across groups was not satisfied, repeated measures ANOVA were not

conducted or modifications of the analysis models were required (e.g. number of

levels of repeated measures were reduced from 6 levels to 5 in analysing the

experimental groups).

However, the problem with homogeneity of variance among groups only occurred

when repeated measures ANOVA were conducted for all subjects combined (i.e.

subjects in the experimental and clinical groups) in investigating the change of log

reading rate and interaction effect among groups across time. This was mainly due to

the deterioration of visual acuity at the last visit (week 20), introducing a

confounding variable affecting the log reading rates with STM across time among

groups. As such, the data from the last visit were not included, resulting in 3 levels

rather than 4 levels in the analysis model (repeated measures) to investigate the

change of log reading rate across time. Paired t-tests were used to compare the

reading measures at the last visit (week 20) and that at week 8 to identify any

statistically significant difference.

As a consequence of vision deterioration at the last visit (week 20), log reading rates

with and without STMs were reduced. To control the change of vision as a

confounding variable affecting log reading rate across time, repeated measures

analyses of covariance (ANCOVA) were conducted (Pallant, 2001). However, the

homogeneity of variance was violated when repeated measures ANCOVA was

performed for all subjects combined (i.e. subjects in the experimental and clinical

groups) in investigating the change of log reading rate across time. For this reason,

this analysis could not conducted to investigate the change in log reading rate across

time with vision as a covariate.

4.4.2 Vision measures

Threshold print size achieved with the STM (both word and text) at some visits were

significantly different from a normal distribution (Kolmogorov-Smirnov Goodness

of Fit test, p<0.05), which violated the assumption of normality for repeated

measures ANOVA and multiple regressions. For this reason, a logarithm


171

transformation of threshold print size (N-notation) with the STM was used for

analyses. As vision measures were not significantly different from a normal

distribution (Kolmogorov-Smirnov Goodness of Fit test, p>0.1), parametric statistics

– repeated measures ANOVA and Pearson's product moment correlation – were

used. Repeated measures ANOVA was used to compare vision measures (distance

and near visual acuities) across time so as to monitor any significant changes in

vision between experimental visits. Although vision measures for the majority of

subjects were reduced at the last visit (week 20) the reduction of vision did not

invalidate the assumption of the homogeneity of inter-correlations in the analysis

when repeated measures ANOVAs were conducted for all subjects combined. Unlike

repeated measures of log reading rate, vision measures at the last visit were included

in the analysis. Pearson’s correlation was used to investigate any correlation between

vision and reading measures.

4.4.3 Questionnaire

As the scales used in the questionnaires to classify the subjective responses made by

subjects were ordinal variables rather than continuous variables, repeated measures

on the responses of the frequency, duration and usefulness of the STM could not be

analysed by a non-parametric analysis (Friedman test). For this reason, the scales of

these variables were recoded into continuous variables by time-weighted factors of

the responses. This recoding system was determined based on an assumption that the

scale approximately reflected the actual meaning of the scales. Details of the

recoding of the responses are given in Appendix 5.

4.4.4 Correlation between variables

The relationships between log reading rate (with and without magnifiers) and vision

measures were also investigated by Pearson's correlation coefficients. In addition,

stepwise forward multiple regression analysis was performed to identify the potential

explanatory variables of log reading rate with and without STMs. The independent

variables for each subject were as follows: distance visual acuity, near visual acuity,

contrast sensitivity, log reading rate without STMs and visual field size of remaining

field. As the number of correlation tests (22) was less than the number of degrees of


172

freedom (42), correction for type I errors (the probability of finding a significant

difference by chance alone) was not considered necessary (Keppel, 1991). Statistical

significance per comparison was accepted at the 0.05 probability level.

4.5 Results

4.5.1 Effect of reading practice on log reading rate with STM over time

4.5.1.1 Comparison of baseline measures among groups

Table 4.7 shows the mean vision and reading measures at the visit when the STM

was supplied for home use for each of the four groups: control (N), large print

practice (P1), large print with reduced field of view practice (P2) and clinical groups

(C). The majority of vision and reading variables showed no significant differences

between groups (multivariate one-way ANOVA, Groups (4), p≥0.05), with the

exception of distance visual acuity (F3,39=3.59, p=0.02) and visual field loss in terms

of steradians (F3,39=3.48, p=0.03).

Mean distance visual acuity for subjects in the clinical group was significantly worse

than that in the control group by 0.31 logMAR (Bonferroni post hoc analysis,

p=0.04). In addition, the visual field defects in terms of steradians for subjects in the

clinical group were significantly larger than the P1 group using a 5-mm white target

(Bonferroni post hoc, p=0.045).

Although other differences in vision measures, such as word near visual acuity

(F3,39=1.72, p=0.18) and contrast sensitivity (F3,39=2.33, p=0.09) between groups

were not significantly different, some approached significance, indicating the vision

of the subjects in the clinical group was worse than vision of the subjects in the

experimental groups.


173

Table 4.7 Comparison of vision and reading measures for different groups at week 2 for all groups

EXPERIMENTAL GROUPS

PRACTICE GROUPS Variables CONTROL GROUP (N) P1 P2

CLINICAL GROUP (C) ANOVA SIGNIFICANCE

Number 10 11 11 11 F P

Age 80.80 ± 4.08 80.46 ± 4.50 79.64 ± 4.84 83.0 ± 5.44 1.00 0.40

Length of visual impairment (months) 14.90 ± 12.13 19.27 ± 20.18 18.18 ± 10.83 16.45 ± 9.61 0.20 0.89

Length of AMD (months) 38.70 ± 21.72 42.34 ± 51.57 24.09 ± 13.36 23.09 ± 7.71 1.26 0.30

Distance visual acuity (logMAR) 0.58 ± 0.21 0.76 ± 0.27 0.61 ± 0.29 0.89 ± 0.21 3.59 0.02

Word near visual acuity (logMAR) 0.70 ± 0.26 0.82 ± 0.26 0.78 ± 0.34 0.96 ± 0.21 1.72 0.18

Text near visual acuity (logMAR) 0.63 ± 0.22 0.74 ± 0.26 0.67 ± 0.35 0.91 ± 0.22 2.22 0.10

Critical print size (logMAR) 0.97 ± 0.28 1.05 ± 0.30 0.99 ± 0.35 1.20 ± 0.25 1.31 0.29 Threshold word reading print size achieved with

STM (log) 0.68 ± 0.12 0.72 ± 0.16 0.69 ± 0.09 0.75 ± 0.12 0.62 0.61

Threshold text reading print size achieved with STM (log) 0.57 ± 0.10 0.64 ± 0.16 0.59 ± 0.08 0.66 ± 0.10 1.41 0.20

Critical print size achieved with STM (log) 0.85 ± 0.09 0.86 ± 0.18 0.88 ± 0.13 0.93 ± 0.11 0.65 0.59

Contrast sensitivity (log) 1.15 ± 0.12 1.17 ± 0.15 1.22 ± 0.20 1.05 ± 0.10 2.33 0.09

Visual field loss steradians (5mm) 0.02 ± 0.03 0.02 ± 0.02 0.02 ± 0.04 0.06 ± 0.05 3.48 0.03

Visual field loss as % of sphere (5mm) 0.18 ± 0.26 0.11 ± 0.15 0.16 ± 0.30 0.49 ± 0.39 3.77 0.02 Log passage reading rate without STM at critical

print size (log wpm) 1.91 ± 0.23 1.88 ± 0.31 1.88 ± 0.33 1.82 ± 0.19 0.23 0.87

Log passage reading rate with STM at critical print size (log wpm) 1.88 ± 0.19 1.80 ± 0.26 1.82 ± 0.25 1.66 ± 0.20 1.71 0.18


174

4.5.1.2 Comparison of log reading rate with stand magnifier as a function of

time for experimental groups

Log reading rates with STM on passages at CPS improved significantly over time

from weeks 0 to 8 but decreased significantly at the last visit (week 20) (repeated

measures ANOVA, Groups (3) x Time (6); F5,25=4.46, p=0.005). Post hoc analysis

showed that the significant improvements in log reading rates with STM were found

at the first few visits (from weeks 0 to 2) (p<0.02) (Figure 4.6). There were no

significant differences in log reading rates between the three experimental groups

(F2,29=0.03, p=0.96) and the interaction effect between groups across time was not

significant (F10,50=1.13, p=0.36) with respect to the change in reading rate. This

indicates that the improvement in log reading rate with STM across time was similar

for each experimental groups.

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

Time (weeks)

Log

read

ing

rate

(wpm

)

Control group

Large print practicegroup

Large print withreduced field of viewpractice group

Clinical group

Figure 4.6

Log reading rate with STM on passages at CPS across time (log-scale). Error bars show one standard error of the meane.

Log reading rates for the experimental and clinical groups increased significantly across time at the initial visits, followed by a reduction at week 20. However, there was no significant difference in log reading rate across time between the experimental groups.

e For Figures 4.6 to 4.12, curves for each group were translated horizontally for easier discrimination between curves.

0 1 2 4 8 20

STM supplied Experimental groups p=0.005

Clinical group p=0.002


175

At the last visit (week 20), log reading rate was significantly reduced compared to

that at the previous visit (week 8) (post hoc analysis, p=0.043). Despite the non-

significant difference in the change in log reading rates between the experimental

groups (F2,29=2.14, p=0.14), there appears to be a trend towards a greater reduction in

log reading rate with STM in the control group than in the practice groups (P1 and

P2) (Figure 4.6).

While there was a tendency for the practice groups (P1 and P2) to show greater

improvement than the control group in log reading rate with STM over time, there

was no interaction effect between the experimental groups. The reason for the non-

significant effect may have been the small sample size for each group and the large

variation among subjects in the change in log reading rate across time (Figure 4.7).

The individual results illustrate that reading rate with STM for some subjects from

the P1 and P2 groups improved substantially at the initial visits (Figure 4.7). Subjects

24 and 31 from P1 and subjects 2 and 7 from P2 are examples of this. In contrast, the

magnifier reading rate for subjects from the control group revealed a gradual

increment in magnifier reading rate across time (Figure 4.7). This suggests that

reading practice is beneficial in improving log reading rate with STM when the STM

is first prescribed.

To account for the wide variation in baseline log reading rates and visual acuities

across individuals, relative log reading rate - the difference between log reading rate

at a particular visit and the log reading rate at week 0 was examined for the control

(N) and practice groups (P1 and P2). There was a significant improvement in log

reading rates from weeks 0 to 8 followed by a significant reduction at week 20

(Figure 4.8) (repeated measures ANOVA, Groups (3) x Time (5); F4,26=5.06,

p=0.004). Improvement in the log reading rate was significantly greater between

weeks 1 and 2 (post hoc analysis, p=0.009). Although there was no significant

interaction effect (F8,52=1.17, p=0.34), Figure 4.8 shows a trend that the relative

improvement in log reading rate with STM at weeks 0 to 8 for the practice groups

(P1 and P2) seems to be greater than that for the control group. Although Figure 4.8

shows a higher relative log reading rate in the practice groups (P1 and P2) than the

control group, the overall difference in relative log reading rate between groups was

not significant (F2,29=2.79, p=0.08).


176

Figure 4.7

Change in log reading rate across time for subjects from control (N), large print practice (P1), large print under reduced field of view practice (P2) and clinical groups (C).

As a consequence of substantial vision deterioration for Subject 22 in the control group, reading rate with STM could not be measured. Due to the scale limitation in the y-axis, reading rate for this subject at week 20 was not displayed in the graph.

Reading rate with STM (control)

1

1.2

1.4

1.6

1.8

2

2.2

2.4

STM-0 STM-1 STM-2 STM-4 STM-8 STM-20

Log

read

ing

rate

(wpm

)

13

14

16

17

18

19

20

21

22

23

Reading rate with STM (P1)

1

1.2

1.4

1.6

1.8

2

2.2

2.4


Log

read

ing

rate

(wpm

)

24

25

26

27

28

29

30

31

32

33

34


1

1.2

1.4

1.6

1.8

2

2.2

2.4


Log

read

ing

rate

(wpm

)

1

2

3

4

5

6

7

9

10

11

12

Reading rate with STM (clinical)

1

1.2

1.4

1.6

1.8

2

2.2

2.4

STM-2 STM-4 STM-8 STM-20

Log

read

ing

rate

(wpm

)

35

36

37

39

40

41

42

43

44

45

46


177

At the last visit, there was a significant reduction in log reading rate with STM (post

hoc analysis, p=0.048). The reduction in the control group appeared to be greater in

the control group than the practice groups (P1 and P2), however, the post hoc

analysis showed that it was not statistically significant (F2,29=2.12, p=0.14). The

likely explanation for the reduction in relative log reading rate at the last visit was the

deterioration of vision, which was a consequence of progressing AMD.

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

Time (weeks)

Cha

nge

in lo

g re

adin

g ra

te (w

pm)

Control group

Large printpractice group

Large print w ithreduced f ield ofview practicegroup

Figure 4.8

Change in log passage reading rate with STM as a function of time (log-scale) for the experimental groups. Error bars show one standard error of the mean.

The relative reading rate with STM significantly improved from weeks 0 to 8 but reduced at week 20. There is a tendency that the increase in log reading rate in the practice groups (P1 and P2) to be larger than that in the control group. However, the statistical analysis showed no significant difference in the change in log reading rate between groups.

STM supplied

0 1 2 4 8 20

p=0.004


178

Table 4.8 Summary of the mean vision measures across time for each group

TIME (Weeks) N GROUP 0 1 2 4 8 20

10 Control (N) 0.55 ± 0.17 0.60 ± 0.19 0.58 ± 0.21 0.61 ± 0.21 0.63 ± 0.25 0.86 ± 0.39

11 Large print practice (P1) 0.67 ± 0.24 0.72 ± 0.25 0.76 ± 0.27 0.75 ± 0.29 0.74 ± 0.30 0.74 ± 0.32

11 Large print with reduced field of view practice (P2) 0.59 ± 0.25 0.61 ± 0.28 0.60 ± 0.27 0.59 ± 0.29 0.60 ± 0.30 0.64 ± 0.32

Distance visual acuity (logMAR)

11 Clinicalf (C) 0.89 ± 0.21 0.89 ± 0.23 0.94 ± 0.23 1.04 ± 0.26

10 Control (N) 0.70 ± 0.24 0.73 ± 0.23 0.70 ± 0.26 0.71 ± 0.26 0.70 ± 0.28 0.92 ± 0.37



Near word visual acuity (logMAR)

11 Clinical (C) 0.96 ± 0.21 0.95 ± 0.21 0.97 ± 0.18 1.08 ± 0.21

10 Control (N) 0.68 ± 0.14 0.67 ± 0.15 0.68 ± 0.12 0.68 ± 0.10 0.68 ± 0.13 0.93 ± 0.34



Threshold print size (word) with STM (log n-

notation)

11 Clinical (C) 0.75 ± 0.12 0.65 ± 0.09 0.76 ± 0.14 0.87 ± 0.19

f For subjects in the clinical group, the first visit was equivalent to week 2 for subjects in the experimental groups as STM was prescribed at this visit for subjects in all groups.


179

4.5.1.2.1 Change in vision measures for experimental groups

Distance visual acuity, near word visual acuity, and threshold print size achieved

with the STM were selected as the parameters used to monitor the stability of vision

throughout the study (Table 4.8). Due to different numbers of visits included in the

experimental and clinical groups, the statistical analyses (repeated measures) were

separated.

There was significant difference in distance visual acuity across time (repeated

measures ANOVA; Groups (3) x time (6), F5,25=3.87, p=0.01) but there was no

significant interaction effect between the experimental groups (control and two

practice groups) (F10,50=1.87, p=0.07). However, there is a tendency of greater vision

deterioration for subjects in the control group compared with P1 and P2 groups at the

last visit (week 20) from that at week 8 (Figure 4.9).

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

Time (weeks)

Visu

al a

cuity

(log

MA

R)

Control group


Large print withreduced field ofview practicegroupClinical group

Figure 4.9

Distance visual acuity across time (log-scale). Error bars show one standard error of the mean.

Distance visual acuity was significantly reduced at the last visit (week 20). There was a trend towards a greater reduction in distance visual acuity for the control group than the other groups, but this was not significant.

0 1 2 4 8 20


Experimental groups p=0.01


180

For near visual acuity (word), no significant difference was found across time

(F5,25=1.52, p=0.22), however there was a significant interaction effect between

groups (F10,50=1.92, p=0.045). The post hoc analysis revealed that the significant

difference between groups occurred at the last visit (F2,29=3.68, p=0.038). Similar to

the change in distance visual acuity over time, there was a greater reduction in near

visual acuity in the control group than that in the practice groups (Figure 4.10)

Similarly, threshold print sizes (word) achieved with the STM did not change

significantly across time (F5,25=2.22, p=0.08). Similar to the near visual acuity,

Figure 4.11 shows that there was a trend towards a greater reduction in threshold

print size measured with STM in the control group than other groups. However the

difference in the change in threshold print sizes with STM across time was not

significant between groups (F10,50=1.32, p=0.25 ).

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

Time (weeks)

Visu

al a

cuity

(log

MA

R) Control group



Figure 4.10

Near (word) visual acuity across time (log-scale). Error bars show one standard error of the mean.

Near visual acuity was not significantly changed across time. However, there is a trend towards a reduction in near visual acuity, especially for the control group, at the last visit (week 20).



0 1 2 4 8 20


181

0.6

0.7

0.8

0.9

1

1.1

Time (weeks)

Thre

shol

d pr

int s

ize

with

STM

(lo

g n-

nota

tion)

Control group



Figure 4.11

Word threshold print size achieved with STM across time (log-scale). Error bars show one standard error of the mean.

Word threshold print size achieved with STM was not significantly reduced at the last visit. Among the experimental groups, the reduction in the control group seemed to be greater than the other groups.

4.5.1.2.2 Change in log reading rate with stand magnifier controlled for change in

visual acuity

Previous studies have shown that reading rate is highly correlated with visual acuities

(Goodrich et al., 1977; Legge et al., 1985 (b); McMahon et al., 1991; Leat and

Woodhouse, 1993; Bullimore and Bailey, 1995; Bowers, 1998 (a); Lovie-Kitchin et

al., 2000 (a)) (refer to Table 1.5). Therefore, it is reasonable to argue that the

significant reduction in log reading rate with STM at the last visit could be due to the

deterioration in vision since distance visual acuity was significantly reduced at the

last visit. In order to control for the reduction in vision which would affect the log

reading rate with STM, analysis of covariance (ANCOVA) was conducted. Among

the three vision measures monitored (distance and near visual acuities, threshold

print size with STM), change in distance visual acuity (i.e. the slope) was chosen as

the covariate in the analysis model since it is the commonly used standard for

monitoring change in vision (Lovie-Kitchin, 1993; Enoch, 1998). Log reading rate

with STM for the experimental groups improved from weeks 0 to 4 and then became



0 1 2 4 8 20


182

stable from weeks 4 to 20 (repeated measures ANCOVA, Groups (3) x Time (6);

F5,24=3.01, p=0.03). The post hoc analysis showed that the significant increase in log

reading rate occurred from weeks 0 to 2 (F1,28=9.88, p=0.004). However, the change

in log reading rate was not significantly different among groups (F10,48=0.88, p=0.56

for interaction effect across time; F2,28=0.85, p=0.44 for log reading rate between

groups).

Another way of examining the change in log reading rate with STM without vision

change affecting results was to exclude subjects from analysis whose distance visual

acuity at the last visit decreased by 0.2 logMAR or more compared to the visual

acuity at week 2 (refer to section 4.3.8). This analysis provided further supportive

evidence that log reading rate with STM was stable after week 4 provided that vision

was stable. Six subjects in the control group, ten subjects in the large print practice

group, nine subjects in the stand and large print practice group were included in this

analysis. Log reading rate with STM significantly changed across time (repeated

measures ANOVA, Groups (3) x Time (6); F5,18=3.74, p=0.02) with a significant

improvement between weeks 1 and 2 (post hoc analysis, p=0.04) prior to the

provision of STM for home use, followed by a gradual improvement until week 4.

From weeks 4 onwards, log reading rate with STM had reached a plateau (p>0.11)

(Figure 4.12). However, the significant improvement in reading rate with STM was

not significantly different between the experimental groups (F10,36=1.25, p=0.29).

4.5.1.3 Comparison of log reading rate with STM and vision measures as a

function of time for clinical group

Log reading rate with STM on passages for the complete clinical group showed

significantly changes over time (repeated measures ANOVA, Group (1) x Time (4);

F3,8=12.04, p=0.002). Post hoc analysis showed that the improvement in log reading

rate was significant between weeks 2 and 4 (p=0.002) but it became stable from

weeks 4 to 20. While the change in log reading rate between weeks 4 and 20 was not

significant, there was a tendency for log reading rate to reduce at the last visit (Figure

4.6).


183

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)

Log

read

ing

rate

(wpm

)Control group



Clinical group

Figure 4.12

Log reading rate with STM across time (log scale) excluding subjects whose distance visual acuity reduced by 2 lines or more. Error bars show one standard error of the mean.

Log reading rate with STM improved at the first few visits and then was stable from weeks 4 to 20. The improvement in magnifier reading rate was not significantly different among the experimental groups. However, significant improvement in log reading rate in the clinical group was found in the clinical group, which was significantly greater than that in the experimental groups between weeks 2 and 4.

For the clinical group, all vision measures except near visual acuity (repeated

measures ANOVA; Group (1) x time (4); F3,8=3.38, p=0.08) changed significantly

different across time (F3,8=6.61, p=0.02 for distance visual acuity; F3,8=13.6, p=0.002

for word threshold print size with STM). Post hoc analysis showed that the

significant reduction of these vision variables was found at the last visit (p<0.05)

(Figures 4.9 and 4.11).

As subjects’ vision was significantly reduced at the last visit, the comparison of the

log reading rate with STM was re-analysed including the change in distance visual

acuity (i.e. the slope) as the covariate. Log reading rate with STM for the clinical

groups improved significantly from weeks 2 to 4 and then became stable from weeks

4 to 20 (repeated measures ANCOVA, Groups (1) x Time (4); F3,7=9.22, p=0.008).

The post hoc analysis showed that the significant increase in log reading rate

occurred from weeks 2 to 4 (F1,9=8.56, p=0.002).



0 1 2 4 8 20


184

4.5.1.4 Comparison of log reading rate with stand magnifier and vision

measures as a function of time for all groups

The reason for recruiting subjects into the clinical group was to compare the reading

performance of subjects with a STM supplied by the low vision clinics with that of

subjects in the experimental groups who were prescribed a STM by the

experimenterg. The aim was to investigate whether the experimental interventions

(such as the repeated measures of log reading rate prior to the provision of a STM)

had any impact on the log reading rate with STM across time. As the number of

visits for different groups varied due to the design of the experiment, analysis of the

log reading rate with STM was only conducted for the common visits with group as

an independent between-subjects variable. Therefore, repeated measures were

examined for the visits when the magnifier was supplied for home use and the

subsequent follow up visits (weeks 2 to 20). The differences in initial log reading

rates and visual acuity among groups are discussed in this section.

4.5.1.4.1 Change of vision measures over time for all subjects

In general, most of the vision variables (distance and near visual acuities and

threshold print size with STM) were reduced at the last visit for subjects in all groups

(repeated measures ANOVA; Groups (4) x time (4), F9,31=2.42, p=0.03) h . The

reduction in vision with and without STM at the last visit in the control group were

significantly greater than that in the two practice groups and clinical group at the last

visit as a significant interaction effect was found (F27,99=2.5, p=0.04) (Figures 4.9 to

4.11).

4.5.1.4.2 Change of log reading rate with stand magnifier over time for all subjects

As the majority of the subjects' vision among the three groups decreased at the last

visit, analysis of log reading rate with STM across all subjects and all visits could not

be performed due to the violation of the homogeneity of variance (refer to section

g The STM prescribed by the low vision clinics were all illuminated STMs and the fixed acuity reserve method was used to calculate the tentative magnification. However, STMs prescribed for subjects in the clinical group might be different from the calculated magnification as they were allowed to try STMs with different EVDs. h Distance and near visual acuities together with vision with STM were included in the repeated measure analysis model. The results are reported for the multi-ANOVA statistic.


185

4.4.1.2). This problem of changes in visual acuity was dealt with in two ways: 1. data

for week 20 were excluded; 2. data for subjects who had a significant reduction in

distance visual acuity were excluded. Log reading rate with STM improved across

time (weeks 2 to 8) for reading passages at CPS (repeated measures ANOVA,

Groups (4) x Time (3); F2,38=7.58, p=0.002). The improvement in log reading rate

was significantly greater between weeks 2 and 4 (F1,39=15.03, p<0.001) than weeks 4

to 8. However a significant interaction effect across time between groups (F6,76=2.94,

p=0.012) was found, which indicated that the increment in log reading rate in the

clinical group was greater than the other groups (Figure 4.6). Despite deterioration of

vision, the change in log reading rate with STM between weeks 8 and 20 did not

quite reach statistical significance (paired t-tests, t=1.85, df=42, p=0.07).

In the experimental groups, provided that there was no significant change in vision at

the last visit, log reading rate with STM was stable compared to the log reading rate

at week 8. Instead, subjects who had a reduction in distance visual acuity by 0.2

logMAR or more at the last visit (week 20) were excluded from the analysis. In

addition to the 25 experimental subjects included (section 4.5.1.2.2), seven subjects

in the clinical group were included in this analysis model. Log reading rate with

STM significantly improved across time (repeated measures ANOVA, Groups (4) x

Time (4); F3,26=2.83, p=0.04). Post-hoc analysis found that the significant increase of

log reading rate occurred at week 4 (F1,28=7.77, p=0.009) which was 2 weeks after

the STM was supplied for home use. From week 4, the change in log reading rate

with STM was not significant (p=0.6). The change in log reading rate with STM

across time was not significantly different among different groups (F9,63=0.81,

p=0.61).

4.5.2 Change in log reading rate without stand magnifier across time

The changes in log reading rate (for large print) without STM mirrored the vision

changes over time (Figure 4.13). Prior to the prescription of STM (weeks 0 to 2),

there was no significant difference in log reading rates without STM over time for

passage reading in the control and practice groups (repeated measures ANOVA,

Groups (3) x Time (3); F2,289=1.98, p=0.16). Four months after the STM was

prescribed, the log reading rate without STM at the last visit (week 20) was slower


186

than that at week 2, but the reduction did not quite reach statistical significance

(paired t-test, t=1.97, df=31, p=0.06).

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

Time (weeks)

Log

read

ing

rate

(wpm

)

Control group



Clinical group

Figure 4.13

Log reading rate without STM across time (log-scale). Error bars show one standard error of the mean.

Log reading rate without STM was not significantly different for subjects in the control and practice groups (P1 and P2) before the STM prescribed (from weeks 0 to 2). However, it reduced at the last visit but the reduction was not statistically significant. Subjects in the clinical group showed a similar reduction in log reading rate at the last visit, which was statistically significant.

For the clinical group, log reading rate without STM at the last visit (week 20)

decreased by 0.17 log wpm compared to the log reading rate at the first visit

(equivalent to the visit at week 2 for the experimental groups). This reduction in log

reading rate was statistically significant (paired t-test, t=3.13, df=10, p=0.011).

Figure 4.13 shows the log reading rate without STM for each group across time,

which indicates a reduction in log reading rate at week 20 for all groups. The

reduction for the control group appears greater than for the other groups, however no

statistical analysis (repeated measures) could be conducted to examine this due to the

violation of the assumption of homogeneity of variance.

0 1 2 20

Experimental groups (weeks 2 and 20) p=0.06



187

Ideally, distance visual acuity should be controlled as a confounding variable in

assessing the change in log reading rate without STM across time. However,

repeated measures ANCOVA could not be used due to the violation of the

assumption of homogeneity of variance (refer to section 4.4.1.2). Instead, when

subjects whose distance visual acuity had reduced by 0.2 logMAR or more were

excluded from the analysis, log reading rate without STM at week 2 (before the STM

was prescribed) was not significantly different from log reading rate at week 20

among all the subjects (paired t-test, t=1.74, df=31, p=0.09).

4.5.3 Comparing log reading rate with and without stand magnifier

Table 4.9 shows the mean log reading rates with and without STM across time for

each group. For the experimental groups (control and two practice groups), the STM

did not significantly reduce the log reading rate across time (repeated measures

ANOVA, reading conditions i (2) x reading rates (6) x Groups (3), F5,25=2.40,

p=0.07). However, comparison between visits showed that the log reading rate with

STM was significantly reduced compared with the log reading rate without STM at

the first visit, week 0 (F2,29=3.51, p=0.04). Log reading rates with and without STM

were not significantly different among the experimental groups across visits from

week 1 (F10,50≥1.45, p≥0.19). For subjects in the clinical group, the use of STM

significantly reduced log reading rate compared with log reading rate on large print

across time (F3,8=6.3, p=0.02). Post hoc analysis showed that the significant

difference in reading rate with and without STM occurred at week 2 (which was the

first visit for subjects in the clinical group). From week 4 onwards, the STM did not

have any significant effect on log reading rate.

i The reading conditions included were reading rate with and without STMs. In order to minimize the effect of fatigue on reading performance, reading rates without STM were not assessed at weeks 4 and 8. However, for analyses purpose (repeated measures ANOVA), reading rate without STM was assumed to be stable and log reading rate at week 2 was used to be the reading rates without STM for weeks 4 and 8.


188

Table 4.9 Summary of the measures of mean log reading rate (log wpm) across time for each group

TIME (Weeks)

Log Reading rate

(wpm) N GROUP 0 1 2 4 8 20

10 Control (N) 1.91 ± 0.29 1.88 ± 0.30 1.91 ± 0.22 1.57 ± 0.69

11 Large print practice (P1) 1.85 ± 0.36 1.88 ± 0.31 1.89 ± 0.31 1.85 ± 0.31

11 Large print with reduced field of

view practice (P2) 1.83 ± 0.34 1.87 ± 0.34 1.88 ± 0.32 1.79 ± 0.37

Without STM

11 Clinical (C) 1.82 ± 0.19 1.65 ± 0.21

10 Control (N) 1.85 ± 0.26 1.86 ± 0.26 1.88 ± 0.19 1.87 ± 0.17 1.90 ± 0.19 1.60 ± 0.65



view practice (P2) 1.73 ± 0.33 1.78 ± 0.3 1.82 ± 0.25 1.84 ± 0.27 1.86 ± 0.27 1.76 ± 0.35

With STM

11 Clinical (C) 1.66 ± 0.20 1.77 ± 0.15 1.74 ± 0.18 1.72 ± 0.16

Without STM

(restricted field of view)


view practice (P2) 1.58 ± 0.34 1.65 ± 0.36 1.69 ± 0.33 1.62 ± 0.38


189

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)

Log

read

ing

rate

(wpm

)

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)

Log

read

ing

rate

(wpm

)

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)

Log

read

ing

rate

(wpm

)

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)

Log

read

ing

rate

(wpm

)

These results indicate that for subjects with no experience in using magnifiers, log

reading rate with STM was significantly lower than without STM at the first visit

(week 0 for experimental groups and week 2 for clinical group) before any training

on large print reading or the use of STM (Figure 4.14). However, this reduction in

log reading rate in the control group appeared to be less than that in the practice (P1

and P2) and clinical groups at week 0.

Large print with reduced field of view Clinical group (C) Practice group (P2)

Figure 4.14

Comparison of log reading rates with (solid line) and without STM (dashed line) across time (log-scale) for each group. Error bars show one standard error of the mean.

0 1 2 4 8 20 0 1 2 4 8 20

0 1 2 4 8 20 2 4 8 20

Control group (N) Large print practice group (P1)


190

STMs were prescribed by the experimenter for the experimental groups at week 2 or

by the clinician for subjects in the clinical group. All subjects were encouraged to use

the STM for reading their target reading materials at home. As time went on, the log

reading rate with STM increased although the extent of improvement varied among

groups. The difference in log reading rate with and without STM became less across

time. For subjects in the experimental groups (control and two practice groups), log

reading rate with STM was not significantly slower than log reading rate without

STM at week 1 (p≥0.18). For subjects in the clinical group, STM did not

significantly affect the log reading rate on large print reading after 2 weeks home

practice on using STM for reading (p>0.2). Interestingly, there was a trend that the

log reading rate with STM was slightly faster than the log reading rate without STM

for subjects in the clinical group at week 20 when most subjects had a significant

reduction in vision and reading performance. However, this difference was not

statistically significant.

4.5.4 Relationship between near visual acuity and EVD (magnification)

prescribed

The required EVD (calculated magnification) in this study was based on the text

visual acuity and target print size of each subject (fixed acuity reserve method).

When the STM was selected for subjects to conduct the repeated laboratory measures

at weeks 0 to 2, eye-lens distance was recommended to subjects from the control and

practice groups to achieve the required magnification. However, no instructions

regarding the eye-lens distance or ways to manipulate the magnifiers were given by

the experimenter to the subjects from the clinical group as they received these

guidelines when they attended the LVC. Based on the eye-lens distance and the

equivalent power of the magnifier, the EVD at all subsequent follow up visits (weeks

4 to 20) was calculated. Analysis showed that the EVD achieved by the subjects was

not significantly different from the EVD of the STM prescribed at week 2 (repeated

measures ANOVA, Groups (4) x Time (4); F3,26=2.0, p=0.14). This indicated that the

working distance and the magnification achieved with the STM were consistent

across time. According to the design of the STM, the shorter the distance between

the eye and lens, the higher the magnification, without considering the

accommodation demand due to a shorter image distance. Subjects in this study did


191

not change their habitual working distance for the sake of increasing the magnifying

effect despite vision deterioration at the last visit. This was probably because subjects

had adapted to the working distance between the lens and eye for clear magnified

images.

4.5.5 Log reading rate with a restricted field of view (P2)

Subjects in P2 were given large print reading practice under a restricted field of view

(approximately 6 characters horizontally). In order to determine whether this reading

practice improved reading rate with a restricted field of view, additional reading

assessment on large print under reduced field of view by using a practice stand was

assessed. This extra reading assessment was conducted at each visit prior to the STM

being prescribed (weeks 0 to 2) as well as at the last visit (week 20) for P2 subjects

(Table 4.9). The log reading rate under a restricted field of view changed

significantly across time (repeated measures ANOVA, Group (1) x Time (4);

F3,8=6.57, p=0.015) with a significant increment between weeks 1 and 2 (F1,10=6.38,

p=0.03). At the last visit (week 20), the log reading rate was not significantly

different from the log reading rate at week 2 (F1,20=2.5, p=0.14) even though vision

had deteriorated.

4.5.6 Log reading rate with and without a restricted field of view compared with

log reading rate with STM

There were no significant differences in log reading rate without STM under normal

and restricted field of view compared with log reading rate with STM for all subjects

at the last visit (one-way ANOVA, Groups (4), F2,125=1.91, p=0.15). The log reading

rates measured under the three reading conditions were re-analysed for each group to

address any differences in the result due to the grouping variable. Log reading rates

under normal and restricted field of view and the log reading rate with STM were not

significantly different for subjects in either the control (F2,26=0.007, p=0.99) or

practice groups (F2,30=0.88, p=0.43 for P1, F2,30=0.66, p=0.52 for P2). However, log

reading rate without STM under a restricted field of view was significantly slower

than that under full field and log reading rate with STM for subjects in the clinical


192

group (F2,30=5.9, p=0.007). Figure 4.15 shows the log reading rates measured under

each reading condition.

0

0.5

1

1.5

2

All N P1 P2 C

Log

read

ing

rate

(wpm

) Reading ratew ithout STM(full f ield)

Reading ratew ith STM

Reading ratew ithout STM(restrictedfield ofview )

Figure 4.15

Log reading rate at week 20 for different reading assessment at CPS. Error bars show one standard error of the mean.

For the control (N) and two practice groups (P1 and P2), log reading rates were not significantly different in each of the reading conditions (reading under full and restricted field of view and reading with STMs). However, for the clinical group (C), log reading rate without STM under a restricted field of view was significantly slower than log reading rates in other conditions.

4.5.7 Reading performance with different reading materials

Reading newspapers or mail was the major goal for subjects with visual impairment

attending the LVC (refer to Figure 4.18b). Reading performance on passages of text

was primarily examined in this study. However, Harper (1999) reported that a large

percentage of people used their magnifiers to read labels on packets and medications

(60%) or telephone books (36%) in addition to reading ordinary print books or

newspapers (57%). According to the three categories of reading requirements

classified by Whittaker and Lovie-Kitchin (1993) (refer to Table 1.4), reading labels

or the telephone book refers to short-term spot-reading, and these are important

reading activities of daily living, for survival. For this reason, measuring reading

p=0.18 p=0.99 p=0.007 p=0.43 p=0.52


193

performance for short sentences was also important to reflect subjects’ reading

performance with their STMs. In this study, sentences on Bailey-Lovie text reading

charts were used, in addition to passage reading, to assess reading performance with

and without STM across time.

Log reading ratej for sentences was significantly faster than that for passages with

STM (paired t-test, t=5.23, df=42, p<0.01) or without STM (t=8.95, df=42, p<0.01).

However, reading rates for these two reading tasks were highly correlated (r=0.89,

p<0.001 for reading without STM, r=0.90, p<0.001 for reading with STM).

The levels of readability of the sentences (Bailey-Lovie text reading chart) and

passages were below 6th grade reading level, but the length of the reading task might

result in different reading performances across time. Table 4.10 summarises the

reading performance with and without STM comparing passages and sentences. In

general, reading performance with and without STM using passages and sentences

were very similar although the change in the reading rate across time differed

between reading materials. The main differences in reading performance with

passages and sentences are described below.

Log reading rate with STM improved across time for the experimental groups with

sentences (repeated measures ANOVA, Groups (3) x Time (6); F5,25=6.25, p=0.001).

Post hoc analysis showed that the significant increase in log reading rates with STM

using sentences (Bailey-Lovie text reading charts) was found from weeks 1 to 8,

which was relatively longer than that with passages (from weeks 0 to 2) (Table

4.10a). This implies that the reading rate for a short-term reading task continues to

improve with practice using the STM at home. In addition, subjects in P1 and P2

showed greater improvement in log reading rate with sentences between weeks 1 to 2

compared with subjects in the control group (Figure 4.16), which was statistically

significant (F10,50=2.24, p=0.03). In contrast, no significant interaction effect between

groups was found when passages were used.

j Week 2 reading rates with passage and sentences were analysed.


194

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)

Log

read

ing

rate

(wpm

)

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)Lo

g re

adin

g ra

te (w

pm)

Control group (N) Large print practice group (P1)

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)

Log

read

ing

ra

te (w

pm)

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

Time (weeks)

Log

read

ing

rate

(wpm

)

Large print with reduced field of view Clinical group (C) Practice group (P2)

Figure 4.16

Comparison of log reading rates with STM for sentences (represented in solid line) and passages (represented in dash line) across time (log-scale) for each group. Error bars show one standard error of the meank.

The patterns of the change of log reading rate with STM for sentences and passages were similar for subjects in control, large print practice, large print with reduced field of view practice and clinical groups.

k The dashed curves for each group were translated horizontally for easier discrimination.

0 1 2 4 8 20 0 1 2 4 8 20

0 1 2 4 8 20 2 4 8 20


195

Table 4.10a Comparison of log reading rates with STM using sentences and passages

Groups Period Analysis Passage reading p-value Sentence reading p-value

Significant improvement from weeks 0 to 2 but reduced at week 20

0.005 Significant improvement from weeks 1 to 8 but reduced at week 20

0.001

Weeks 0 to 20 Repeated measures ANOVA (Groups (3) x Time (6)) No significant interaction

effect 0.36 Significant interaction effect (Improvement of RR in P1 and P2 > control group)

0.03

Significant improvement from weeks 0 to 2 0.01 Significant improvement

from weeks 1 to 8 0.006

Experimental

Weeks 0 to 8

Repeated measures ANOVA (Groups (3) x Time (5)) No significant interaction

effect 0.37 Significant interaction effect (Improvement of RR in P1 and P2 > control group)

0.03

Clinical Weeks 2 to 20 Repeated measures ANOVA (Group (1) x Time (4))

Significant improvement from week 2 to 4 0.002 Significant improvement

from week 2 to 4 0.006

Significant improvement from week 2 to 4 0.002 Significant improvement

from week 2 to 4 0.03

Weeks 2 to 8 Repeated measures ANOVA (Group (4) x Time (3))

Significant interaction effect (Improvement of RR in clinical > experimental groups)

0.012

Significant interaction effect (Improvement of RR in clinical > experimental groups)

0.001 All

Weeks 8 and 20 Paired t-test No significant change across

time 0.07 Significant reduction at week 20 0.01

Experimental groups include control (N), large print practice (P1) and large print with reduced field of view practice (P2) groups. All groups include control (N), large print practice (P1), large print with reduced field of view practice (P2) and clinical (C) groups. p-value = significance level RR = reading rate


196

Table 4.10b Comparison of log reading rates without STM using sentences and passages


Experimental

Prior to the STM

prescribed (weeks 0 – 2)

Repeated measures ANOVA (Groups (3) x Time (3))

No significant change across time 0.16 No significant change

across time 0.061

Experimental Paired t-test No significant reduction 0.06 Significant reduction at week 20 0.03

Clinical Paired t-test Significant reduction 0.01 Significant reduction at week 20 0.045

All

After 4 months (weeks 8 and

20) Paired t-test Significant reduction 0.01 Significant reduction

at week 20 0.009

Table 4.10c Comparison of log reading rates with and without STM using sentences and passages


Significant difference in RR with and without STM at week 0

0.04

Significant difference in RR with and without STM at weeks 0 to 2

0.03 Experimental Weeks 0 to 20

Repeated measures ANOVA (Reading (2) x Groups (3) x Time (6)) No significant

interaction effect 0.19 No significant interaction effect 0.27

Clinical Weeks 2 to 20

Repeated measures ANOVA (Reading (2) x Group (1) x Time (4))

Significant difference in RR with and without STM at week 2

0.02

Significant difference in RR with and without STM at weeks 2 and 4

0.02


197

For reading performance without STM on large print, subjects in the experimental

groups showed a significant reduction in log reading rate at the last visit (week 20)

with sentences (p=0.03). Although the reduction in log reading rate at the last visit

for passages (by 0.15 log wpm) was slightly less than that for sentences (by 0.17 log

wpm), the reduction in log reading rate with passages was close to significance

(Table 4.10b).

In comparing log reading rates with and without STM, the duration for the log STM

reading rate to reach maximum (reading rate without STM) was longer for sentences

compared to passages (Table 4.10c). Log reading rate for sentences with STM for

subjects in the experimental groups was not significantly different from log reading

rate on large print from week 4 onwards, which was two weeks after home practice

with STM. Similarly, subjects in the clinical group required more than 2 weeks home

practice using STM to reach the point that the log STM reading rate was equivalent

to log reading rate on large print which occurred by week 8.

4.5.8 Results of questionnaire and subjective report

There are a number of low vision aids which can be used by people to assist reading.

Among these available vision aids, near spectacles (53.5%) and high additions

(addition of more than +3 D) (25.58%) were most commonly used by the participants

in this study before they received the STM as their new assistive reading aids. In

addition, 65% of the subjects preferred to read under either sunlight or strong

illumination (e.g. reading lamps) as they found the extra illumination provided them

with better reading vision.

4.5.8.1 Frequency and duration of reading

Before the STM was prescribed (week 2), 60.5% of all subjects read more than once

per day while 20.9% and 9.3% of the subjects read regularly (once per day) and

sometimes (2-3 times per week) respectively. After they had been given the STM for

home reading, subjects read more frequently such that a higher percentage of

subjects (>81%) read “more than once per day” at weeks 4 and 8 (Figure 4.17a).

Vision deterioration due to the progression of AMD after 20 weeks probably


198

explains the significant reduction in the proportion of people who read frequently

from 81% to 65% (non-parametric Friedman test, α2=60.47, df=3, p<0.001). In

addition, 4.7% of people reported that they did not read, which again was probably

due to the poor vision.

Duration of reading with and without magnifiers varied within the study. Before the

STM was prescribed (week 2), 65.1% of subjects read for 10 minutes to more than

one hour with their own low vision aids (Figure 4.17b). As time went on, the

duration of reading without STM was significantly shorter (non-parametric Friedman

test, α2=26.47, df=3, p<0.001). At week 20, 38% of people still read for 10 minutes

to more than one hour without their STM while more than 40% people chose to read

‘rarely’ when no STM was used.

0

10

20

30

40

50

60

70

80

90

100

Week 2 Week 4 Week 8 Week 20

Perc

ent o

f sub

ject

s (%

)

>1 per day

1 per day

2-3 times perw eekRarely

Never

Figure 4.17 a

Frequency of reading across time.

Participants read significantly more frequently after the provision of a STM to assist reading. However, the frequency of reading reduced significantly at week 20, which could possibly be due to the reduction of vision.


199

0

5

10

15

20

25

30

35

40

45

50

Week 2 Week 4 Week 8 Week 20

Perc

ent o

f sub

ject

s (%

)

> 1 hour

10 mins - 1hour< 10 mins

Sometimes

Rarely

Figure 4.17 b

Duration of reading without STM across time.

After the provision of stand magnifier, the duration of reading without STM was significantly shorter. This was even more severe at week 20 when more subjects preferred to read ‘rarely’ without STM.

0

5

10

15

20

25

30

35

40

45

50

Week 4 Week 8 Week 20

Perc

ent o

f sub

ject

s (%

)

> 1 hour

10 mins - 1hours< 10 mins

Sometimes

Rarely

Figure 4.17 c

Duration of reading with STM across time.

The majority of the subjects preferred to use their STMs for 10 to 60 minutes. At week 20, the reading duration with STM was significantly reduced compared with that at previous visits. This could be because of the deterioration of vision, when the STM could no longer provide sufficient magnification to achieve fluent reading of target reading materials.


200

After the subjects had been prescribed the STMs as their primary low vision aids for

reading at home (week 2), more than 62% people read for 10 minutes to more than

one hour continuously with STM (Figure 4.17c). However, presumably due to the

vision deterioration at the last visit, the length of time that people read with their

STM was significantly diminished (non-parametric Friedman test, α2=18.08, df=2,

p<0.001). At week 20, only 44.2% of the participants used their STM for longer than

10 minutes while 48.8% used it sometimes but for less than 10 minutes.

4.5.8.2 Reading materials

Before they were given the STM to assist their reading, they usually read large print

books (30%), newspaper headlines (20%) or magazines (15%) with their own low

vision aids (Figure 4.18a). Figure 4.18b shows the goal reading materials that the

subjects hoped to read. The majority of the participants aimed to read newspaper

(59%) and mail (14%) with their STM.

In general, subjects did use their STM for reading newspaper (52%), mail (16%),

normal print books (12%) or magazines (9%) (Figure 4.18c). Even though a large

proportion of people (>95%) were capable of reading passages of 8 point print

(newspaper print size) or smaller in the experimental visits (Table 4.11), only 52%

and 12% of the participants reported that they used their STM for reading newspaper

or normal print books at home. This suggests that the near visual acuity measured in

the laboratory or clinic may not always reflect what reading materials subjects

usually read with their STM at home. The discrepancy between the print size of the

materials read at home and the print size achieved with STM in laboratory was less at

week 20 compared to week 4 or 8. Seventy-seven percent of participants could

achieve threshold print size of N8 or better with STM but 58% of them reported they

read newspaper or normal print books.


201

Mail12%

News headlines

20%

News18%

Magazines15%

Normal print books

5%

Large print books30%

Figure 4.18a

Reading material without STM read by subjects.

Before subjects were given the STM to assist their reading, they usually read large print books, newspaper headlines or magazines with their own low vision aids.

News59%

Bible2%

Magazines11%

Telephone books

3%

Large Print books

5%

Mail14%

Normal print book6%

Figure 4.18b

Goal reading material defined by subjects.

The majority of participants intended to read newspaper and mail when they received their STM as their primary low vision aids.


202

Mail16%

Magazines9%

Normal print books12%

Large print books

2%

News headlines

9%

News52%

Figure 4.18c

Primary reading material with STM at weeks 4 and 8.

Most of the subjects used their STM to read newspaper, mail, normal print books or magazines.

Table 4.11 Percentage of subjects whose threshold print size with STM was N8 or better at different visits

Percent subjects (%)

≤ N8 point print Week 4 Week 8 Week 20

Word threshold print size with STM 100 % 95 % 76.7 %

Text threshold print size with STM 100 % 100 % 83.7 %

4.5.8.3 Subjective response on the effectiveness of STM

Threshold print size and log reading rate with the prescribed STM were the objective

measures of the reading performance with magnifiers. However, these measures

might not reflect how effectively the STM helped the AMD subjects with reading. In

this study, more than 40% and 30% of subjects rated the effectiveness of STM as

'excellent and ‘good’ respectively in assisting them to read their target reading

materials (Figure 4.19). Satisfaction with the STM was significantly correlated with

subjects’ measured reading rate and use of STM (refer to section 4.5.8.4). The

majority of subjects reported that the STM could increase their reading duration

(26.9%) and allow them to read smaller print (23.1%). In addition, the use of a


203

reading stand was highly preferred by the subjects when they read with the STM, as

it could provide a better posture for long duration of reading. A minority of the

subjects reported that the benefit received from the STM was 'average' or 'little'. Only

three subjects whose vision had dropped significantly at the last visit considered the

STM to be ‘useless’ as the magnification provided by the STM was no longer

sufficient. However, as the questionnaire was conducted verbally by the

experimenter, the reported usefulness of the STM might not reflect the actual

usefulness of the STM in reading because of the relationships established between

the experimenter and subjects over the course of the study. Subjects’ reports of the

effectiveness of STM was significantly reduced at week 20 compared to that at

previous visits (weeks 4 and 8) (Friedman test, α2=11.3, df=2, p=0.004), probably

because of the vision deterioration.

05

1015

202530

3540

4550

Week 4 Week 8 Week 20

Perc

ent o

f sub

ject

s (%

)

ExcellentGoodAverageLittleUseless

Figure 4.19

Usefulness of STM across time reported by subjects.

The majority of subjects reported that STM was excellent or good for reading. Only three subjects reported that the STM was ‘useless’. This might be because of the significant deterioration in vision at week 20 when the magnification provided by the STM was no longer sufficient.

Subjects reported the limitations of the STM in response to an open-ended question

about the effectiveness of their STMs (Figure 4.20). A large proportion of these

subjects found that the STM was hard to manipulate for reading (28.2%) and the

internal illumination of the STM was not stable or was insufficient (25.6%). These


204

qualitative results suggest that there are some problems that subjects might encounter

during use of their STM.

No reading stand21%

Prefer to read large print

books3%

Others8%

Problem with lighting (flickers

/insufficient)25.6%

Uncomfor-table to use

15.4%

Difficult to use27%

Figure 4.20

Limitations of the STM reported by the subjects.

The major limitations of the STM reported by the subjects were the difficulty manipulating the magnifier and the unstable and insufficient internal illumination provided by the STM.

4.5.8.4 Correlation between reading variables and the reported use of STM

There were significant correlations between the perceived effectiveness of STM,

frequency of reading (Spearman’s correlation, rs=0.38, p=0.01) and duration of STM

use (rs=0.32, p=0.04), which were the variables derived from the questionnaire.

However, these variables were not strongly correlated. In addition, there was no

significant correlation between frequency of reading and duration of STM use

(rs=0.08, p=0.6).


205

4.5.9 Factors affecting log reading rate with STM

4.5.9.1 Correlation between vision and reading variables

The correlations of log reading rate for both passages and sentences and vision

measures are summarised in Table 4.12. As expected, log reading rates without STM

were highly correlated with log reading rates with STM using both reading tasks at

week 2 (Figure 4.21). In addition, both log reading rates with (r=0.71, p<0.001) and

without STM (r=0.77, p<0.001) were significantly correlated with near word visual

acuity (Figure 4.22).

y = 0.7963x + 0.2685R2 = 0.7714(passage)

y = 0.9294x + 0.0097R2 = 0.7517 (sentence)

0.8

1.3

1.8

2.3

2.8

1.2 1.4 1.6 1.8 2 2.2 2.4

Log reading rate without STM (wpm)

Log

read

ing

rate

with

STM

(w

pm)

Passage

Sentence

Linearregression fit(Passage)

Linearregression fit(Sentence)

Figure 4.21

Correlation of log reading rates with and without STM at week 2 for passages and sentences.

The graph shows that the correlation of log reading rates with and without STM is strong and highly significant.


206

Table 4.12 Correlations (Pearson r) between log reading rates with and without STM and clinical measures without STM at week 2

** Correlation was significant at the 0.01 level (2-tailed). * Correlation was significant at the 0.05 level (2-tailed).

PASSAGES SENTENCES

Log reading

rate on passages

Log reading rate without

STM

Contrast sensitivity


Near visual acuity

Text visual acuity

Visual field loss

Log reading rate without

STM



Near visual acuity

Text visual acuity

Visual field loss

(log wpm) (log wpm) (log) (logMAR) (logMAR) (logMAR) (Steradians) (log wpm) (log) (logMAR) (logMAR) (logMAR) (Steradians)

With STM (n=43) 0.84** 0.49** -0.43** -0.71** -0.74** -0.36* 0.84** 0.52** -0.66** -0.82** -0.83** -0.46*

At week 2

Without STM(n=43) 0.35* -0.48** -0.77** -0.73** -0.33* 0.44* -0.66** -0.76** -0.75** -0.52*


207

y = -0.7377x + 2.6064R2 = 0.4955

(reading w ithout STM)

y = -0.6205x + 2.2474R2 = 0.5491

(reading w ith STM)

0

0.5

1

1.5

2

2.5

00.511.5

Word near visual acuity (logMAR)

Log

read

ing

rate

(wpm

)Reading ratewithout STM

Reading rate withSTM

Linear regressionfit (RR withoutSTM )

Linear regressionfit (RR with STM )

Figure 4.22a

Correlation between log reading rate for passages and near visual acuity at week 2.

Near visual acuity was strongly correlated to log reading rate with and without STM on large print.

y = -0.6885x + 2.6002R2 = 0.5698

(reading w ithout STM)

y = -0.8414x + 2.5793R2 = 0.6714

(reading w ith STM)

0

0.5

1

1.5

2

2.5

3

00.511.5

Word near visual acuity (logMAR)

Log

read

ing

rate

(wpm

)

Reading ratewithout STM

Reading rate withSTM

Linear regressionfit (RR withoutSTM )

Linear regressionfit (RR with STM )

Figure 4.22b

Correlation between log reading rate for sentences and near visual acuity at week 2.

Near visual acuity was strongly correlated to log reading rate with and without STM on large print.


208

4.5.9.2 Factors predicting log reading rate with and without STM

As near visual acuities for word and text were highly correlated, only text visual

acuity was used in the multiple regression analyses. Of the clinical measures at week

2 (distance visual acuity, near text visual acuity, contrast sensitivity, visual field

loss), text visual acuity was the only significant predictor of log reading rate without

STM for passages (stepwise multiple regression: adjusted R2=0.58, p<0.001, Table

4.13).

Of the clinical measures (distance visual acuity, near text visual acuity, contrast

sensitivity, visual field loss and log reading rate without STM), log passage reading

rate without STM was the best predictor of log reading rate with STM for passages

(stepwise multiple regression: adjusted R2=0.74, p<0.001, Table 4.13). The addition

of contrast sensitivity improved the regression model with the two variables

accounting for 76% of the variance in log reading rate without STM for passages

(p<0.001). If log reading rate without STM was not included in the regression model,

near text visual acuity was the only significant predictor of log reading rate with

STM (adjusted R2=0.50, p<0.001, Table 4.13). The clinical parameters which

significantly predicted reading rate for sentences were similar to those for passages

although the predictive values (adjusted R2) and the levels of significance (p-value)

were slightly higher (Table 4.13).

4.5.9.3 Factors predicting the change in log reading rate with STM

As most subjects’ vision and log reading rates decreased at the last visit (week 20),

the analysis investigating factors predicting the change in log reading rate with STM

did not include log reading rate at the last visit. This was to ensure that the change of

log reading rate was not affected by the vision deterioration. The slope of the log

reading rate with STM from weeks 2 to 8 was determined for each subject. Positive

slope indicated an improvement in log reading rate.


209

Table 4.13 Summary of the multiple regression analyses

PASSAGES SENTENCES Dependent variable Independent variables Predictors Adjusted R2 P-value Predictors Adjusted R2 P-value

Log reading rate without STM

Distance visual acuity Near text visual acuity Contrast sensitivity Visual field loss (steradians)

Near text visual acuity 0.58 <0.001

Near text visual acuity and Visual field loss (steradians)

0.55

Additional factor improved

R2 to 0.62

<0.001

Log reading rate with STM

Distance visual acuity Near text visual acuity Log reading rate without STM Contrast sensitivity Visual field loss (steradians)

Log reading rate without STM and Contrast sensitivity

0.74

Additional

factor improved R2

to 0.76

<0.001

Log reading rate without STM and Near text visual acuity

0.71


R2 to 0.78

<0.001

Distance visual acuity Near text visual acuity Contrast sensitivity Visual field loss (steradians)

Near text visual acuity 0.50 <0.001

Near text visual acuity and Visual field loss (steradians)

0.66


R2 to 0.69

<0.001

Change of log reading rate with STM (slope)

Distance visual acuity Near text visual acuity Log reading rate without STM Contrast sensitivity Visual field loss (steradians) Groupsl

Visual field loss (5 mm ster-radian) and Near text visual acuity

0.07

Additional factor

improved R2 to 0.14

0.02

Visual field loss (ster-radian) and Near text visual acuity

0.15


R2 to 0.28

0.005

0.001

l Each group was coded to a binominal variable as either 1 or 0.


210

The regression model showed that visual field loss (expressed in terms of steradians)

together with near visual acuity significantly predicted the improvement in log

reading rate with STM, however, these two factors explained only 14% of the

variance in the change of log reading rate with STM (Table 4.13).

4.6 Discussion

4.6.1 Effect of reading practice on log reading rate with STM

Log reading rates with STM for the experimental groups (control and two practice

groups) improved significantly for passage reading at critical print size from weeks 0

to 2, stabilised to week 8 and then reduced at week 20 (refer to Figure 4.6). Provided

that vision remained stable through the experimental period (five months), reading

rate with STM stabilised after a significant improvement from weeks 0 to 2 (refer to

Figure 4.12). This is in agreement with previous studies which have found a greater

improvement in magnifier reading rate at the first 2 visits followed by a gradual, but

still significant, improvement afterwards (Goodrich et al., 1977; Bowers, 2000 (b)).

Significant improvement in reading rate with STM for sentences was found from

weeks 1 to 8. The duration of improvement in reading rate with STM for sentences

was longer than for passages.

There were a number of other results which failed to reach statistical significance but

indicated trends which may have been confirmed with a large subject sample sizem.

These results are discussed below.

Although the improvement in log reading rate across time for the practice groups (P1

and P2) was not significantly greater than the control group (Figure 4.12), there was

a trend towards a greater improvement in the practice groups. Individual results

suggested that large print reading practice under full (P1) or reduced field of view

(P2) enhanced the improvement in log reading rate with STM (Figure 4.7). If the

sample size was increased, there might be a positive effect of large print reading

practice on magnifier reading rate.


211

Relative log reading rate represented the net change in individuals’ log reading rates

across time (Figure 4.8). Although the change in relative log reading rates was not

significantly different between groups, there was a trend suggesting a greater relative

improvement in log reading rate with STM for the practice groups (P1 and P2)

compared to the control group. Large print reading practice (with or without

restricted field of view) before the provision of the STM may be beneficial to reading

performance with STM but further research is required.

With regards to the two different types of reading practice prescribed for subjects in

P1 and P2, the results showed that the reading practice with a simulated STM, the

practice stand to reduce the field of view, did not confer any greater benefit over the

large print reading practice with unrestricted field of view. The non-beneficial

practice effect on page navigation and reading under the restricted field of view on

reading performance with STM might be due to insufficient practice, inappropriate

strategies for page navigation or perhaps non-compliance. Reading rate was

significantly reduced for P2 group when the practice stand was first introduced (i.e.

reading under a restricted field of view). This may be due to inexperience in reading

under a restricted field of view and manipulating the practice stand during reading

large print. It is arguable whether 10 minutes daily practice was sufficient to become

familiar with the use of a practice stand in reading large print. In addition, the

practice stand reduced both the horizontal and vertical fields of view and no

instruction on how to manipulate the practice stand for the large print reading

practice was given to the subjects in P2. For these reasons, subjects might still be

unfamiliar with the ways to manipulate the practice stand such that their reading

performance could be enhanced. As a consequence, the extra reading practice did not

give any advantage on improving the subsequent reading performance with STM.

m The calculation for the required sample size was summarized in Appendix 2.


212

For subjects in all groups (control, P1, P2 and clinical groups), log magnifier reading

rate improved significantly from weeks 2 to 8 and then reduced at week 20 (Figure

4.6). The increase in the log reading rate between weeks 2 and 4 for the clinical

group was significantly greater than that in the other groups. The reasons for the

substantial improvement in the log reading rate of the clinical group are probably the

different experimental interventions among groups and the baseline vision in the

clinical group.

Experimental interventions

Although the STM was not prescribed for home use between weeks 0 and 2 for

subjects in the experimental groups (control and practice groups), repeated measures

of log reading rate with STM were made in the laboratory. The improvement in log

reading rate within this period was greater than the improvement achieved after the

STM was prescribed for home use (weeks 2 to 8). This suggests that a small amount

of practice, as given with the measurements of reading rate in the laboratory, is

effective in achieving significant improvement in reading rates with STMs. For the

log reading rate analyses involving all groups, the data for weeks 0 to 2 were

excluded. Therefore, although there was little improvement in log reading rate with

STM for the control and practice groups from weeks 2 to 4 (as most of the

improvement had already occurred during weeks 0 to 2), comparatively there

appeared to be a much greater improvement in log reading rate for the clinical group,

as the clinical group had had no exposure to the use of STM prior to week 2.

In addition, differences in recruitment procedures and magnifier prescription for

subjects in the clinical group may be another reason for the substantial improvement

in log magnifier reading rate between weeks 2 and 4 in the clinical group. Firstly,

clinical group subjects were recruited following their consultation in the LVC.

Although some resting time was given before they started the experimental measures

for the study, they might still have been tired after a 2 to 3 hour consultation in the

LVC. Because of fatigue, their reading performance at the initial visit may have been

lower than if they had been assessed before their clinic consultation, resulting in a

slower log reading rate with STM as the baseline measure (week 2). However, for

the subsequent assessments from weeks 4 to 20, the majority of subjects (9 out of 11)

in the clinical group had the research assessment scheduled at a different day or prior


213

to their follow up consultations in the clinic. Therefore, the effect of fatigue is

unlikely to have been a factor affecting their reading performance between weeks 4

and 20. Thus, the improvement in log reading rate from weeks 2 to 4 may be

artificially high. Secondly, often a number of low vision aids were selected for trial

by patients in the clinic consultation. Time for the clinical group subjects to practise

using the STM specifically might have not been sufficient, whereas the subjects in

the experimental groups (control and practice groups) were more familiar with the

STM as these subjects had had prior use of STM in the laboratory measurements

(although they were not given the STM for home practice for the first two weeks).

Due to familiarity in using the STM, log magnifier reading rate for participants in the

experimental groups may have been faster at the baseline assessment.

Difference in baseline vision

While the difference in log reading rate with STM between groups was not

significant, log reading rate with STM in the clinical group was slightly slower than

that in the experimental groups (N, P1 and P2). Subjects from the clinical group were

directly recruited from patients attending the LVC while subjects from the

experimental groups were recruited from the database in the VRC or referral from

ophthalmologists. Even though similar resources were used for recruitment, subjects

in the experimental groups were not randomly selected into control or one of the two

practice groups. They were allocated into different groups according to their near

visual acuity and age. Therefore, it was to be expected that there was no significant

difference in the baseline measures between the experimental groups. However, this

procedure was not possible for subjects in the clinical group. Provided the patients

fulfilled the recruitment criteria and agreed to participate, they were recruited into

this study. They were a "convenience sample" and visual acuity and age was not

matched with those of the other groups. Thus, the distance visual acuity in the

clinical group was significantly worse than that of the control group. Subjects in the

clinical group also had significantly larger visual field defects than those in P1. As

there was a high correlation between reading rate and near visual acuity in this study

(Figure 4.22) and previous studies (Rubin, 1987; McMahon et al., 1991; Bullimore

and Bailey, 1995; Bowers, 1998 (a); Lovie-Kitchin et al., 2000 (a)), the slightly

poorer vision of the participants in the clinical group would suggest that log reading

rate of this group might also be worse (slower) than the other groups. There was a


214

tendency for this to be the case, although the differences in log reading rates with and

without magnifiers among groups were not statistically significant because of the

range of reading rates in a small sample.

Log reading rate with STM was significantly reduced at the last visit for subjects in

all groups. AMD is a progressive eye disease, so it was not surprising that vision and

reading rate deteriorated 5 months after the subjects were recruited in this study.

Monitoring of the retinal changes and progression rate of AMD was not performed in

this study but vision measures (distance and near (word) visual acuities, threshold

print sizes achieved with STM) were assessed. Among these measures, distance

visual acuity was significantly reduced at the last visit (Figure 4.9). The reductions in

vision and reading rate at week 20 reflect the progression of AMD over time.

The distance visual acuity appears to be reduced at the last visit to a greater extent in

the control group than that in the other groups. Distance visual acuity measured by

Bailey-Lovie distance acuity chart has been described as the “gold standard” (Enoch,

1998) for visual acuity measures and for defining any significant change in vision

because of its repeatability and reliability (Lovie-Kitchin, 1993). In AMD the highest

rate of acuity loss occurs for eyes with better distance visual acuity than in eyes with

moderate or severe visual disability (Sunness et al., 1999; Sunness et al., 2002).

Comparing the baseline vision measures (week 2) among groups, distance visual

acuity for subjects in the control group was significantly better than that in the

clinical group (Table 4.7). This suggests that people in the control group had an

earlier stage AMD while those in the clinical group had more advanced AMD. In line

with the findings of Sunness and colleagues (1999; 2002), subjects in the control

group, who started with better visual acuity, appeared to show greater reduction in

acuity than that in the clinical group over time even though this was not statistically

significant (Figures 4.9 to 4.11). Threen of ten subjects from the control group

showed a reduction in distance visual acuity by 0.2 logMAR or more between weeks

8 and 20, which was reflected in a marked decrease in their reading rates (Figure 4.7

and Appendix 2). This supports the argument that AMD progresses faster for people

at the early stage with better visual acuity than people in the moderate or advanced

n Participants whose distance visual acuity reduced by 0.2 logMAR or more were subjects 13, 20 and 22.


215

stage of AMD. In addition, more subjects in the control group had wet AMD (four

out of 10 subjects) than the other groups (two out of 11 subjects). There have been

numerous studies stating that the reduction in visual acuity for people with wet AMD

is sudden and more severe compared with people with dry AMD (Lovie-Kitchin and

Bowman, 1985; Alexander, 1993; Vinding, 1995; Schuchard et al., 1999).

4.6.2 Comparison of log reading rate without STM with log magnifier reading

rate

Previous studies have shown that reading rate reduces significantly when magnifiers

are introduced for reading (Mancil and Nowakowski, 1986; Dickinson and Rabbitt,

1991; Cohen and Waiss, 1991 (a); Cohen and Waiss, 1991 (b); Bowers and

Ackerley, 1994; Dickinson and Fotinakis, 2000). Participants in these studies had

either normal vision (Mancil and Nowakowski, 1986; Cohen and Waiss, 1991 (a);

Cohen and Waiss, 1991 (b); Dickinson and Fotinakis, 2000) or simulated low vision

(Dickinson and Rabbitt, 1991) rather than “real” low vision. This current study

recruited people with low vision who had no experience in using STMs for reading.

The findings in this study agreed with previous studies (refer to section 4.5.3),

indicating that log reading rate was reduced when the STMs were first introduced to

subjects in both the experimental and clinical groups (Figure 4.11).

4.6.2.1 Maximum reading rate without STM

If AMD did not progress significantly and vision was stable, log reading rate without

STM did not significantly change across time (Figure 4.13). Reading practice did not

have any significant impact on the log reading rate without STM. This indicated that

neither large print reading practice nor repeated measures of the log reading rate gave

any increase in subjects’ maximum reading performance without STM. This result is

in agreement with the suggestion by Lovie-Kitchin et al. (2000 (a)) that the reading

rate without magnifier reflects an individual’s maximum reading performance.

Because of its repeatability, maximum reading rate (without STM) could be one of

the parameters used to monitor the progression of the disease in addition to vision

measures.


216

Log reading rates without STM for participants in the experimental groups were

measured repeatedly for 3 visits before the provision of the STM (week 0 to 2). In

contrast log reading rate was measured at the first visit (equivalent to week 2 in the

experimental groups) for subjects in the clinical group with no repeated measures.

Although the results for the experimental groups showed no significant difference in

the repeated measures of log reading rate without STM, it is still arguable whether

the log reading rate without STM can truly reflect the maximum reading

performance of the participants in the clinical group as the measure was limited to

one visit. It was suggested previously that (refer to section 4.6.1), the initial measures

of log reading rate without STM from the clinical group were probably slower than

maximum reading rate because of fatigue. The effect of fatigue on reading rate

should be controlled in future study.

4.6.2.2 Reading rate with STM

Lovie-Kitchin et al. (2000 (a)) stated that the difference in reading rates with and

without magnifiers reflected the extent to which the reading rate with the magnifier

could be improved. This study agrees with Lovie-Kitchin et al.’s suggestion (2000

(a)). In the current study, log reading rate with STM increased across time while log

reading rate without STM did not change. For subjects in the experimental groups,

there was a significant reduction in log reading rate when the STM was first

introduced. In support of one of the experimental hypotheses, the log magnifier

reading rate improved such that it was not significantly different from log reading

rate without STM after one week reading practice (P1 and P2 groups) or repeated

measures (N group) (refer to sections 4.5.1.2 and 4.6.1). Although there was no

significant interaction between the experimental groups in the significant

improvement of log magnifier reading rate, there was a trend for the practice groups

to show greater improvement in log reading rate with STM over time (Figures 4.6

and 4.8). In addition, the reduction in log reading rate for the control group was

comparatively less than that for the practice group at week 0 when a STM was used

(Figure 4.14). As a consequence of a smaller reduction in reading rate at the initial

visit, there was not much room for the reading rate to improve to the maximum

reading rate (without STM) that the subject could achieve. This suggests that the


217

reading practice is sufficient to improve reading rate with STM to maximum reading

rate (i.e. reading rate without STM).

For subjects in the clinical group, log reading rate with STM was no longer

significantly different from the log reading rate without STM after 2 weeks practice

in using the STM at home (at week 4). This suggests that patients who have not had

prior reading practice require about 2 weeks practice using the STM at home before

their log reading rates with STM reach their maximum log reading rates (without

STM). This result agrees with the findings of McMahon and Spigelman (1989) who

showed a significant improvement in reading rate with STM with 2 weeks practice

for young subjects with normal vision. As reading with STM was not assessed

between weeks 1 and 2 after the provision of a STM, no definite conclusion can be

drawn about the minimum period of time that is required for patients to reach

maximum reading rate with the STM, but the results indicate that it is two weeks or

less.

Comparing log reading rates with and without STM over time, subjects in the control

group behaved slightly differently from subjects in the two practice (P1 and P2) and

clinical groups. There was a significant reduction in log reading rate when the STM

was first introduced but the reduction for the control group seemed to be less than

that for the other groups (Figure 4.14). This suggests that the difficulties imposed on

reading by using a STM (reduced field of view and navigational demands) were less

for the control group, which could possibly be because of the slightly (but not

significantly) better vision in this group. Magnification (in terms of EVD) was

calculated from subjects’ near visual acuity and their target reading material. For

people with better vision, the required magnification was less (larger EVD) to read

the same print size with the magnifiers. Although the magnification prescribed was

not statistically different among groups (one way ANOVA, Groups (4), F2,40=1.37,

p=0.27), the mean magnification prescribed for subjects in the control group was

slightly less, i.e. EVD greater (EVD: 12.57 ± 5.22 cm) than that for the clinical

(EVD: 8.65 ± 4.52 cm) and both practice groups (EVD for P1: 10.9 ± 6.74 cm, EVD

for P2: 11.98 ± 7.82 cm). Because of the inverse relation between lens diameter and

the equivalent power of the lens, a STM with lower magnification will have a larger


218

lens diameter than a STM with higher magnification. Difficulty in reading under a

restricted field of view and difficulty manipulating the STM have been suggested as

reasons for the reduction in reading rate when a low vision aid is first applied. For

the subjects (in the control group) who were prescribed magnifiers with larger lens

diameters, the impact of reduced field of view may have been less than for the

subjects who were prescribed magnifiers with narrower lens diameters. As such, log

reading rate for subjects in the control group might have been reduced to a lesser

extent when the STM was introduced for reading. Other factors, which may affect

reading rate when the magnifier is first supplied, such as motor control and page

navigation skills (see Chapter 5) were not assessed in this experiment.

4.6.3 Comparison of log reading rates under different fields of views

Previous literature has stated that reduced field of view and navigational problems

are the major difficulties that people with low vision encounter when they use STMs

for reading (McMahon and Spigelman, 1989; Spitzberg et al., 1989; Beckmann and

Legge, 1996; Fine et al., 1996; Bowers, 2000 (b)). Familiarity with the “practice

stand” introduced these impediments to reading – reduced field of view, the

requirement for manipulation of a stand and navigation across the text during reading

– and thus simulated reading with a STM. The hypothesis was that practice with this

practice stand while reading large print would improve the log reading rate with the

STM. However, the results showed that reading performance for subjects from P1

and P2 were not significantly different (Table 4.7). This finding suggests that reading

practice with a “practice stand” does not give any additional benefit over simple

large print reading practice in the improvement in reading performance with STM.

However, it should be noted that this result was based on a relatively small sample

size and there was no objective determination of the extent to which P2 subjects

complied with the use of the practice stand during the home practice period or the

large print reading practice.

In this study, all subjects were supplied an appropriate STM to read their target

material at home. With 18 weeks practice in using the STM, the participants became

more experienced in using their low vision aids for reading. At the last visit, there

was no significant difference in the log reading rates without STM under normal and


219

restricted field with a practice stand compared to the log reading rate with STM for

subjects as a whole group. This result may be important for clinical purposes. The

effect of navigational problems and reduced field of view imposed by the practice

stand on reading did not significantly affect reading rate provided that subjects were

familiar with reading with a narrow field, were experienced in manipulating low

vision aids and appropriate magnification was given for reading (with STM and with

practice stand). However, this conclusion is limited because there are no supporting

data of log reading rates with the practice stand for all subjects at the first visit which

indicates a reduction in reading rate on large print when the field of view was

narrowed down by the practice stand.

A range of fields of view, from 5 to 24 characters, have been found in previous

studies to be required to achieve optimal reading rate without STM (Legge et al.,

1985 (b); Lovie-Kitchin and Woo, 1987; Whittaker and Lovie-Kitchin, 1993;

Beckmann and Legge, 1996; Den Brinker and Bruggeman, 1996; Fine et al., 1996).

Comparison of reading rates (with and without STM) as a function of field of view

was not included in this study. The print size of the passages used to measure the

reading rate under a restricted field was the individuals’ critical print size. Therefore,

sufficient acuity reserve (or appropriate magnification) was provided for reading

with the practice stand even though the field of view was narrowed down to

approximately 6 characters horizontally. The non-significant difference in log

reading rates under normal and restricted field of view agrees with the argument that

field of view is not a significant factor limiting reading rate provided sufficient

magnification is given (Lovie-Kitchin and Whittaker, 1998 (b); Lovie-Kitchin and

Whittaker, 1999 (b)).

The analysis for the separate groups showed that log reading rates with and without

STM (under normal and restricted field of view) were not significantly different for

the control and practice groups. However, participants in the clinical group showed a

significant reduction in log reading rate under a restricted field of view when

compared with the log reading rate under full field or the log reading rate with STM

at week 20. The significant reduction in log reading rate under a restricted field of

view for the clinical group could be due to the poorer vision of this group. The

majority of subjects in this study showed a significant reduction in visual acuity at


220

the last visit; subjects in the clinical group were no exception. Despite the fact that

greater vision deterioration was found in the control group, the vision levels in the

clinical group were still comparatively worse than the vision levels in the other

groups at week 20.

4.6.4 Reading performance with different reading materials

Carver (1990) has shown that reading rate for short sentences is faster than reading

rate for long passages. Results from Chapter 3 and this chapter agree with this

finding. Because of the faster reading rate with sentences, any disturbance to habitual

reading may result in a greater reduction in reading rate compared with reduction in

reading rate with passages. This may explain why subjects required longer duration

using the STM to improve reading rate with STM for sentences to maximum reading

rate (without STM). In clinical consultations, reading performance with both reading

materials should be assessed since the results reflect reading behaviour for different

daily reading tasks although reading rates for the two tasks are highly correlated.

4.6.5 Use of stand magnifiers

The provision of a STM increased subjects’ frequency and duration (with STM) of

reading, as they could read their desired reading materials more often (81% read

more than once per day) or for longer durations either reading large print without

STM and small print with STM. The high frequency and duration (with STM) of

reading reported in this study could possibly be due to the characteristics of the

recruited subjects. Subjects participating in the study might be more motivated to

seek ways to alleviate their reading disability, so they might contribute more time

and effort to read compared to other patients who attend the LVC.

However, both the frequency and duration of reading were significantly reduced after

18 weeks. The reason for the reduction could be attributed to the deterioration of

vision across time. As a consequence of reduction in vision, the prescribed low

vision aids might not be as helpful as before to allow subjects to achieve small print

reading.


221

Despite the reduction in vision at week 20, about 65% of the subjects still read more

than once every day with or without their STM. This finding was similar to the

results from previous studies, in which the majority of subjects with low vision (>

60%) reported that they used their low vision aids regularly (Temel, 1989;

McIlwaine et al., 1991; Van Rens et al., 1991; Leat et al., 1994; Bischoff, 1995;

Davis et al., 1995; Shuttleworth et al., 1995; Harper et al., 1999).

Before any STM was prescribed, 65% of subjects preferred to read for longer than 10

minutes at a time. After they received a STM, the duration of reading without STM

became less. More participants changed the reading duration without STM from '10

minutes or longer’ to ‘less than 10 minutes’. There are two explanations for this

result. Firstly, most of the participants used near spectacles (79.1%) for reading

before any STM was prescribed, which probably did not provide the best

magnification to read their desired reading materials. Due to this limitation, they

could hardly read their preferred reading materials such as newspaper or mail.

Instead, they read large print books and newspaper headlines of larger print sizes

(Figure 4.18a). When a suitable STM was given, people could read their desired

reading materials, as print size was not a limiting factor affecting their choice of

reading tasks. As a result, subjects might invest more time reading smaller print

newspaper rather than reading large print materials. Therefore, the duration of

reading without STM reduced across time. Secondly, AMD progresses gradually

over time, which results in a slow reduction in vision and reading ability. Due to the

vision limitation, subjects might not be able to read the large print materials which

they could manage before, and hence the duration of reading without STM reduced.

Frequency of magnifier use and duration of reading with STM are indirect measures

of the usefulness of the newly prescribed low vision aids. Just after STM provision,

62% people used the STM for longer than 10 minutes. Eighteen weeks later, only

44% of participants read for longer than 10 minutes with their STM. This result

agrees well with the findings of Leat et al. (1994) who reported that 49% of their 57

patients with low vision aids prescribed for six months to three years continued to

use their low vision aids for 10 minutes or longer. However, the proportion reported

in current study was smaller than the result obtained Bischoff (1995) and Harper et

al. (1999). Bischoff (1995) conducted a questionnaire on 112 patients with AMD


222

who had been prescribed low vision aids one to five years previously. His survey

found that 58% of the participants used their magnifiers for longer than 10 minutes

each day. The discrepancy between these two studies might be due to the different

visual acuities and perhaps different progression rates of AMD. The majority of

subjects in the current study showed decreased vision after 20 weeks. In contrast, no

vision measures were conducted in Bischoff’s study when the low vision aids were

prescribed and in the follow up assessment. Perhaps the progression rate in

Bischoff’s study was not as significant as that in the current study because the

majority of subjects still used their low vision aids for reading. Also, reading

materials read with the low vision aids were not reported in Bischoff's study.

Therefore, it was unclear that the subjects could achieve the reading goals defined at

the time of the initial consultation with their low vision aids. Harper et al. (1999)

followed up 56 AMD subjects (mean age of 81.4 years) who had been prescribed low

vision aids (for different visual tasks) one year after their initial visit and reported

that 55.4% of the subjects used the low vision aids for at least 5 minutes or more.

The higher percentage shown in Harper’s study (1999) was mainly due to the

different design of the questionnaire. In the question addressing the duration of low

vision aid use, the response categories for the first two options in Harper et al.’s

study were as follows: (1) for 15 minutes or more; or (2) for at least 5 minutes or

more but less than 15 minutes. In contrast, the categories used in the current study

were: (1) for 1 hour or more; or (2) for at least 10 minutes or more but less than 1

hour. The discrepancy in the minimum duration required for the second category

between the two studies could be the reason for the different percentages found.

4.6.6 Satisfaction with stand magnifiers

The majority of subjects (72%) were initially satisfied with the prescribed STM as

their response to the question on the usefulness of the STM was “excellent or good”.

However, even though their vision was reduced at the last visit, 65.5% of the

participants still reported that the STMs were useful to them. This result agrees with

the responses in Harper et al.’s (1999) study that 58% of subjects ranked their low

vision aids in the categories “quite” to “extremely important” one year after they had

been prescribed the low vision aids. Satisfaction with STM was significantly


223

correlated with log magnifier reading rate, frequency of reading and duration of STM

use in the current study (refer to section 4.5.8.4).

Even though the log reading rate with STM was not significantly different from the

log reading rate without STM (from weeks 4 to 20), participants did report

limitations in using STM. A number of subjects complained that it was difficult to

manipulate the STM accurately to find the start of the next line, or the field of view

of the STM was too narrow to provide a comfortable environment for reading. To

address the first complaint, a number of commercially available low vision aids have

incorporated a tracking deviceo to assist people with low vision who have reading

difficulties. An experiment to assess this type of assistance is reported in Chapter 5.

Other complaints were related to the design of the STM in terms of illumination.

Some participants were not satisfied with the amount of illumination provided by the

STM with the 2-batteries handle. They preferred using 3 batteries or the mains

alternate current power to increase illumination. In addition, reflections from the lens

and flickering of the light were the other limitations reported.

4.6.7 Predictors of reading rate with and without magnifiers

Near visual acuity was the single significant predictor of log reading rate without a

STM for subjects with AMD, accounting for 58% of the variance in log reading rate.

This finding agreed with the results of previous studies in which near visual acuity

was a strong predictor of reading rate for subjects with macular degeneration

(Bullimore and Bailey, 1995; Lovie-Kitchin et al., 2000 (a)). Bullimore and Bailey

(1995) measured the eye movements of 13 subjects with AMD and found that word

reading acuity was the single strongest predictor of both maximum reading rate

(without magnifier) and reading rate for passages that were twice the threshold print

size (0.3 log acuity reserve). Near visual acuity accounted for 53% of the variance in

each of the reading measures. Similarly, Lovie-Kitchin et al. (2000 (a)) recruited 22

people with macular degeneration and reported that maximum oral reading rate was

o Visual tracking magnifier from Coil (equivalent power of 18.9D); flip up reading line with stand magnifiers (12 to 20D) and specially illuminated reading line (7.6 and 11.4D) from Eschenbach.


224

well predicted by near visual acuity, accounting for 72% of the variance in reading

ratep.

Log reading rate without STM was a good predictor of reading performance with

STM as reported in previous studies (Ahn and Legge, 1995; Lovie-Kitchin et al.,

2000 (a); Bowers et al., 2001 (b)). Log maximum reading rate (without STM)

accounted for 74% of the variance in the log magnifier reading rate. The inclusion of

contrast sensitivity into the model slightly improved the regression coefficient, to

account for 76% of variance in log reading rate with the magnifier. This agrees with

the findings of (Leat and Woodhouse, 1993) who reported that contrast sensitivity at

0.5 cycles per degree was a good predictor of log reading rate with magnifier for

people who were visually impaired. Contrast sensitivity is a measure of the lowest

letter contrast that a person can detect. These results confirm that deficits in contrast

sensitivity have a small but significant effect on magnifier reading rate.

Near visual acuity was also a strong predictor of log magnifier reading rate, when log

maximum reading rate (without STM) was not included in the regression analysis. In

clinical consultations, maximum reading rate without magnifier may not be assessed

if insufficient time is allowed. The result from this study agrees with Lovie-Kitchin

et al. (2000 (a)) that near visual acuity can give an indication of the magnifier

reading rate when the magnifier is prescribed for patients with low vision.

In this study, near visual acuity could explain a high percentage of the variance in log

reading rates with and without magnifiers achieved by subjects with AMD. There

have been numerous studies investigating the reasons for slower reading rates among

subjects with visual impairment compared with subjects who are normally sighted. In

general, visually impaired people read more slowly than people with normal vision

even when the print size is optimal. The reason for the reduction in reading rate for

people with low vision has been largely attributed to the reduction in number of

letters per forward saccade or shorter forward saccades during reading (Rumney and

Leat, 1994; Bullimore and Bailey, 1995; Bowers et al., 2001 (b)). Legge et al. (1997)

argued that the reduction in reading rate for people with low vision was mainly due

p The reading rate reported by Lovie-Kitchin et al. (2000 (a)) was the silent reading rate rather than oral reading rate. However, these two reading rates were highly correlated (r=0.9).


225

to the reduction in visual span. They defined visual span as the number of characters

that can be recognised on each fixation. They hypothesised that fewer letters were

recognised on each fixation, therefore forward eye movements (saccades) were

shorter and more numerous, and reading rate was reduced for visually impaired

people (Legge et al., 1997). This implies that as near visual acuity for visually

impaired people decreases, visual span decreases and therefore reading rate

decreases.

Vision and reading variables were the objective measures in this study to assess

subjects' baseline performance before the magnifier was prescribed. The

questionnaire addressed the subjective reports on the effectiveness of the magnifier

for reading. Apart from these factors, there are many non-visual variables that could

affect the reading behaviour with STM, which in turn could affect the improvement

of reading rate with magnifiers, such as subjects' physical (e.g., health and

intelligence) and psychological status (e.g., motivation and acceptance of using a

STM as their reading aid). However, not all factors could be included as measures in

this study. Visual field loss and near visual acuity could only explain a small

proportion of the variance in the change in magnifier reading rate. People with visual

impairment who have been prescribed hand-held or stand magnifiers have to

manipulate the magnifiers during reading. As such, an individual’s motor function

may have an impact on reading performance with the magnifiers and the

improvement in reading rate as they learn to use a magnifier. Further research to

address the role of motor function in reading with magnifiers should be conducted

(Bowers et al., 2002 (a)). In the next experiment (Chapter 5) magnifier movements

were recorded as subjects moved a STM across a passage of text. For some subjects,

these recordings highlighted problems with motor control during manipulation of the

STM.


226

4.7 Conclusions and recommendations

Prescribing optical low vision aids is the most common therapeutic approach to assist

people with visual impairment to read. Previous studies have found that reading rate

is reduced when a hand-held or stand magnifier is introduced. Results of this study

support this finding. Reading rate is significantly reduced when a STM is first

prescribed for subjects with AMD. However, one week large print reading practice

(with or without a reduced field of view) at home or in-office practice with magnifier

could benefit AMD subjects’ future use of a STM such that reading rate quickly

returns to maximum (equivalent to reading rate on large print). The results suggested

that the large print reading practice prior to magnifier prescription may give a more

rapid improvement in reading rate than that achieved by the subjects who did not do

any large print reading practice, although this did not reach statistical significance.

Despite a significant reduction in reading rate when the STM was first introduced,

magnifier reading rate increased over a short period of time. Provided that there is

sufficient acuity reserve (appropriate magnification), the results of this study suggest

that with two weeks home practice with the magnifier, reading rate with magnifier

can reach maximum reading rate (without magnifier) when no prior reading practice

is given (control group subjects).

For clinical purposes the low vision practitioners could perhaps recommend AMD

patients read large print books of their critical print size (CPS) before their first visit

at the clinic. The CPS could be estimated from the visual acuity and reading goals

provided by the patient or referring practitioner. The results of this study suggest that

reading large print daily could reduce the time for magnifier reading rate to return to

maximum reading rate. In the low vision consultation, reading rates with and without

magnifiers should be assessed as the baseline reading measures Reading rates from

the clinical group in this study suggested that the assessment of reading rate should

be done at the beginning of the consultation to avoid the effects of fatigue and to

reflect patient's maximum reading performance. The reading measures should be

conducted with sentences and passages as the result reflects reading performances for

different daily reading tasks. A follow up visit is recommended after two weeks trial

with the magnifier to enable assessment of any improvement in magnifier reading


227

rate or any difficulty in using the magnifiers for reading. If the reading rate with the

magnifier shows no improvement compared with the reading rate at the previous

visit or the reading rate with the magnifier is still significantly slower than the

reading rate without magnifiers, further investigations of the magnifier would be

necessary such as the calculated magnification and the technique of manipulating the

magnifiers for reading. Lastly, most subjects in this study showed significant vision

deterioration 20 weeks after their first visit. This suggests that a review visit to

measure vision and reading performance with and without magnifiers should be

scheduled between 3 and 6 months after the patients’ first clinical consultation to

assess whether the prescribed low vision aids were still sufficient to achieve their

reading goals. However, this may be costly for low vision rehabilitation services to

provide such frequent review (Robbins, 1981). Patients could simply be telephoned

and advised to have a full assessment if they noticed the low vision aid was not

suitable to their needs.

Chapter 5 Does a line guide improve reading performance with stand magnifiers?

228

CHAPTER 5

Does a line guide improve reading performance with stand

magnifiers?

5.1 Introduction .................................................................................................... 229 5.1.1 Navigation difficulties in using stand magnifiers .................................... 229 5.1.2 Assisting navigation with magnifiers....................................................... 230 5.1.3 Quantifying navigation performance with magnifiers ............................. 231

5.2 Subjects ........................................................................................................... 232

5.3 Methods ........................................................................................................... 233 5.3.1 Vision assessment .................................................................................... 233 5.3.2 Field of view............................................................................................. 234 5.3.3 Design of line guide ................................................................................. 235 5.3.4 Measurement of magnifier movements.................................................... 237

5.3.4.1 Analysis of magnifier movements.................................................. 238 5.3.4.2 Magnifier movement parameters ................................................... 240 5.3.4.3 Categorisation of magnifier movement.......................................... 242 5.3.4.4 Navigation errors............................................................................ 245

5.3.5 Reading passages...................................................................................... 247 5.3.6 Procedures ................................................................................................ 249 5.3.7 Questionnaires.......................................................................................... 249

5.4 Analysis ........................................................................................................... 250

5.5 Results ............................................................................................................ 251 5.5.1 Comparison of navigation and reading performance using stand magnifier

with and without line guide...................................................................... 251 5.5.2 Magnifier movement strategy .................................................................. 253 5.5.3 Frequency of magnifier use and subjective feedback about navigation

difficulties................................................................................................. 254 5.5.4 Subjective preference for the line guide................................................... 257

5.6 Discussion........................................................................................................ 262 5.6.1 Comparison of navigation and reading performance using stand magnifier

with and without line guide...................................................................... 263 5.6.2 Comparison of stand magnifier manipulation strategy with and without line

guide 264 5.6.3 Subjective preference for the line guide................................................... 266 5.6.4 Design of the line guide ........................................................................... 267 5.6.5 Adaptation to the line guide ..................................................................... 268

5.7 Conclusion....................................................................................................... 269


229

5.1 Introduction

Prescribing optical magnifiers is one of the most common ways to overcome reading

disability for people who are visually impaired. The results of the questionnaire in

Chapter 4 indicated that some subjects with low vision still encountered difficulties

when using a stand magnifier (STM) for reading even though training on large print

reading and practice in using STMs had been given. These difficulties were mainly

self-reports and related to the narrow fields of view imposed by the STMs and

repositioning the STMs at the start of the next line accurately.

5.1.1 Navigation difficulties in using stand magnifiers

Reading with a STM involves two processes: processing the magnified visual image

and manipulating the magnifier across the text. Even though magnification facilitates

the resolution of smaller stimuli, the field of view of a magnifier can be quite

restricted. Field of view can be considered as the number of characters seen through

the magnifying lens. Manual control of the reduced field of view along each line of

text (forward movement) and back to the correct position at the beginning of the next

line (retrace movement) is required when reading with STMs. Beckmann and Legge

(1996) defined this manipulation process as “page navigation”. Navigation across the

reading material has been suggested as a major difficulty in using low vision aids

(LVAs) (McMahon and Spigelman, 1989; Spitzberg et al., 1989; Beckmann and

Legge, 1996; Fine et al., 1996; Bowers, 2000 (b)).

It is reasonable to expect that people who have no or minimal experience in using

magnifiers may encounter difficulties manipulating the magnifier in both forward

and retrace phases. Results from Chapter 4 indicated that reading rate was

significantly reduced (compared to large print reading rate) when subjects with low

vision first used STMs for reading. Inexperienced users may easily lose the place

where they read or move to an inappropriate line during retrace. As a consequence of

difficulty in page navigation, low vision people may read the same line twice or miss

a line when they read with their magnifiers. Therefore, any assistance to people who

are inexperienced in using magnifiers to facilitate page navigation and orient the


230

place where they are reading may improve navigation performance and hence

improve reading rate.

5.1.2 Assisting navigation with magnifiers

Kuyk et al. (1998 (a)) introduced a mechanized reading stand that provided to-and-

fro text movements with speed control and automated line indexing. A stationary

pointer made of white matte paper was attached to the reading stand. The pointer

indicated where the next text line began and directed readers’ gaze during reading to

where the text was drifted beneath the pointer to assist people with low vision using

optical aids. Subjects with low vision read significantly faster with this automated

reading stand than with their customary reading methods (Kuyk et al., 1998 (a)). The

automated stand removed the requirement for page navigation movements when

using an optical magnifier and also provided line orientation assistance; both these

factors could have contributed to the improvement in reading rate.

It may be inconvenient for patients to sit before a mechanised reading stand to read,

as this eliminates the portability of magnifiers and does not provide long-term

rehabilitation for reading. A few LVAs suppliersa incorporate a reading guide into

some of their magnifiers (in the form of a line that can be placed under the text being

read). Such reading guides provide similar line orientation assistance to that

incorporated in the mechanized reading stand, but the major difference is that the

guide and magnifier are under the manual control of the reader. The extent to which

line orientation assistance might improve navigation performance when magnifiers

are under manual control has not been assessed. LVA suppliers, not surprisingly,

suggest that their line reading-guides are beneficial for magnifier reading. However,

no published study has been done to investigate the effect of this reading accessory

on navigation and reading performance of people with low vision using STMs.

a http://www.eschenbach.com/home.asp (May, 2002) http://www.coil.co.uk/index.html (September, 2002)


231

5.1.3 Quantifying navigation performance with magnifiers

In order to fully assess navigation performance when magnifiers are under manual

control, the actual movements of the magnifiers should be recorded. This approach

enables navigation errors, navigation times and magnifier movement strategies to be

investigated. Page navigational movements when subjects with visual impairment

use CCTVs have been recorded (Beckmann and Legge, 1996; Den Brinker and

Bruggeman, 1996), but the method of manipulating the x-y platform of a CCTV is

somewhat different to the method of manipulation of an optical magnifier. Among all

available LVAs, simple optical magnifiers rather than electronic ones are more

frequently prescribed and preferred by patients in low vision rehabilitation

(Humphrey and Thompson, 1986; Leat and Rumney, 1990; Virtanen and

Laatikainen, 1991; Cole, 1993; Virtanen and Laatikainen, 1993). With the exception

of one study using normal subjects reading with hand-held magnifiers (Neve, 1989

(b)), it is only very recently that the navigational movements of subjects with low

vision using optical magnifiers have been investigated (Bowers et al., 2002 (a);

Bowers et al., 2002 (b)).

The purpose of this study was to investigate the effect of a simple reading guide (line

guide) on navigation performance and reading rate in subjects with low vision using

their habitual STMs. Performance with and without a line guide attached to the STM

was evaluated. To the author’s knowledge, there has been no previous study

investigating the effect on navigation and reading performance of a line guide

attached to a STM. It was therefore important to provide a complete assessment of

navigation performance and to investigate any changes in magnifier manipulation

strategies when the line guide was attached to the STM. Objective assessments of

navigation performance and magnifier movement strategies were derived from

magnifier movement recordings, while subjective responses to the line guide were

determined through a questionnaire.


232


1. The attached line guide would improve navigation performance (navigation

times and errors) with STMs and hence increase reading rate.

2. Subjects with less experience would have a greater preference for the line

guide than subjects who were experienced in using STM.

5.2 Subjects

Twenty-nine subjects with low vision aged between 65 and 89 years (mean age 80.41

± 5.68 years) were selected from the Queensland University of Technology (QUT)

Vision Rehabilitation Centre (VRC). All subjects were diagnosed by

ophthalmologists to have age-related macular degeneration (AMD) and had been

prescribed STMs as their main LVAs for reading (Table 5.1). Preceding their

recruitment, a comprehensive vision examination was conducted. Only subjects with

near visual acuity that had changed by less than 1 line (0.1 log unit) since their last

assessment in the VRC were recruited. This criterion was applied to ensure that the

magnification of their STMs was sufficient to achieve the subjects’ reading goal.

Subjects were all fluent English speakers and had no reported cognitive problems

(either indicated in the record card or self-reported by the subjects).

The inclusion and exclusion criteria for subjects in this experiment were the same as

those described in section 3.2. Additionally, the threshold print size and the critical

print size achieved with the STM for subjects participating in this study was better

than N10 and N16 respectively measured with Bailey-Lovie text reading chart.

Subjects who had had their magnifiers for 3 months or more were classified as

“experienced” while subjects who had had their magnifiers for less than 3 months

were defined as “inexperienced” users. All subjects participating in this study had

their STMs prescribed in the low vision clinics. The instructions given to the subjects

for manipulating their magnifiers were unknown.


233

Table 5.1 Details of subjects’ vision measures

ID Age Length of

visual impairment

Length of STM use


Near visual acuity


Threshold print size with STM

CPS with STM

Location of scotoma

(years) (months) (months) (logMAR) (logMAR) (log) (N-point) (N-point) 1 80 12 5 0.40 0.50 1.05 3 8 S 2 88 24 5 1.06 1.30 1.00 6 10 L, R 3 83 24 8 0.38 0.40 1.30 8 10 R 4 79 48 5.5 0.72 0.90 1.10 5 10 S, R, L 5 76 60 6 0.92 1.20 0.85 4 8 S, R, I 6 76 18 7 0.12 0.10 1.45 8 6 0 7 81 18 5 0.30 0.50 1.45 8 5 S, R 8 79 12 3 1.30 1.50 0.55 6 10 S, R, I 9 79 30 7 0.74 0.82 1.25 4 8 R, L

10 83 27 1 0.92 1.03 0.85 8 8 S 11 65 30 5 0.70 0.96 1.25 8 10 S 12 84 24 4 0.50 0.50 1.30 4 8 0 13 84 6 2 0.74 1.10 0.90 4 10 R, I, L 14 81 12 5 0.98 1.14 0.95 4 8 I, R, L 15 86 24 1.5 1.36 1.40 0.65 5 10 Central 16 74 4 0.25 1.02 0.98 1.05 4 10 Central 17 78 20 5 1.08 1.10 1.10 8 8 S 18 82 15 5 1.12 1.23 0.45 6 8 Central 19 70 36 0.1 0.42 0.70 1.15 3 10 0 20 76 4 0.75 0.26 0.30 1.15 3 6 0 21 78 0.8 0.5 0.78 1.20 0.70 5 10 S, R 22 71 20 0.25 1.12 1.04 0.55 3 8 S 23 89 2 0.5 0.86 1.00 1.15 5 6 S, R , I 24 88 72 60 1.12 1.50 0.85 6 10 S, L, I 25 84 9 0.25 0.96 1.20 0.65 6 6 S, R, I 26 82 12 0.5 0.46 0.70 1.15 8 8 0 27 86 60 36 1.08 1.45 0.85 5 8 S, L 28 85 42 5 1.12 1.00 1.05 3 8 S, R 29 85 72 4 0.94 1.40 0.95 5 8 Central

STM = Stand magnifier VA = Visual acuity CPS = Critical print size + Qualitative categorisation of visual field loss: Superior (S); Inferior (I); Right (R); Left (L); Central

(C)

5.3 Methods

5.3.1 Vision assessment

Vision and reading assessments were conducted monocularly with the eye used for

reading or the eye with better near visual acuity if the subject usually read

binocularly. If both eyes had similar near visual acuity, the eye with less scotoma on

the right side of the visual field was selected as the “reading eye”. This was


234

confirmed by the visual field testing described below. Best corrected distance visual

acuity was measured using the Bailey-Lovie distance visual acuity chart at 3.0 metres

scored to the nearest letter and recorded in logMAR (Bailey and Lovie, 1976). The

Bailey-Lovie word chart was used to measure both subjects’ near visual acuity at the

working distance equivalent to their habitual near additions and the print size

achieved with their own STMs (Bailey and Lovie, 1980). The charts were

illuminated by overhead fluorescent tubes and the illuminance ranged from 300 to

400 lux. Details of the Bailey-Lovie distance and near acuity charts have been given

in sections 3.3.2 and 2.3.2.2 respectively.

The central 250 visual field was assessed monocularly using a tangent (Bjerrum)

placed at 1 metre (Henson, 1993) and contrast sensitivity was assessed by the Pelli-

Robson chart (Elliott et al., 1990 (b)). Measurements were conducted for the 'reading

eye'. For people with similar vision in each eye, the visual field was assessed

monocularly for each eye. Procedures of the assessments and the recording methods

have been described in sections 4.3.2.

5.3.2 Field of view

Field of view of the STMs was measured using N12 print size. This print size was

the same size as the reading passages used in the reading assessments. As the object

distance of the STM is usually shorter than the focal length of the lens, a change of

working distance between eye and the lens surface (eye-to-lens) alters the equivalent

power (or EVD). In addition, there is an inverse relation between the eye-to-lens

distance and the field of view. To ensure the same reading environment was provided

in measuring field of view and reading performance with STMs, the eye-to-lens

distance was the same in both measures. In measuring the field of view, subjects

were asked to count the total number of characters they could read clearly (with no

distortion) through the magnifier with their “reading” eye while the magnifier was

kept stationary. The number of characters reported was the field of view achieved

with STM.


235

5.3.3 Design of line guide

Some types of STMs supplied by Eschenbach (Table 5.2) contain a specially

illuminated reading line (transparent) or a moveable red reading line for highlighting

one line of printb. Eschenbach states “these accessories allow easy orientation on the

reading materials which substantially improve reading rate”b. However, these designs

are only available in some of the series of Eschenbach illuminated STMs. There are

no commercially available STMs which include a reading guide (line guide) across

the whole range of magnification. Therefore, a modified reading line was designed

for this study, which could be attached to the base of all STMs, so that subjects could

use their habitual STM and there was no limitation on the level of magnification.

Table 5.2 Eschenbach illuminated stand magnifiers

Descriptions Labelled dioptric Power (D)

Labelled magnification (X)

Lens diameter (mm)

Biconvex lenses 5 2.2 119 × 67 Biconvex lenses 10 2.5 119 × 94

105 x 80 mm (+4D swivel lens) 7 2.8 119 × 67

100 × 50 mm (+4 D swivel lens) 7.6 3 71

Aspheric 60 mm round * 12 3 71 Aspheric 70 mm round * 16 4 88 Aspheric 60 mm round * 20 5 71 Aspheric 50 mm round 23 6 62 Aspheric 35 mm round 28 8 52 Aspheric 35 mm round 38 10 52 Aspheric 35 mm round 50 12.5 52

* The stand magnifier has pivotable reading line (supplied by Eschenbach) (Refer to

Table 4.5 for the optical parameters of the stand magnifiers).

As one of the overall goals of this research was to evaluate rehabilitation methods

that could easily be implemented in any low vision rehabilitation clinic, the design of

the modified reading line, or “line guide”, was simple, inexpensive and based on a

line guide that had been developed and used by a low vision trainer (Ross Still,

personal communication) who provided home training for people who were visually

b Eschenbach Catalogue 2001/2002


236

impaired. R. Still (personal communication) reported that, from his clinical

experience with low vision patients, many STM users found the line guide to be

useful. The design of the line guide used in this study was the same as the

“prototype” used by R. Still (personal communication), comprising a white line with

a black fixation patch that could be easily attached and removed from the base of the

STM by adhesive tape. The rationale for this line guide was that it would help the

magnifier users to concentrate on the line being read and direct their fixation to each

group of words in turn as they moved the magnifier along the line they were reading.

The line guide was made from a white plastic strip with a black fixation patch in the

middle of the strip. Based on R. Still’s design (personal communication), the width

of the line guide was 10 mm (which covered two lines of N12 print), while its length

was determined by the horizontal diameter of the STM. Neve (1989 (b)) showed that

people mostly used only the central part of the vertical field when they used their

magnifiers to read. Therefore the vertical position of the line guide was standardised

for every subject with its upper edge placed along the mid-line of the STM (Figure

5.1) so that the line being read was visible, but the next two lines of text were

blocked. The black fixation patch was rectangular with a width of 10 mm but the

length varied from 6 to 20 mm depending on the field of view of each subject. The

length of the black patch was set so that it was approximately 30% of the measured

field of view of the STM for each subject. Based on R. Still’s practical experience

(personal communication) and pilot measurements with patients who were visually

impaired, this length was found to be adequate to draw subjects’ attention to the

location of reading.


237

Figure 5.1

The design of the line guide attached on the stand magnifier.

The line guide was made from a white plastic strip with a black fixation patch in the middle of the strip. The width of the line guide was 10 mm and its length was determined by the horizontal diameter of the stand magnifier. The black fixation rectangular patch was 10 mm in width but the length varied from 6 to 20 mm depending on the field of view of each subject.

5.3.4 Measurement of magnifier movements

The movement of the STM was recorded using a 3 SPACE Isotrak system (Polhemus

Navigation Sciences Division, Kaiser Aerospace, Vermont, USA). The Isotrak

comprises a system electronics unit (SEU), an IBM personal computer, one source

and one sensor (Figure 5.2). This instrument was used to measure the position and

orientation of a sensor in relation to a source in three-dimensional space, providing

six degrees of freedom (Polhemus 3SPACE Isotrak User's Manual, 1987; Percy and

Hindle, 1989; Trott et al., 1996). The source, which generated a low-frequency (10

Line guide of 10 mm width, which blocks the next two lines of text

Black fixation patch of 10 mm width and 6 mm length which highlights a group of 3 characters.


238

Hz) electromagnetic field, was securely attached to the rear of the reading stand. The

X, Y and Z planes of the source were aligned to the relative X, Y and Z planes of the

reading stand. The sensor, which was used to detect the low-frequency magnetic

field generated by the source, was firmly mounted onto the handle of STM such that

it did not interfere with subjects grip of the magnifier. It was located as close as

possible to the magnifying lens. This was to eliminate as much as possible any

relative skewing of the magnifier orientation that could introduce artefacts in

magnifier movement. Data collected from the sensor were transferred to the

electronic unit and recorded by an IBM computer reading the position of the

magnifier at a frequency of 10 times per second (10 Hz). Before the experiment

started, calibration of the instrument was conducted to assess the linearity and

sensitivity (resolution) of the equipment. The resolution of the Isotrak was better than

2 mm across the area where the magnifiers moved during the experiment. This

finding was better than the resolution claimed by the manufacturer (Polhemus

3SPACE Isotrak User's Manual, 1987) of 2.2 mm. An example of calibration traces

made in 2 mm steps is shown in Figure 5.3 and the procedures of calibration are

given in Appendix 6.

5.3.4.1 Analysis of magnifier movements

Magnifier movements (X, Y and Z) were recorded by the Isotrak. Movement in the

X-plane indicated the forward (horizontal) movement along the line of text while

movement in the Y-plane showed the vertical movement of the magnifier between

lines. Movement in the Z-plane represented the distance of the magnifiers from the

page on the reading stand. As subjects were instructed to place their STMs on the

reading stand during reading, movement in the Z-plane did not change and so those

data were not analysed. Movements of the STM position during reading were

analysed using programs (Excel and Matlab) that displayed the horizontal movement

(magnifier along X-plane: left to right position on page) and the vertical movement

(magnifier along Y-plane: vertical position on page) as a function of time (Figure

5.4). In some circumstances, it was difficult to decide whether a subject searched for

the beginning of a line or started reading a new line. In order to resolve this

ambiguity, a scatter plot (x, y) of the position of the STM on the text was included

(Figure 5.5).


239

Figure 5.2

3-SPACE Isotrak system.

The Isotrak comprises a system electronics unit, an IBM computer, one source and one sensor. Data collected from the sensor were transferred to the electronic unit and recorded by an IBM computer reading the position of magnifier at a frequency of 10 Hz.

Figure 5.3

Calibration traces of 2 mm steps movement (x-direction)

The sensor was used to detect the low-frequency magnetic field generated by the source. It was mounted onto the handle of the stand magnifier where it did not interfere with subjects holding the magnifier firmly for reading.

The source generated the low-frequency (10Hz) electro-magnetic field. It was securely attached to the rear of the reading stand.

The unit was connected to an IBM computer

30

30.5

31

31.5

32

32.5

33

33.5

34

Mov

emen

t of s

enso

r (cm

)


240

In order to define the exact time when the subjects started and finished reading with

their STMs, the STM movement recordings for the first and last lines read were

excluded from the analysis (Bowers, 2000 (a)). The magnifier movement parameters

as well as the movement strategy were then quantified and categorised for the

remaining 10 lines for both forward and retrace movements. The starting point of the

analysis (second line) was identified as the first point with a clear forward horizontal

movement at the start of the second line whilst the last point of the analysis was the

termination of the retrace movement after the 11th line, just before the start of the

forward movement at the start of the 12th line.

5.3.4.2 Magnifier movement parameters

In the analyses of the magnifier movement parameters, it was crucial to locate which

was forward and retrace movement in order to calculate the distance and the time for

forward and retrace. These parameters were derived from the plots of magnifier

movement in the X-plane (Figure 5.4). The upper trace of Figure 5.4 shows the saw-

tooth movements of the STM across the page in the X-plane when subject 27 read

the first three lines of the passage.

Figure 5.4 comprises forward movements (A to F, B to G and C to H) and retrace

movements (F to B, G to C and H to D) with the peaks and troughs indicating the

start and end of lines respectively. The forward movement was slower than that for

the retrace as cognitive analysis of the information from reading was involved in the

forward but not the retrace movement for any line. The start of the forward phase

was defined as the first clear rightwards movement at the start of a line, which meant

at the end of each of the peaks in the X-plot (e.g. A). The start of the retrace was

identified as the first clear leftwards movement at the end of a line, which meant the

end of the valleys in the X-plot (e.g. F). Forward time segment was defined as the

time difference between the beginning (A) and the end of the forward movement (F)

while the retrace time segment was the time difference between the start (F) and the

end of retrace movement (B). The average forward and retrace times for a line were

calculated. Total reading time was the sum of the forward and retrace reading times.

In addition, forward and retrace distances (horizontal distance moved by the


241

magnifier) were calculated based on the same definition used to determine the

forward and retrace time segments.

Horizontal stand magnifier movement (x) - Subject 27

17

19

21

23

25

27

29

31

33

Time (seconds)

Hor

izon

tal p

ositi

on o

f mag

nifie

r in

text

(c

m)

Vertical stand magnifier movement (y) - Subject 27

-2-1.5

-1-0.5

00.5

11.5

2

Time (seconds)

Vert

ical

pos

ition

of

mag

nifie

r in

text

(cm

)

Figure 5.4

Traces of movements of a stand magnifier in horizontal and vertical directions during reading.

Horizontal movements (X-plot) across the page are shown in the upper graph while vertical movements (Y-plot) down the page are shown in the lower graph.

Forward movement: AF, BG, CH Retrace movement: FB, GC, HD

R was the regression made in forward movement while T was the regression made in retrace movement.

CB D

G H

TR

A

F

0 10 20 30 40 50 60 70

LEFT

RIGHT

0 10 20 30 40 50 60 7 0


242

Some subjects had difficulty finding the beginning of the next line. This is illustrated

for subject 27 by the ‘flat’ section of trace before the point C in the X-plot in Figure

5.4. A retrace movement is initiated at point G which moves the magnifier to the left

hand side of the page in a single movement, however there is then a period of about 6

seconds (the ‘flat’ section) during which there is little change in X-position before

the forward movement is clearly started at point C. Examination of the Y-plot reveals

several movements in the Y-plane during the time period corresponding to the ‘flat’

section on the X-plot. Subject 27 moved the magnifier back to the left hand side of

the page and then had difficulty finding the beginning of the correct line and moved

the magnifier up and down in the Y-plane at the left hand side until the correct line

was found. In determining the retrace time for each line the time taken up by any

regressive movements during retrace and difficulties in positioning the magnifier at

the start of the line were included in the retrace time. Hence the reason for defining

the start of the forward movement (i.e. end of the retrace movement) as the first clear

rightwards movement.

5.3.4.3 Categorisation of magnifier movement

The magnifier had to be moved along the line and retraced to the beginning of the

next the line in order to complete one “navigation”. The strategy of magnifier

movement that subjects used in reading was categorised from the X-Y plots of

magnifier movements (Figure 5.5). In general, the magnifier movement was

differentiated into three main classifications for the forward and retrace phases –

straight, downhill and uphill. This categorisation was performed to identify which

strategy was preferred by subjects with low vision when manipulating magnifiers

during reading. Based on the general perception of the main magnifier movements

shown on the X-Y plot, any significant change in vertical position along the line

(forward and retrace) was ranked as a non-straight movement, either uphill (diagonal

upward) or downhill (diagonal downward) depending on which point – the start or

the end point on the same line - had a relatively higher position.


243

Line plot of stand magnifiers Subject 12 -with the line guide

-7

-6.5

-6

-5.5

-5

-4.5

1517192123252729

Horizontal distance (cm)

Vert

ical

dis

tanc

e (c

m)

Line plot of stand magnifiers Subject 12 -without the line guide

-7

-6.5

-6

-5.5

-5

-4.5

Horizontal distance (cm)

Vert

ical

dis

tanc

e (c

m)

Figure 5.5

Traces of a stand magnifier movement during reading.

For subject 12 the application of the line guide onto the STM changed the forward movement from downhill (top line plot) to straight (lower line plot).

29 27 25 23 21 19 17 15

LEFT RIGHT

LEFT RIGHT


244

Most participants were elderly people with visual impairment, therefore it was

reasonable to allow some trembling when they manipulated the magnifier in reading,

which resulted in a small vertical movement across the line. As a consequence, if the

change of the vertical position in magnifier movement between the start and the end

across a line was less than 3 mm, this movement was classified as straight.

Otherwise, the movement was determined as either uphill or downhill movement.

The reason to choose 3 mm as the criterion to differentiate the movement into

straight or not was related to the vertical distance between the centres of adjacent

lines (refer to Appendix 6). Twelve-point print (N12) was selected as the print size of

the reading passages with single line spacing, so the vertical distance between lines

was 5 mm. Therefore, it was reasonable to state that any vertical movement that was

more than half of the vertical distance between lines was no longer a straight

movement.

For example, the x- and y-coordinates of the starting point of a line when one subject

used his STM for reading were 29 and -6.5 cm; however, the coordinates changed to

18.8 and -7.1 cm respectively at the end of the same line. This implied that this

subject, instead of moving his STM straight, tended to move his STM slightly

downward (downhill movement) in the forward phase, as vertical position of the

starting point was higher than that of the end point by 0.6 cm. Then he moved the

STM straight to the beginning of the next line with no apparent vertical shift.

Comparing the y-coordinates of the end point of the previous line and the start of the

new line, the vertical shift was 0.25 cm. This suggested that this subject preferred to

move his STM in a straight pattern for retrace. In summary, the movement strategy

for this subject was downhill for forward phase and straight for retrace. For subjects

who did not consistently use the same strategy for all lines, the classification was

based on the strategy used for 50% or more of lines.

Table 5.3 summaries the classification of the magnifier movements for forward and

retrace phases. In the retrace phase, in addition to the three main categories, there

were some sub-categories of magnifier movements which reflected the specific

strategy that people used in retrace that they adopted to manipulate the STM. For

example, some subjects moved the magnifier in a diagonal downhill direction until

the middle of the retrace and then moved the magnifier in an uphill direction to the


245

start of the line. This type of retrace movement was categorised as diagonal

downward movement with a V-shaped pattern. Also, some magnifier users preferred

to retrace the magnifiers in a straight movement along the current line (the line that

was read) and then moved down to the start of the next line or moved the magnifier

down at the end of the current line and then retraced straight along the next line.

These specific strategies for retrace were classified as straight movement in the

retrace together with a vertical movement. The vertical movements related to these

specific straight retrace strategies were considered to be part of the retrace strategy.

The strategy related to vertical movements were not counted as navigational errors.

However, other Y-corrective movements that commonly occurred at the end of

retrace movements were considered as navigational errors and were counted in the

error analyses (see below).

5.3.4.4 Navigation errors

In addition to determining forward and retrace times and categorising magnifier

movements, other magnifier movement parameters known as navigational errors

were quantified according to the method of Bowers et al. (2002 (a)). This included

number of missed or repeated lines, the number of pauses, regressive movements,

corrective vertical movements, overshoots and undershoots during the forward and

retrace phases. In the forward reading phase, subjects occasionally stopped or even

moved back (regressed) if insufficient information was retrieved. These errors were

defined as pauses and regressions respectively. If a subject could not fixate on the

same line of text and moved in between lines, vertical movement was necessary to

bring the magnifier back on track. Any vertical movement to bring the magnifier

back to the correct line was coded as “forward Y-correction”. In the retrace

movement, regressive manoeuvres or vertical movements between lines to locate the

beginning of the next line to be read were all classified as navigational errors –

regression, pauses and regressive Y-corrections. In addition, any overshoots or

undershoots during retrace were counted as navigational errors. An overshoot was

defined as magnifier movement to a position where it was beyond the start of the

next line (end of retrace). This resulted in compensatory movements in the opposite

direction to bring the magnifier back to the correct position to read the line.

Undershoot was when the magnifier was not moved far enough to the start of the


246

line, which resulted in compensatory movements in the same direction to bring the

magnifier to start of the line. The definitions of each of the magnifier movement

parameters which were classified as navigational errors are summarised in Table 5.4.

The magnifier movement plots from which the errors were identified are also given.

Table 5.3 Categorisations of magnifier movements

Types of movement Description Difference in

Y-coordinate * Schematic diagram

Forward phase

Straight Movement was mainly horizontal ≤ 0.3 cm

Downhill Movement was mainly diagonal downward > 0.3 cm

Uphill Movement was mainly diagonal upward > 0.3 cm

Retrace phase

Movement was mainly horizontal ≤ 0.3 cm

• Move downward at the start of retrace

• Straight retrace along the next line

≤ 0.3 cm

Straight

• Straight retrace along the line • Move downward at the end of

retrace to the next line ≤ 0.3 cm

Movement was mainly diagonal downward > 0.3 cm

• Diagonal downward retrace • Move upward at the end of

retrace to the next line > 0.3 cm

• Diagonal downward retrace • Move downward at the end of

retrace to the next line > 0.3 cm

Downhill

V shaped (Move diagonal downward and then

diagonal upward) > 0.3 cm

Movement was mainly uphill > 0.3 cm

• Diagonal upward retrace • Move downward at the end of

retrace to the next line > 0.3 cm Uphill

Inverted U-shaped > 0.3 cm

* The difference of the Y-coordinate between the start (left-hand-side) and the end (right-hand-side) along one horizontal line.


247

Table 5.4 Definitions of navigation errors

Plot Types of errors Magnifier movements

X Pauses Stationary period of at least 2 seconds

X & XY Regression Reversed horizontal movement (from right to left) with more than 5 mm

Forward movement

Y & XY Y correction A vertical corrective movement of more than 3 mm between lines

X & XY Overshoot

Horizontal movement of 10 mm or more beyond the start of line with a

compensatory movement in the opposite direction

X & XY Undershoot

Horizontal movement of 10 mm short of the start of line with a

compensatory movement in same direction

X Pauses Stationary period of at least 2 seconds

X & XY Regression Reversed horizontal movement (from left to right) with more than 5 mm at

the start or the end of a retrace

Retrace movement

Y & XY Y correction Vertical corrective movement of more than 3 mm

5.3.5 Reading passages

Passages used in this study were similar to the passages used in Chapter 4 but the

print layout was slightly different such that the new passages all had a similar

number of characters and were of the same font size. Text for the passages was

selected from Oxford Progressive English Readers- Grades 3 (1995) and 4 (1995),

reproduced in N12 print size. Reading ability required was below sixth grade reading

level; this was analysed by the Flesch-Kincaid Grading Level System. The passages

contained 1044.8 ± 27.14 characters (174.13 ± 4.52 standard words) and were

divided into 12 lines. The length for each line was 15 cm. Six different passages were

printed across A4 cards with landscape orientation on a laser printer in black on

white print (Figure 5.6). Text of passages was left justified with single-spacing

between lines.


248

He climbed down, and came to the wall which went round the valley. He could see a number of men and women resting on piles of grass in the meadows, and nearer the village some children were lying on the grass, asleep. Closer to him three men were carrying buckets along a little path that ran from the boundary wall towards the houses. They walked slowly, one behind the other, like men who had been working all night. After waiting for a moment Peter went forward and gave a loud shout that echoed round the valley. The three men stopped and moved their heads as though they were looking round them. But they did not appear to see him. After a while Peter shouted again, and then once more. And again the word 'blind' came into his thoughts. 'The fools must be blind,' he said to himself, and began to walk towards them. The three stood side by side, not looking at him, but with their ears turned towards him, judging him by his unfamiliar steps. They stood close together like men a little afraid.

Passage 1 (Number of characters: 1001 characters)

Figure 5.6

Example of a reading passage for reading assessment


249

5.3.6 Procedures

Reading rates with STMs for single sentences were measured using Bailey-Lovie

text reading charts (refer to section 2.3.1) to determine the critical print size (CPS)

for each subject. This assessment was used to confirm that the print size of passages

(N12) used for this study was equal to or larger than the CPS for each subject, so that

print size would not affect reading rate. Reading rates and magnifier movements

were then measured as subjects read passages using their STMs, with and without the

line guide attached. For each reading measure, reading errors and number of lines

missed per passage were recorded.

Each subject completed two passages for each condition, alternating the order of

conditions between “with” and “without” line guide for each successive subject.

Subjects were given a few minutes to become familiar with the line guide attached to

the STM by reading through practice passages. Simple instructions about how to use

a STM with a line guide attached were given to subjects. They were asked to rest the

STM completely on the reading stand and to keep the line guide underneath the line

they were reading as they moved the magnifier forwards along the line. However, so

that spontaneous retrace strategies could be examined, they were not given any

instructions about how to move the magnifier with the line guide from one line to the

next line during retrace. Four out of six passages were selected for testing and the

magnifier movement recordings were stored in a disk file for later analysis.

Procedures of reading assessment and the analyses of reading measures have been

described in section 2.3.2.3.

5.3.7 Questionnaires

In addition to the objective assessment of reading performance, subjects’ subjective

assessment of reading with and without a line guide attached to the magnifier was

determined using a simple questionnaire. The questionnaire (Appendix 5) was not a

validated questionnaire but a modified version of the Manchester Low Vision

Questionnaire (MLVQ) (Harper et al., 1999). It addressed the frequency and length

of time for which the STM had been used, reading tasks read with the STM,

difficulties in manipulating the STM and the preference for reading with or without


250

the line guide attached to the base of the STM. This questionnaire was administered

on completion of all the reading assessments.

5.4 Analysis


version 10. As vision measures, reading rates, forward and retrace times, errors in Y-

corrective movements in retrace were not significantly different from a normal

distribution (Kolmogorov-Smirnov Goodness of Fit test, p>0.1), parametric statistics

were used - Pearson's correlation and paired-sample t-tests. As some of the measures

of navigation errors such as the numbers of overshoots, undershoots, regressions and

lines missed did not conform to a normal distribution (Kolmogorov-Smirnov

Goodness of Fit test, p<0.05), navigation errors (except the Y-corrective movements

in retrace) were analysed using non-parametric statistics – Wilcoxon Signed Ranks

test and Spearman correlation test. A probability of less than 0.05 was taken to

indicate statistical significance for all analyses. However, because of the number of

statistical tests conducted in this study, the probability of finding a significant result

by chance (type I error) is increased (Keppel, 1991.). In order to control the

magnitude of type I error, a Bonferroni correction was applied to the probability

associated with each test by dividing it by the number of tests executed. The number

of t-tests (35) exceeded the degrees of freedom for the variables (28), so 0.00143 was

the probability level for significance. Additionally, probability values of 0.00167

indicated significance when correlation tests were conducted.

For the reason outlined in section 3.4 and section 4.4.1, reading rate was transformed

to the logarithm (log) of reading rate. After data transformation, log reading rates

were not significantly different from a normal distribution (Kolmogorov-Smirnov

Goodness of Fit test, p=0.1). There were no statistically significant differences in the

measures for the two trials with and the two trials without line guide (p>0.1);

therefore, the data used in the analyses was the average of the two trials. Reading rate

described in the results and discussions of the current study always refers to log

reading rate.


251

5.5 Results

5.5.1 Comparison of navigation and reading performance using stand magnifier

with and without line guide

Table 5.5 summarises the mean results for the reading measures, magnifier

movement parameters and navigational errors made by the low vision and normal

vision subjects using STMs with and without a line guide.

Log reading rates when using a STM with a line guide were significantly reduced

compared to log reading rates without a line guide (paired t-test, t=4.3, df=28,

p<0.001). Although the effect of the line guide on reading performance was

relatively small - it reduced the reading rate by only 5 wpm (10%) - it was

statistically significant. The time since the STM had been prescribed for home use

had no significant impact on the change in reading rate with and without the line

guide (repeated measures ANOVA, F1,27=1.16, p=0.29) (Figure 5.7). In contrast,

there were no significant differences in the number of reading errors (t=0.69, df=28,

p=0.5) and lines missed with and without the line guide (Wilcoxon Signed Rank test,

z=0.73, p=0.46).

The reduction in log reading rate with a line guide was in part due to the significant

increase in distance that the magnifier was moved during the forward (paired t-tests,

t=5.44, df=28, p<0.001) and retrace phases (t=5.5, df=28, p<0.001). Although the

increase in both forward and retrace distances in turn did not significantly increase

forward (t=2.13, df=28, p=0.04) and retrace times (t=2.92, df=28, p=0.007), the

combined effect of increased forward and retrace times was the main reason for the

significant reduction in log reading rate. There was no significant difference in the

rate of forward movement (total forward distance divided by forward time) with and

without a line guide (t=0.22, df=28, p=0.83). Therefore, the magnifier was moved at

the same rate in the forward direction with and without a line guide, but the increase

in x-distance (forward distance) moved with the line guide increased the actual

reading time. For the retrace movement, the rate of moving the magnifier with a line

guide was slower than that without (t=2.26, df=28, p=0.03) although it did not quite

reach the significance level (after Bonferroni adjustment). This implies that some


252

other factors in addition to the increase in retrace distance contributed to the increase

in retrace time. One of these other factors was the significant increase in the number

of Y-corrective movements (t=4.0, df=28 p<0.001) which were made at the end of

the retrace movements with the line guide compared to retrace without the line guide.

The extra retrace corrections could be partly explained by the smaller vertical field of

view resulted in the attachment of the line guide. This indicates that subjects had

more difficulty finding the correct vertical position for the magnifier at the start of

each line when the line guide was attached. There were no significant differences in

the other forward or retrace navigation errors with and without the line guide

(Wilcoxon Signed Rank test, z<1.9, p>0.06).

Table 5.5 Descriptive statistics of navigation and reading measures using stand magnifier with and without a line guide

Without line guide With line guide P value*

Log reading rate 1.77 ± 0.26 log wpm 1.73 ± 0.27 log wpm <0.001

Forward time 15.58 ± 11.15 sec 16.86 ± 11.71 sec 0.04 Forward distance 12.1 ± 1.69 cm 12.85 ± 1.37 cm <0.001

Forward rate 1.07 ± 0.52 cm/sec 1.054 ± 0.53 cm/sec 0.83 Retrace time 4.06 ± 2.22 sec 5.07 ± 3.01 sec 0.007

Retrace distance 12.07 ± 1.71 cm 12.88 ± 1.31 cm <0.001 Retrace rate 3.72 ± 1.61 cm/sec 3.2 ± 1.28 cm/sec 0.03

Percent retrace time 22.67 ± 8.87 % 24.8 ± 7.52 % 0.17

Reading errors 33.81± 39.86 characters

38.47 ± 39.23 characters 0.50

Reading performance

Number of lines missed 0.24 ± 0.34 lines 0.26 ± 0.37 lines 0.46

Forward regression 4.67 ± 6.3 4.0 ± 5.23 0.37

Forward pauses 1.22 ± 1.55 1.5 ± 2.05 0.26 Forward Y- corrective movement

1.5 ± 1.82 0.95 ± 1.53 0.06

Overshoot 0.12 ± 0.34 0.12 ± 0.39 1.0 Undershoot 0.12 ± 0.22 0.03 ± 0.13 0.10

Retrace regression 1.66 ± 1.8 1.91 ± 2.55 0.84

Navigation errors (in terms of

number per passage)

Retrace Y- corrective movement

2.59 ± 2.53 4.02 ± 3.46 0.002

* The adjusted probability for significance was 0.00143 (see section 5.4).


253

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

Time since prescription (months)

Cha

nge

in lo

g re

adin

g ra

te (w

pm)

Figure 5.7

Comparison of the change in log reading rate due to the use of line guide and the length of STM use (log-scale). Positive direction on the Y-axis represents an increase in log reading rate with use of the line guide while the negative direction represents a reduction in log reading rate when the line guide was used.

The percent retrace times, which were the proportion of reading time devoted to

retrace, were compared for subjects reading with the STM with and without the line

guide. Mean percent retrace time without the line guide was not significantly

different from that with the line guide (t=1.4, df=28, p=0.17). This indicates that the

line guide attached to the STM did not alter the proportion of reading time used for

retrace, as both retrace and forward time increased when the line guide was used.

5.5.2 Magnifier movement strategy

Without the line guide attached, the majority of subjects (62.1%) moved their STMs

downhill (diagonal downward) while the others (37.9%) moved their magnifiers in a

straight line when they read along the line (forward movement) (Table 5.6). Log

reading rate without line guide did not differ significantly among subjects using these

two different forward movement strategies (one way ANOVA, F1,27=0.24, p=0.63).

The introduction of the line guide significantly changed some subjects’ movement

strategy from downhill to a straighter forward motion (Chi-square, α2=5.63, df=1,

p=0.021). An example of this change in movement strategy is shown for subject 12

(Figure 5.5). For retrace without the line guide, downhill movement was the most

0 1 10 100


254

common strategy used by low vision subjects (55.2%) followed by straight back

(24.1%) and uphill or inverted U-shaped movement (20.7%). Interestingly, magnifier

reading rate for subjects who manipulated the magnifiers diagonally downward was

significantly faster than those who moved the magnifiers straight back (one way

ANOVA, F2,26=7.69, p=0.002). This implies that a diagonal downward retrace

movement might be more efficient for magnifier reading. The attachment of line

guide did not result in any significant change in the movement strategy in retrace

(α2=1.18, df=2, p=0.55). Therefore the line guide had more impact in changing the

strategy of magnifier manipulation for the forward movement than for the retrace

movement.

Table 5.6 Strategy of magnifier movements with and without a line guide attached

Forward movement Retrace movement

Line guide attached Straight

Downhill (Diagonal

downward) Straight

Downhill (Diagonal

downward) Upward

Without line guide 37.9% 62.1% 24.1% 55.2% 20.7%

With line guide 69% 31% 27.6% 62.1% 10.3%

5.5.3 Frequency of magnifier use and subjective feedback about navigation

difficulties

Reading newspaper and mail were the main reading tasks that the low vision subjects

performed with their STMs. These tasks occupied approximately 50% of all the

reading tasks for which subjects used their magnifiers each day. The majority of

subjects (62%) in this study used their STMs frequently, at least once a day. Only

one of twenty-nine subjects rarely used his STM and this was because he used his

closed-circuit television for reading. The majority of subjects (89%) read with their

STMs for up to 30 minutes duration. Very few subjects (11%) read for longer than

30 minutes at a time. There was a high correlation between frequency and duration of


255

magnifier use (Spearman correlation, r=0.65, p<0.001) which implied that the more

often the magnifier was used, the longer the duration for which it was used.

With regard to difficulties manipulating their STMs, approximately 70% of subjects

reported that they did not have any difficulty in moving the STM when reading along

a line (forward movement), but only 18% of subjects reported no difficulty with

retrace (Figure 5.8a). Two thirds of subjects stated that they had slight to moderate

problems with retrace when they used their STMs. Presumably because of this

difficulty, subjects sometimes missed lines of text while reading (Figure 5.8b). On

the basis of the length of time for which the STM had been prescribed, 12 subjects

(41.4%) were classified as “inexperienced” (magnifiers prescribed for less than 3

months). A higher proportion of these inexperienced subjects (50%) reported more

difficulty in manipulating their magnifiers, in particular during the retrace

movementc compared with that reported by “experienced” subjects (35.3%), however

this difference was not statistically significant (Chi-square, p>0.05).

Among the vision and reading variables, there were no significant correlations with

subjective difficulty in forward movement (Table 5.7). There was a moderately

strong correlation (r=-0.42) between monocular field of view (number of characters)

of the STM and the difficulty in retrace movement (p=0.02), but this was not

statistically significant with Bonferroni correction. This does suggest that subjects

using magnifiers with larger fields of view had less difficulty in retrace movement.

c Subjects who reported moderate to extreme difficulty in retrace movement were considered as those who had problems with retrace for analysis purposes.


256

0

10

20

30

40

50

60

70

80

no diff iculty slightlydiff icult

moderatediff icult

signif icantlydiff icult

extremelydiff icult

Perc

ent o

f sub

ject

s (n

=29)

Forw ardmovement

Retracemovement

Figure 5.8a

Difficulty in using stand magnifiers (subjective response reported by the subjects).

Approximately 70% of subjects reported they had no difficulty in moving the STMs during forward movement. However, the percentage who found no difficulty in retrace movement reduced to 18%.

0

10

20

30

40

50

60

veryfrequently

frequently sometimes rarely never

Perc

ent o

f sub

ject

s (n

=29)

Figure 5.8b

Frequency of missing lines in using stand magnifiers.

A large percentage of subjects reported that they sometimes missed a line during reading with STMs. This could be due to the difficulties manipulating the magnifiers in retrace.


257

Table 5.7 Spearman correlations between subjective response of difficulty in using magnifier and clinical measures without the attachment of line guide

Variables Difficulty in forward movement

Difficulty in retrace movement

Distance visual acuity 0.17 0.16 Near visual acuity 0.11 0.27 Log reading rate -0.25 -0.34

Change in log reading rate -0.02 0.40 EVD of the STM 0.16 -0.26 Visual field loss 0.17 0.18

Field of view (monocular) -0.04 -0.42 Length of visual impairment -0.24 -0.41

Vision and reading

Length of STM use -0.25 -0.34 Forward regression 0.03 0.24

Forward pauses -0.04 0.04 Forward Y corrective

movement 0.12 0.35

Retrace undershoot 0.21 -0.04 Retrace overshoot 0.05 0.17

Navigation errors

Retrace Y-corrective movement 0.13 0.29

With the adjusted probability for significance of 0.00167 (see section 5.4), these correlations were not significant.

5.5.4 Subjective preference for the line guide

Approximately 50% of subjects preferred to have the line guide and about 33%

preferred to have the black fixation patch to assist their reading with STM (Figure

5.9a). The reasons for selecting or refusing the attachment of the line guide and

fixation patch are summarised in Figures 5.9b and c. A higher proportion (58%) of

participants who had less experience (less than 3 months since the STMs were

prescribed) compared with 41.2% of experienced subjects preferred these accessories

to aid their reading. However, there was no significant relationship between the

reported usefulness of the line guide and fixation patch and either the length of time

since the magnifier was prescribed (α2= 1.71; df=4, p=0.27) or the frequency of

magnifier use (α2= 2.04; df=4, p=0.25).


258

0

10

20

30

40

50

60

70

Preferred Not preferred Line-guide Fixation Line-guide Fixation

patch patch

Perc

ent o

f sub

ject

s (n

=29)

Figure 5.9a

Subjective preferences for the line guide and fixation patch.

For the line guide, approximately half of the subjects preferred to have this reading assistance with their STMs. In contrast, the preference for the black fixation patch was less.

Easier to follow the line

20.8%

Provide better orientation

58.3%

Enhance concentration

16.7%

more confidence in

reading4.2%

Figure 5.9b

Reasons for preference for the line guide and fixation patch.

The majority of the subjects (58%) recognised that the line guide and fixation patch provided better orientation in reading with STMs. These reading assistances allowed them to follow the line they read more easily (21%).


259

Not necessary19%

Not working8%

Harder to read31%

Vertical field of view was

reduced 19%

Habitual reading

strategy was disturbed

23%

Figure 5.9c

Reasons for rejection of the line guide and fixation patch.

The two main reasons that participants rejected the line guide and fixation patch were the disturbance of the habitual magnifier-aided reading strategy caused by these devices and the introduction of the orientation assistance made the reading harder.

Subjective reports of difficulties manipulating the STM also had no significant

relationship with the preference for the line guide (α2= 4.8, p=0.09). However

Figures 5.10a and b do highlight a trend that is apparent in the data, namely that

subjects who preferred having the line guide as the reading accessory tended to

report greater difficulties in manipulating the STM compared with those who refused

having this reading accessory, particularly during the retrace movement. In contrast,

subjects who had reported difficulty in tracing along the line (forward reading) had a

significant preference to have the black patch as their reading guide (α2= 6.54;

p=0.04).


260

0

5

10

15

20

25

30

No difficutly Slightlydifficult

Moderatedifficult

Significantlydifficult

Extremelydifficult

Perc

ent o

f sub

ject

s (n

=29)

Diff iculty inforw ardmovement

Diff iculty inretracemovement

Figure 5.10a

Distribution of reported difficulty in using STM for subjects who preferred the line-guide.

0

5

10

15

20

25

30

35

40

45

50

No difficutly Slightlydifficult

Moderatedifficult

Significantlydifficult

Extremelydifficult

Perc

ent o

f sub

ject

s (n

=29)

Diff iculty inforw ardmovement

Diff iculty inretracemovement

Figure 5.10b

Distribution of reported difficulty in using STM for subjects who did not prefer the line-guide.


261

The subjective preference for the line guide did not directly reflect the outcomes of

the objective measures of navigation and reading performance. Although subjects

reported that they thought that the line guide was helpful, this was not associated

with a measurable improvement in reading or navigation performance. For subjects

who preferred the attachment of the line guide, there was no significant improvement

in objective measures (such as reading rate, retrace Y-corrective movement) when

using the line guide. Although the attachment of the line guide appeared to improve

the control of vertical positioning of the magnifier during the forward phase for

subjects who preferred the line guide, it was not statistically significant with

Bonferroni correction (Wilcoxon signed ranks test, z=-2.51, p=0.01). In addition, the

baseline vision and reading measures for the subjects in favour of the line guide were

not significantly different from those for the subjects who rejected the line guide

(Table 5.8, unpaired t-tests or Mann-Whiney U testd, p>0.1). Therefore, none of the

objective measures used in this study or subject’s baseline vision measures were

predictive of the preference for the line guide in STM use.

d Unpaired t-tests were used to compare variables which were not significantly different from normal distributions, such as log reading rate, distance and near visual acuities, EVD and field of view. Mann-Whitney U test was used to compare variables, such as length of STM use, length of visual impairment, which were significantly different from normal distributions.


262

Table 5.8 Comparison of objective performance measures using stand magnifier with and without a line guide for subjects who preferred and rejected the line guide

* The adjusted probability for significance was 0.00143 (see section 5.4).

5.6 Discussion

In this study navigation and reading performance were evaluated with and without a

simple line guide that could easily be attached to any STM. The white line guide and

black fixation spot were designed to help subjects with low vision to direct their

attention to the correct line and words by blocking the next line and assisting

orientation to the place where they were reading. Magnifier movement recordings

provided useful insights into magnifier manipulation strategies and navigation

performance with and without the line guide.

Preference for line guide (n=14)

Rejected the line guide (n=15)

Without line guide

With line guide

p-value*

Without line guide

With line guide p-value

Log reading rate (wpm) 1.73 ± 0.29 1.70 ± 0.30 0.051 1.81 ±0.24 1.75 ± 0.25 <0.001

Reading errors

(characters) 41.3 ± 47.56 16.69 ± 38.25 0.68 26.8 ± 31.12 30.8 ± 39.86 0.54

Forward regressions 5.68 ± 7.46 3.75 ± 4.13 0.17 3.73 ± 5.08 4.23 ± 6.22 0.94

Forward pauses 0.96 ± 1.37 0.82 ± 1.12 0.53 1.47 ± 1.72 2.13 ± 2.52 0.06

Forward Y-correction 2.0 ± 2.11 0.75 ± 1.05 0.01 1.03 ± 0.23 1.13 ± 1.88 0.79

Overshoot 0.11 ± 0.29 0.07 ± 0.18 0.70 0.13 ± 0.40 0.17 ±0.52 0.56

Undershoot 0.11 ± 0.29 0 0.08 0.13 ± 0.23 0.07 ±0.18 0.41 Retrace

regression 2.0 ± 2.09 2.11 ± 2.29 0.75 1.33 ± 1.47 1.73 ± 2.84 0.94

Retrace Y- corrective movement

2.79 ± 2.19 4.0 ± 3.32 0.025 2.4 ± 0.29 4.03 ± 3.71 0.01


263

5.6.1 Comparison of navigation and reading performance using stand magnifier

with and without line guide

In addition to the changes in forward manipulation strategy and the increase in

retrace navigation errors (Y corrections) made when looking for the start of the next

line, both forward and retrace distances and navigation times increased significantly

when the line guide was introduced. It is possible that the increase in forward

distance was due to the presence of the black patch in the middle of the line guide.

The main purpose of including the black patch was to direct the subjects’ fixation to

the group of words being read during the manipulation of the STM. Therefore it is

possible that, when using the line guide, subjects kept their eyes centred in the

middle of the magnifier lens above the black patch until they read the last group of

words on a line before the STM was moved to the next line (retrace). Comparatively,

the forward movement of the STM and the forward distance might be less if the

subjects viewed the last group of words through the edge of the magnifying lens

(rather than the centre) when no line guide was attached. The increase in forward

distance moved when the line guide was attached to the STM had two consequences,

firstly an increase in forward navigation time and secondly a corresponding increase

in retrace distance. The increase in retrace time with the line guide was a result both

of the increase in retrace distance moved and the increase in the number of vertical

corrective movements at the end of the retrace.

The increase in navigation times lead to an increase in overall reading time with a

consequent reduction in log reading rate. The reduction in log reading rate was small

but statistically significant, equivalent to about 5 wpm (10%). This result is similar to

the findings of a study by Fitz and colleagues (2000) who investigated the effect of

an electronic ‘highlighter’ superimposed on magnified text on the reading rate of

patients with AMD using a magnifier. Their results indicated that the user-controlled

electronic highlighter did not provide any clinically significant short-term

improvement in the reading rates of patients with AMD.

Kuyk and colleagues (1998 (a)) found that people with low vision read significantly

faster with the assistance of a stationary pointer and a mechanized reading stand

when using their optical LVAs. The discrepancy in findings is mainly due to the use


264

of the mechanized reading stand and the practice effect with the reading assistance in

Kuyk et al.’s (1998 (a)) study. This reading stand was an automated x-y viewing

table in a tilted position and provided to-and-fro text movements with speed control

and automated line indexing. Since the retrace navigation of the LVAs was managed

by the automated reading stand, the requirement for the users to manipulate the

magnifiers (both forward and retrace movement) was eliminated. The majority of the

subjects in the current study complained of moderate difficulty in navigating their

magnifiers especially for the retrace movement. If reading performance with the

mechanized reading stand (without retrace navigation) had been compared to that

with normal manipulation of the LVAs in Kuyk et al.’s study, the improvement in

reading rate may have been mainly due to the elimination of retrace by the

mechanized reading stand rather than the line orientation assistance of the reading

accessory (pointer). In addition, participants in Kuyk et al.’s study were given five

days to practise using the reading stand before the reading measures were taken.

However, subjects in the current study were only given a short time to practise using

the STM with the line guide until they reported that they were familiar with the use

of line guide in the laboratory.

5.6.2 Comparison of stand magnifier manipulation strategy with and without line

guide

Interestingly, when reading without the line guide the majority of subjects (62%)

moved the STM diagonally downward during the forward movement of the

magnifier along the line, rather than straight along the line. This could be due to the

significantly faster reading rate achieved with this manipulation strategy in retrace

movement. When using the line guide, subjects were instructed to put the white line

exactly under the line they were reading. As might be expected, this instruction

resulted in a modification of the diagonal downwards movement to a straight

movement by half of the subjects when the line guide was introduced. Figure 5.5

compares the movement of STM with and without a line guide. With the line guide,

it appears the positioning of the STM becomes more critical such that movement are

neater and straighter along the horizontal line. The results of this study suggest that


265

the addition of the line guide to the base of the STM changes the strategy of forward

movement regardless of level of experience in using the STM.

Although subjects were given very clear instructions about how to place the line

guide under the line that they were reading during the forward movement, they were

not given any specific instructions about how to move the magnifier with the line

guide attached during the retrace phase. The choice of strategy during retrace was

entirely decided by the subjects and it was of interest to examine whether subjects

would, in the short-term, modify their habitual retrace strategy when using the line

guide. Interestingly very few of the subjects changed their retrace strategy when the

line guide was attached to the magnifiers, despite the fact that the guide obscured the

two lines underneath the line that was being read. However, the magnifier movement

traces clearly demonstrated that the line guide (decreased vertical field) made it more

difficult for the subjects to determine the correct vertical position for the magnifier at

the start of the next line, resulting in an increase in vertical corrective movements (Y

correction) at the end of the retrace movement. This study showed that the common

retrace strategy (diagonal downward) without the line guide was still the best

strategy for retrace with the line guide since more efficient reading performance was

achieved (refer to section 5.5.2). Intuitively one would expect that a “straight” retrace

strategy would be best suited to the design of line guide used in this study (e.g.

retrace straight along the line just read and then move down to the next line). The

majority (76%) of subjects in this study habitually used a non-straight retrace

strategy without a line guide and had never had any formal instruction in using a

straight retrace strategy when their habitual STM was first prescribed. The common

retrace arc might be a function of natural pivoting actions involving the wrist, elbow

and/or shoulder, independent of active visual guidance. The findings suggest that

once subjects have developed their own retrace strategy, they do not modify it for the

line guide. Perhaps, a line guide similar to the one used in this study could be

introduced to new magnifier users who have not previously had any STM prescribed.

Specific instructions about using a “straight” retrace strategy could be given to these

patients such that their retrace movement was more suited to the use of the line

guide.


266

5.6.3 Subjective preference for the line guide

In low vision rehabilitation, the subjective response from patients on the use of their

LVAs is one of the important criteria that should be considered in order to achieve a

successful rehabilitation outcome. These subjective responses may provide

information that is not reflected in objective clinical measurements but provide

valuable information to interpret the rehabilitation outcome. In this study, half of the

subjects preferred to use the line guide to assist reading with STM, despite the

reduction in log reading rate and increase in navigation errors imposed by the line

guide. The reported preference for the line guide by the participants supports the

trend, which was not statistically significant (refer to section 5.5.4), that a higher

proportion of subjects with less experience using magnifiers preferred the inclusion

of the line guide. This suggests that the line guide may be a useful training tool for

new magnifier users.

Comparison of the subjective reports and objective analysis of navigation errors and

log reading rates indicated that the number of subjects who preferred the line guide

(subjective response) was more than the number who showed an improvement in

reading performance with the line guide (objective analysis). This suggests that the

subjective responses do not necessarily agree with the objective findings. Previous

studies have demonstrated that there are discrepancies between self-reported and

measured functions (Elam et al., 1991; Hickson et al., 1999; Friedman et al., 1999

(b)). Hickson et al. (1999) pointed out that more participants in their population

study of older Australians reported vision difficulties than had a measured

impairment. Similarly, Friedman et al. (1999 (b)) suggested that the substantial

discrepancy between self-reported reading difficulty and measured reading rate

among participants in their community based study may be due to different

underlying expectations and experiences. In addition, the objective measures might

not be sensitive enough to predict the subjective response or the reasons for the

participants preferring to have the line guide as their reading assistance. Also, other

factors that were not measured in this study may have influenced subjects’

preferences in selecting the line guide. One of the possibilities may be the

psychological support provided by the line guide on orientation. The major reason

that people chose to include the line guide was the better orientation provided, which


267

made them feel that it was easier to follow the line when they read. This accessory

provides sufficient support to users and increases their confidence by ensuring that

they are reading at the correct position. This may in turn improve people's reading

comprehension when they are reading passages, however comprehension was not

measured in this study. This type of support is important in achieving successful low

vision rehabilitation.

5.6.4 Design of the line guide

It is possible that modifications to the design of the line guide and more instructions

on how to use it could minimise some of the problems that have been identified from

the magnifier movement recordings as reasons for the reduction in reading rate when

the line guide was used.

As the upper edge of the line guide designed for this study was attached along the

mid-line of the STM and the strip was 10 mm wide, the vertical field of view with

the magnifier was greatly reduced. This could have contributed to the straighter

tracking of magnifier and extra retrace corrections. Previous studies have shown that

the vertical field of view should be no less than 3 lines to achieve maximum reading

performance, otherwise reading rate was reduced significantly (Duchnicky and

Kolhners, 1983; Neve, 1989 (a); Myrberg et al., 1996). Because of the reduction in

vertical field of view, some subjects rejected the line guide when they were asked to

comment on its usefulness. In addition the positioning of the line guide forced the

lines of text to be located above the horizontal midline (widest horizontal field).

Perhaps the line guide should be placed below midline, so the line of text being read

can be in exact centre of the field of view.

As the text from the next line was completely blocked by the line guide, more retrace

difficulties in reading were induced compared with reading without the line guide. If

subjects had been able to preview where the next line starts, the retrace problem

caused by the line guide may have been reduced. A transparent yellow line guide

rather than the opaque white strip may be more useful. The use of a transparent strip

could orientate readers to the line (and text) and allow them to preview the text as

they retrace to the line below. Although the majority of subjects in this study did not


268

find any advantage in the use of the black patch for reading, a pilot experiment

should still be conducted to evaluate the benefits of including the black patch in a

new design of the line guide. If the result from this further research shows no

significant improvement with the black patch, the new design of the line guide would

not include this accessory. Moreover, a reduction in width of the line guide to cover

1 line (e.g. 5 mm) rather than 2 lines may be better for reading materials of around

12-point print. Also, if such a line guide is supplied, instructions of a specific

“straight” retrace strategy could be included to suit the design of the line guide. For

example, at the end of the forward movement when using a magnifier with the line

guide, it is suggested that the magnifier should be moved vertically downward to the

next line followed by a horizontal movement along the new line to the start.

Alternatively, people could move the STM with the line guide horizontally backward

(from right to left) along the completed line to the start of the line before moving the

magnifier vertically downward to the next line.

5.6.5 Adaptation to the line guide

Irrespective of the frequency of magnifier use or length of time since the magnifier

was prescribed, all subjects were unfamiliar with the use of their STM with a line

guide attached as none of them had been given a line guide attachment when their

magnifier had been prescribed. Therefore, the reduction in reading performance

when they read with the line guide may possibly be due to their lack of experience

with this new accessory and insufficient time for them to adapt to using the line

guide. If someone had used a magnifier for a period of time and developed their own

manipulation strategies, it would be quite difficult to change their habitual way of

reading when a new feature is added to the magnifier.

The results from this study suggest that the short-term practice was insufficient to

allow the subjects to modify their manipulation technique such that it was suitable

for reading with their STMs when the line guide was attached. However, the effect of

long-term practice with the line guide was not included in this study. To address this

factor, further research, providing sufficient time for each subject to practise (e.g.

one or two weeks practice) with the use of the line guide attached to the STM, should

be conducted. The modified design of line guide should be used and instructions on a


269

specific retrace strategy to suit the line guide should be given. Navigation and

reading performance should be assessed before and after two weeks home practice

with the line guide attached to the subject’s STM. Additionally, a clinical trial is

essential to investigate the benefit of the line guide for new magnifier users. This

could a longitudinal study to compare reading performance for people who have the

line guide attached onto the STMs and those who have no assistance on the

magnifiers as well as a questionnaire addressing the difficulties of magnifier use.

From the results of this further study, the benefits of the line guide for magnifier

reading could be investigated more thoroughly.

Even taking into account manipulation strategy and experience in magnifiers use for

reading, some users may still have poor magnifier movements. Poor navigation

technique could be related to the pivoting actions of the wrist, elbow and/or shoulder.

The relationships between motor control, page navigation and magnifier

manipulation should be fully investigated.

5.7 Conclusion

Low vision people, in particular those who have less experience in using STMs,

often report problems with page navigation. Common reports are that they either

miss a line or take extra time to look for the beginning of the next line. In order to

relieve these problems, assistance that can direct the location as they read with their

magnifiers might be helpful. No significant improvement in reading performance

was found for the AMD subjects using STMs with a line guide attached in this study.

Reading rate was reduced and retrace navigational errors increased probably due to

the limitation of the vertical field of view, lack of preview of the next line induced by

the line guide and the short period of familiarisation with the line guide. Despite this

more than fifty percent of the subjects preferred to have the line guide attached onto

their own STMs. The line guide used in this study is inexpensive and can be fitted to

any STM. Simple modifications to the design and specific retrace instructions are

suggested to overcome some of the limitations identified. Results indicate the line

guide may be useful as a preliminary training aid when low vision patients with

AMD are first prescribed STMs as it did induce straighter forward reading


270

movement with the magnifier. Further investigations of the design and possible

benefits of a line guide in improving navigation performance are needed.

Chapter 6 Conclusion and recommendations

271

CHAPTER 6

Conclusions and recommendations

6.1 Introduction.................................................................................................. 272

6.2 Main findings................................................................................................ 272 6.2.1 Validation of the Bailey-Lovie text reading chart.................................. 272 6.2.2 Validation of the fixed acuity reserve method for prescribing

magnification for subjects with AMD ................................................. 273 6.2.3 Large print reading practice and reading performance with stand

magnifiers (Chapter 4) ......................................................................... 275 6.2.4 Line-guide gives no objective improvement in reading performance but is

preferred by more than fifty percent of subjects with AMD (Chapter 5).............................................................................................................. 277

6.3 Future research ............................................................................................ 279 6.3.1 Inclusion of subjects with different causes of low vision ...................... 279 6.3.2 Schedule for the review of reading performance with magnifiers......... 279 6.3.3 Training in page navigation with a restricted field of view ................... 280 6.3.4 Page navigation with magnifiers............................................................ 280 6.3.5 Improved design of the magnifier line guide ......................................... 281

6.4 Clinical recommendations ........................................................................... 282 6.4.1 Prescribe large print reading practice before prescribing magnifiers .... 282 6.4.2 Measure reading performance and near visual acuity at the first visit... 283 6.4.3 Calculate the required magnification ..................................................... 283 6.4.4 Select appropriate magnifier and assess reading performance .............. 283 6.4.5 Loan the magnifier and review performance before prescribing ........... 284 6.4.6 Follow up visit ....................................................................................... 285

6.5 Summary....................................................................................................... 285


272

6.1 Introduction

Low vision affects a person’s daily activities and quality of life. Of all daily

activities, difficulty with reading is the major disability reported by people who are

visually impaired (Elliott et al., 1997; Watson et al., 1997). Because of the

significant increase in the prevalence of low vision over the past twenty years and its

associated activity limitations, vision rehabilitation aimed at improving reading

ability for people with visual impairment is essential. Providing magnification to

enlarge the retinal image is one way to overcome a person’s reading disability.

This study was designed to address important issues for the effective prescription and

training of patients with visual impairment due to age related macular degeneration

(AMD) in the use of stand magnifiers for reading. The emphasis was on the

development of simple methods of assessment and training that could be easily

implemented, at no great cost, by practitioners in low vision clinical practice. The

specific objectives of the study were to examine the effects of large print reading

practice and use of a line-guide on reading performance with stand magnifiers. The

important findings related to the purposes of this study are summarised in section 6.2

and suggestions for further research as a result of this study are given in section 6.3.

In addition, clinical recommendations for vision rehabilitation for reading are

presented in section 6.4.

6.2 Main findings

6.2.1 Validation of the Bailey-Lovie text reading chart

An assessment of reading rate is one of the most common methods of evaluating

reading performance in clinical vision examinations for people with either normal or

low vision (Ahn et al., 1995). Therefore, a reading chart that gives reliable measures

of reading performance is necessary for clinical and research use. Recently, Bailey

designed two text versions (Bailey-Lovie text reading charts) of the Bailey-Lovie

near word acuity charts using sentences which were modified from the MNRead

charts which have been used extensively in reading research (refer to section 3.3.4).


273

However, compared to the MNRead chart, reading performance measured using the

Bailey-Lovie text reading chart is better related to real world reading because of the

left page justification of the sentences and the smaller size of the chart. These

features also make it more convenient for the experimental set up of a reading study.

However no validation study of the new Bailey-Lovie text reading chart had been

carried out. Therefore the repeatability of the reading parameters obtained from the

Bailey-Lovie text reading chart, compared to the reading measures obtained from the

MNRead chart were examined (Chapter 2). The reading measures included critical

print size (CPS) and maximum reading rate (MRR). CPS is the smallest print size at

which maximum reading rate can be achieved while MRR is the mean of the reading

rates at and above the CPS. There were no significant differences in the repeatability

of reading measures achieved with the Bailey-Lovie text reading and the MNRead

charts for young subjects who were normally sighted. This finding is not surprising

since the designs of these two reading charts are very similar. Because of the better

features of the Bailey-Lovie text reading chart, it was chosen as the tool to measure

reading performance in this thesis. For clinical reading assessment, the two

parameters of MRR and CPS are usually determined from one trial. The result of this

study supports this practice as these measures did not differ over repeated trials.

6.2.2 Validation of the fixed acuity reserve method for prescribing magnification

for subjects with AMD

Determining the appropriate magnification of low vision aids to assist people for

reading is an important task in low vision rehabilitation. In the past, practitioners

have used distance and/or near visual acuity to calculate the required magnification

of low vision aids (Kestenbaum and Sturman, 1956; Cole, 1993; Dickinson, 1998).

Magnification calculated this way usually under-estimated the final magnification

prescribed (Cole, 1993; Flom et al., 1993; Elam, 1997), because an acuity reserve

allowance for reading rate was not included in the initial calculation. Recently, there

have been two different approaches proposed to determine the appropriate acuity

reserve for calculating magnification (Legge et al., 1992; Whittaker and Lovie-

Kitchin, 1993; Ahn et al., 1995; Lovie-Kitchin and Whittaker, 2000). These include a

fixed acuity reserve of 0.3 log unit as described by Lovie-Kitchin and Whittaker

(1999 (b); 2000) or an individual determination of optimum acuity reserve as


274

described by Legge and colleagues (Legge et al., 1992; Ahn et al., 1995). The fixed

acuity reserve method is relatively quick and simple to apply compared to the

individual acuity reserve method as only near visual acuity and print size of the

target reading task are required. The two methods are described in detail in section

3.1.4. Although the fixed acuity reserve method was first suggested in 1993 (based

on a review of the literature (Whittaker and Lovie-Kitchin, 1993)), the method had

not been validated. Therefore the aim of this study was to determine whether the

fixed acuity reserve method gave adequate magnification for maximum reading rate

for AMD subjects and to compare the magnification and reading rates with those

obtained by the individual acuity reserve method.

There were no significant differences in the equivalent viewing distance (EVD) (i.e.

magnification) or reading rates achieved with the low vision aids determined by the

fixed acuity reserve and the individual acuity reserve methods (Chapter 3). Therefore

the fixed acuity reserve method of 0.3 log unit was used for the prescription of

magnifiers in the main experiment (Chapter 4), and is recommended for use in

calculating the appropriate magnification for low vision aids in clinical practice. The

reading performance achieved with the magnification calculated by this fixed acuity

reserve method was as expected for 30 of the 32 experimental subjects with AMD

(Chapter 4). For two subjects, the reading rate achieved with the magnification

calculated by the fixed acuity reserve method was not satisfactory, i.e. the reading

rate with the new low vision aid was reduced compared to the reading rate for large

print without a low vision aid. In this case, the individual acuity reserve method was

used to modify the magnification prescribed. The individual acuity reserve method

requires an assessment of reading rates at different print sizes to determine individual

acuity reserve required for fluent reading rate and to calculate the appropriate

magnification (Ahn and Legge, 1995); 0.5 log units acuity reserve was required for

each of these two subjects.

This finding agrees with the clinical recommendation of Lovie-Kitchin and

Whittaker (2000) that a fixed acuity reserve of 0.3 log units (3 lines) for fluency

should be used to determine the tentative magnification. If the reading rate with this

magnification is noticeably reduced compared with the reading performance on large

print without a magnifier (at print size which is at least 0.3 log unit larger than the


275

near visual acuity), individual assessment of required acuity reserve for fluent

reading is necessary. Using this prescribing regime, reading rates versus different

print sizes do not need to be measured for most patients as a routine procedure for

the purpose of determining magnification.

6.2.3 Large print reading practice and reading performance with stand

magnifiers (Chapter 4)

Optical low vision aids are commonly used by people with visual impairment to

assist in reading. However, reading performance with the low vision aid may not

improve without practice or training in magnifier use. Instead of intensive training

programs in the use of magnifiers as suggested by some authors (Goodrich et al.,

1977; Nilsson and Nilsson, 1986 (b); Nilsson, 1988; Nilsson, 1990; Goodrich et al.,

2000), simple reading practice on large print before the prescription of a magnifier

may be beneficial. Watson et al. (1992) showed that training or reading practice was

effective in improving the reading performance of people with low vision. However,

measures of reading rates with magnifiers before and after training were not included

in their study. Other studies have shown an improvement in reading following short-

term reading practice with stand magnifiers for subjects with normal vision

(McMahon and Spigelman, 1989; Bowers, 2000 (b)). The aim of the current study

was to investigate the effect of simple reading practice with large print text, under

either full or reduced field of view (the latter simulated by a practice stand), on

reading rate with stand magnifiers for subjects with AMD.

The results of the experiment reported in Chapter 4 showed that reading rates for

both passages and sentences were significantly reduced when stand magnifiers were

first introduced. The reduction in reading rate could be due to the lack of reading

practice (due to visual impairment) and/or difficulty manipulating the magnifier

while reading. The results of this study indicated that simple large print reading

practice at home, either with or without reduced field of view, could benefit the

people with AMD in their subsequent use of stand magnifiers. This was certainly the

case for some individual subjects in this study but cannot be generalised to all people

with AMD or other low vision causes. After one week of large print reading practice

or repeated measures of reading rate with magnifier, magnifier reading rate improved


276

to the point that it was not significantly different from the reading rate on large print

without a magnifier.

For subjects who received neither large print reading practice nor prior exposure to

the stand magnifier before it was prescribed (i.e. subjects from the clinical group),

home practice in using the stand magnifier was essential to improve the magnifier

reading performance. The results indicated that with two weeks home practice,

reading rate with the magnifier for passages was equivalent to maximum reading rate

on large print (without magnifier). However, as reading performance was not

measured after one week’s practice with the magnifier, it was not known if one week

practice with the stand magnifier would have been sufficient to achieve maximum

reading rate with the magnifier.

Although reading rate for short sentences was significantly faster than that for

passages of text, results of reading performances with stand magnifiers across time

for both reading tasks were similar. The main difference was that a longer duration

was required for subjects to achieve their maximum reading rates with the magnifier

for sentences than for passages. For sentence reading, subjects required two more

weeks home practice with the stand magnifier, in addition to large print reading

practice (as improvement in the two practices was significantly more than that in the

control group), to achieve a maximum magnifier. In contrast, with passage reading

one week large print reading practice was sufficient for subjects to achieve maximum

reading rate. The reason for the difference in the period of practice required might be

the faster reading rates for sentences, hence subjects needed longer for the magnifier

reading rate to improve to maximum. While these reading materials reflect different

reading requirements for different types of daily tasks, the reading performances

measured with these two tasks are highly correlated. In clinical consultations,

assessment of reading performance with either one of these reading materials would

be sufficient.

For clinical purposes, one week of reading practice with large print at or above CPS

may be recommended for AMD patients before any magnifiers are prescribed as this

reading practice can enhance their reading performance with magnifiers. Reading

rates with and without magnifiers and with sentences and/or passages should be


277

measured when the magnifiers are prescribed and also in subsequent review visits.

These reading assessments should be taken at the beginning of the consultation so as

to reflect patients’ maximum reading performance before fatigue. A follow up visit is

suggested two weeks after the provision of a magnifier to assess any change in

magnifier reading rate and any difficulty in using the magnifier for reading. If the

magnifier reading rate shows no improvement compared with the reading rate at the

previous visit, or the reading rate with magnifier is still significantly slower than the

reading rate on large print, further investigations are necessary. These investigations

would include the evaluation of the magnification provided, the manipulation

strategy of the magnifier used by the patients and/or the analysis of other variables

limiting reading performance, such as visual field loss.

6.2.4 Line-guide gives no objective improvement in reading performance but is

preferred by more than fifty percent of subjects with AMD (Chapter 5)

People with low vision, in particular those who have little experience in using stand

magnifiers, often report that they have difficulty with page navigation. However,

some of the subjects who participated in the experiment reported in Chapter 4 also

made similar complaints, even after 18 weeks use of the magnifiers. They reported

that they either missed a line or took extra time to find the beginning of the next line.

In order to alleviate this problem, any reading assistance that can help magnifier

users to find their place while reading may be beneficial for reading performance

with these devices. The aim of the study reported in Chapter 5 was to investigate

whether a line guide attached to the base of the stand magnifier improved navigation

and reading performance.

The introduction of a line guide did not result in any significant improvements in

reading performance in terms of reading rate or number of navigation errors for

subjects with AMD, compared with their reading performance without the line guide

although it did change their STM movement strategies (straighter). In fact there was

a tendency for reading rate to decrease and for retrace errors to increase, probably

due to the design of the line guide interfering with retrace (see section 6.3.5).

Interestingly, these objective measurements of reading rate and navigation

performance did not agree with the subjective responses of the participants, where


278

nearly half of the subjects with AMD stated a preference to have the line guide

attached to their own stand magnifiers. The discrepancy between self-reported and

measured functions suggests that the objective measures might not be sensitive

enough to predict the subjective response. In addition there may have been other

factors that were not measured in this study which influenced subjects’ preferences

in selecting the line guide (e.g., psychological support provided by the line guide in

reading orientation). Among the vision and reading variables, monocular field of

view with the STM was correlated with subjective difficulty in retrace movement,

indicating that subjects with larger fields of view had less difficulty in retrace

movement. However, none of the variables could predict the preference for the line

guide in STM use.

Clinically, the subjective response of patients to the use of low vision aids as well as

their motivation are important criteria for success in low vision rehabilitation. For

subjects who had less experience using STM, a higher percentage of people preferred

the use of the line guide although it was not statistically significant. Before any

clinical trial confirms (or otherwise) the benefits of a line guide on reading with STM

(refer to section 6.3.5), a line guide can be considered as an option to assist patients

with AMD in using their STMs.

Magnifier movement recordings highlighted the specific difficulties that subjects had

with retrace when using the line guide. Modifications to the design of the line guide

such that it could better fit the pattern of patients’ magnifier movements have been

suggested (refer to section 6.3.5). In addition, specific instructions on how to retrace

when using a line guide attached to a stand magnifier should be given when the guide

is suggested to patients who have no previous experience in manipulating the

magnifiers and the line guide (e.g. based on a “straight” retrace strategy, retrace

along the line just read and then move down to the next line).


279

6.3 Future research

6.3.1 Inclusion of subjects with different causes of low vision

The results of this study only apply to people with AMD. To determine if the results

of the magnification calculated by the fixed acuity reserve method, the effect of

reading practice and the effect of a line guide on reading performance with

magnifiers apply to all patients with low vision, a similar study on people with low

vision resulting from diseases other than AMD would be useful. However, it is likely

that the results obtained from the current study would apply to patients with low

vision due to eye diseases other than AMD, especially those with central field loss.

In future, a double blind case-controlled study using different experimenters for

recruiting subjects into different groups and conducting the experimental assessments

should be used. This would help to minimise the possible bias on the training effect

on reading with magnifiers since the investigators performing the measurements

would not know which subjects received the reading practice.

As reported in Chapter 4 and Appendix 2, a sample size of 62 subjects in each

experimental group would be required to achieve statistical significance in such a

study. Because of the difficulty recruiting sufficient numbers of subjects, a multi-

centre study would be necessary to confirm whether the results are widely applicable.

6.3.2 Schedule for the review of reading performance with magnifiers

A reduction in reading rate has been found in previous studies when magnifiers were

first prescribed for people with normal vision and simulated low vision. This result

was confirmed for the subjects with visual impairment due to AMD in this study.

The results from Chapter 4 and the data of Goodrich et al. (1977) and Bowers (2000

(b)) show that there is a sharp increase in reading rate across time following

prescription of a low vision aid. Reading rate then plateaus quickly. The results of

the current study indicated that after two weeks practice using stand magnifiers in the

clinical group, reading rate returned to the subjects’ maximum reading rate (without

magnifiers). However there was no follow up one week after the stand magnifier was

prescribed for home use, therefore the optimal duration for the magnifier reading


280

practice and best required review schedule for monitoring patient’s performance

could not be determined. Future research should assess reading performance with

magnifiers on at least a weekly basis for four consecutive weeks following the

prescription of magnifiers.

6.3.3 Training in page navigation with a restricted field of view

Subjects in the experimental group P2 (Chapter 4) were given large print reading

practice through a restricted field of view which was simulated by a practice stand

(section 4.3.4). This practice stand allowed the subjects to practise reading under a

restricted field and to practise the page navigation movements required when using a

stand magnifier without the optical limitations (such as eye-image distance

requirements, aberrations etc). Surprisingly, this additional practice did not result in

any greater benefit over the large print reading practice with an unrestricted field of

view or over the control group; reasons for this finding have been discussed in

section 4.6.1. Results from Chapter 5 suggested that a specific strategy in page

navigation should be taught when the practice stand/stand magnifier is first

prescribed. For example, move the magnifier back along the same line and then

move it down to the next line. In future research, page navigation training with

specific strategies should be included if a restricted field of view is prescribed for

reading practice. Previous studies have shown that reading rate for people with AMD

is comparatively slower than people with other low vision causes (Legge et al., 1985

(b); Legge et al., 1992). For this reason, the 10-minute reading practice under

reduced field of view may not be sufficient for patients to be familiar with this

condition and manipulating the stand. Therefore, the daily duration of practice should

probably be longer (e.g. 30 minutes) to ensure that patients have enough time to learn

and practise the specific strategies in manipulating the stand when they read large

print books.

6.3.4 Page navigation with magnifiers

The results from Chapter 4 showed that after either one-week of large print reading

practice or measurements with STM or two-weeks of practice in using stand

magnifiers, magnifier reading rates are at maximum (equivalent to reading rates on


281

large print without a magnifier). The study investigated the change in magnifier

reading rate across time, but did not investigate the navigation performance with the

magnifier across time. A clinical longitudinal study should attempt to follow the

navigation performance of naïve subjects in the learning period of magnifier use.

Such a study would provide a better understanding of how subjects change their

strategies in manipulating their magnifiers to obtain faster reading rates. The

magnifier movement recordings (Chapter 5) indicated that some subjects had poor

navigation technique when manipulating their magnifiers. The reasons for the poor

navigation technique should be more fully investigated, for instance the relationships

between motor control, page navigation and magnifier manipulation should be

evaluated.

6.3.5 Improved design of the magnifier line guide

The current study attempted to determine the effect of a line guide on reading

performance with stand magnifiers (Chapter 5). However, the line guide used in the

study reduced the vertical field of view through the magnifying lens and blocked the

text from the next line completely. The participants could not preview where the next

line started, which resulted in more retrace difficulties and there was a tendency for

reading rate to decrease. To alleviate these limitations, a transparent yellow line

guide rather than the opaque white strip might be preferable. This new line guide can

orientate patients to the line they are reading but still allow them to preview the text

from the next line. In addition, a reduction in the width of the line guide to cover no

more than one line (e.g. 2-3 mm) rather than two lines of the reading text would

minimise any reduction in the vertical field of view. In a future study, comparison of

reading performances with this new line guide and with commercially available line

guides (e.g., red line and highlighted bright light from Eschenbach illuminated stand

magnifiers) would be recommended, to investigate the effectiveness of the different

line-guides and the subjective acceptance of line guides.

There was a tendency for subjects with low vision and less experience in using

magnifiers to prefer the use of the line guide. This implies that the attachment of a

line guide may be useful to new magnifier users. A clinical study should attempt to

investigate the effect of a line guide (with improved design) on reading performance


282

compared with reading performance without a line guide for new stand magnifier

users over time.

6.4 Clinical recommendations

The clinical recommendations presented below are derived from this study, with

support from other studies over the past 15 years which have assessed various

aspects of low vision reading. As the current study focused on people with visual

impairment due to AMD who had been prescribed stand magnifiers, emphasis is

given to AMD and on the prescription of stand magnifiers for reading.

6.4.1 Prescribe large print reading practice before prescribing magnifiers

Large print reading practice is very cheap and easy to implement for any low vision

clinic. Before any magnifier is prescribed, it may be beneficial if large print reading

practice is given at home for a minimum of 10 minutes daily for one to two weeks.

The print size should be at or above their critical print size (determined by the

procedures described below).

Prior to the selection of the appropriate print size, the patients’ reading goal should

be defined and their reading performance should be assessed. (Lovie-Kitchin and

Whittaker, 1999 (b)). This not only enables the patient to take an active part in the

rehabilitation process (Rumney, 1995), but also provides information for the

clinician in prescribing low vision aids. The required magnification for reading

depends on the print size of the reading task that the patient wants to achieve (Cole,

1993; Lovie-Kitchin and Whittaker, 1999 (b)). Hence, it is essential to know the goal

reading material and therefore the target print size.

The reading goal could be obtained from the referring practitioners’ information or

by telephone from the patient. Estimation of near visual acuity should be obtained by

a structured telephone questionnaire and practitioners should then suggest or provide

large print that is 0.3 to 0.5 log units larger than estimated threshold print size.


283

6.4.2 Measure reading performance and near visual acuity at the first visit

After one week’s reading practice on large print, reading performance should be

assessed. A text chart (eg. a Bailey-Lovie text reading chart) should be used to

measure reading performance and near visual acuity as coherent text, rather than

unconnected words, better reflects "real world" reading performance. Reading rates

at a working distance determined by the patient’s near addition should be measured

on a range of print sizes to calculate the patient's baseline reading performance

without a magnifier. Maximum reading rate (MRR) on large print (about 0.3 to 0.5

units larger than threshold print size) should be measured (or estimated). In addition,

near visual acuity (NVA), which is the smallest print size that the patient can resolve

and the viewing distance should be measured.

6.4.3 Calculate the required magnification

The reading goal, and therefore the target print size, should be confirmed with the

patient before the calculation of magnification. Required magnification in terms of

equivalent viewing distance (EVD) can be calculated based on the patient’s NVA

and print size of the goal reading task by the fixed acuity reserve method. The fixed

acuity reserve of 0.3 log units would be used to determine the tentative magnification

as it is simple and effective (Cheong et al., 2002).

6.4.4 Select appropriate magnifier and assess reading performance

Depending on the patient’s needs and required magnification, various types of low

vision aids (such as high addition, hand-held and stand magnifiers) giving the

calculated EVD can be selected for trial.

For stand magnifiers, the appropriate eye to lens distance to achieve optimum

magnification and a clear image needs to be demonstrated. Instructions by verbal

communication, visual and/or tactile demonstration and in-office trial of the

magnifiers should be provided when a magnifier is prescribed for home trial. Patients

should be given some time to practise with the magnifier in the consulting room and

then reading rate and threshold print size achieved with the magnifier should be

assessed.


284

Reading rate with the selected magnifier should be measured on the print size of the

patient’s target reading material (e.g. 8 Point print for newspaper reading). Lovie-

Kitchin and Whittaker (2000) recommend simply listening to the patient’s oral

reading, but reading time and the number of words read correctly with and without

the magnifier could be accurately measured. By comparing reading performance with

the magnifier at different visits, the clinician can objectively determine whether there

is any improvement in magnifier reading rate across time and whether maximum

reading rate (as determined without magnifier at CPS) has been achieved.

If the reading rate with the magnifier is significantly reduced compared with the

reading rate without the magnifier (on CPS), the individual acuity reserve method

rather than the fixed acuity reserve method should be used to re-calculate the

required magnification. However, if the reading rate with the new selected magnifier

is still significantly slower than the reading rate without magnifier, visual field

assessment should be conducted to determine if visual field loss is reducing the

magnifier reading performance (Lovie-Kitchin and Whittaker, 1999 (b)).

6.4.5 Loan the magnifier and review performance before prescribing

Before a magnifier is finally prescribed, it is recommended that the patient is loaned

the magnifier for a two-week trial. In addition, detailed information and written

instructions on the use of the magnifier should be given to the patient for reference

(Lovie-Kitchin and Bowman, 1985; Freeman and Jose, 1991) during the loan period.

Review of the reading performance with the prescribed magnifier is recommended

after two weeks. At each visit, the patient is requested to demonstrate how he/she has

been using the magnifier. This allows the practitioner to evaluate whether the patient

is using the magnifier correctly and whether any extra instruction or training has to

be given in manipulating the magnifier. Then near visual acuity and magnifier

reading rate at the target print size should be assessed. This determines any

deterioration in visual acuity and/or any change in magnifier reading rate at the target

print size. If the MRR achieved with magnifier is not similar to the MRR without

magnifier (see section 6.4.1), the magnification provided should be reassessed


285

(section 6.4.3) or further training on the use of magnifier is required (Lovie-Kitchin

and Whittaker, 2000).

6.4.6 Follow up visit

A telephone review of the patient’s vision and reading performance with the

magnifier is recommended three months after the first clinical consultation. If the

patient reports a significant deterioration in vision and reading performance with the

magnifier, further clinical assessment would be recommended. Based on the findings

of this study (Chapter 4), this seems likely for patients with AMD. Otherwise, as

others have suggested (Lovie-Kitchin and Bowman, 1985; Robbins and Murray,

1988), regular follow up in the clinic to review the patient’s vision, any change in

reading goals and the usefulness of the magnifier is recommended every 12 to 18

months.

6.5 Summary

As the prevalence of AMD and low vision are sharply increasing because of the

ageing population, the best possible vision rehabilitation is important to enable

people with visual impairment to maintain their quality of life. The results of this

study have demonstrated that the fixed acuity reserve method for calculating

magnification gives a valid and effective starting point for most patients with AMD.

Also for some patients with AMD, a line guide attached to the base of their stand

magnifier may be useful. The unique contribution of this study to the field of low

vision rehabilitation is that the benefit of short-term reading practice, on large print

or with magnifiers, as simple, cheap methods of enhancing reading performance with

stand magnifiers was demonstrated. The results of this study have led to

recommendations for training patients with AMD when they are prescribed stand

magnifiers. Improved strategies for reading rehabilitation for people who are visually

impaired have been derived from these findings, which can be implemented in

practitioners’ everyday optometry practice in managing patients with low vision.

References

286

REFERENCES

Age Related Macular Degeneration Study Group. (1996). Multicenter ophthalmic

and nutritional age-related macular degeneration study - Part 1: Design, subjects and

procedures. J Am Optom Assoc 67, 12-29.

Age-Related Eye Disease Study Research Group. (2001). A randomized, placebo-

controlled, clinical trial of high-dose supplementation with vitamins C and E, beta

carotene, and zinc for age-related macular degeneration and vision loss: AREDS

report no. 8. Arch Ophthalmol 119, 1417-1436.

Ahn, S. J., and Legge, G. E. (1995). Psychophysics of reading - XIII. Predictors of

magnifier-aided reading speed in low vision. Vision Res 35, 1931-1938.

Ahn, S. J., Legge, G. E., and Luebker, A. (1995). Printed cards for measuring low-

vision reading speed. Vision Res 35, 1939-1944.

Alexander, L. J. (1993). Age-related macular degeneration: The current

understanding of the status of clinicopathology, diagnosis and management. J Am

Optom Assoc 64, 822-837.

Algvere, P. V., Berglin, L., Gouras, P., Sheng, Y., and Kopp, E. D. (1997).

Transplantation of RPE in age-related macular degeneration: observations in

disciform lesions and dry RPE atrophy. Graefes Arch Clin Exp Ophthalmol 235,

149-158.

Aquilante, K. (1998). Repeatability of reading rate measurements in low vision.

Optom Vis Sci (Suppl) 75, 25.

Aquilante, K., Yager, D., and Khmelnitsky, F. (2001). Low-vision patients with age-

related maculopathy read RSVP faster than word duration varies according to word

length. Optom Vis Sci 78, 290-296.

References

287

Aquilante, K., Yager, D., and Morris, R. (2000). Can visual acuity predict the size of

text that low vision readers need to read at maximum rates? In: Vision

Rehabilitation - Assessment, Intervention and Outcomes (C. Stuen, A. Arditi, A.

Horowitz, M. A. Lang, B. Rosenthal, and K. Seidman, eds.). Lisse: Swets and

Zeitlinger, p. 288-291.

Archambault, P., and Colenbrander, A. (1989). Visual results with low-vision aids in

age-related macular degeneration [letter; comment]. Am J Ophthalmol 107, 564-

566.

Arnold, J. J., and Sarks, S. H. (2000). Extracts from "Clinical Evidence": Age related

macular degeneration. BMJ 321, 741-744.

Attebo, K., Mitchell, P., and Smith, W. (1996). Visual acuity and the causes of visual

loss in Australia: The Blue Mountains Eye Study. Ophthalmology 103, 357-364.

Auer, C., Tao Tran, V., and Herbort, C. P. (2002). Transpupillary thermotherapy for

occult subfoveal neovessels in age-related macular degeneration: importance of

patient pigmentation for the determination of laser settings. Klin Monatsbl

Augenheilkd 219, 250-253.

Bailey, I. L. (1980). Magnification for near vision. Optom Monthly 71, 119-122.

Bailey, I. L. (1981 (a)). The use of fixed-focus stand magnifiers. Optom Monthly

72, 37-39.

Bailey, I. L. (1981 (b)). Locating the image in stand magnifiers. Optom Monthly 72,

22-24.

Bailey, I. L. (1984 (a)). Equivalent viewing power or magnification. Which is

fundamental? Optician 188, 32-35.

Bailey, I. L. (1984 (b)). Magnification of the problem of magnification. Optician

187, 14-18.

Bailey, I. L. (1991). Low vision visual acuity. In: Clinical Procedures in

Optometry (J. B. Eskridge, J. F. Amos, and J. D. Bartlett, eds.). Philadelphia: JB

Lippincott, p. 754-761.

References

288

Bailey, I. L., Bullimore, M. A., Greer, R. B., and Mattingly, W. B. (1994). Low

vision magnifiers - their optical parameters and methods for prescribing. Optom Vis

Sci 71, 689-698.

Bailey, I. L., Bullimore, M. A., Hall, A., and Orel-Bixler, D. A. (1989). Reading

characteristics in age-related maculopathy. Invest Ophthalmol Vis Sci (Suppl) 30,

400.

Bailey, I. L., and Lovie, J. E. (1976). New design principles for visual acuity letter

charts. Am J Optom Physiol Opt 53, 740-745.

Bailey, I. L., and Lovie, J. E. (1980). The design and use of a new near-vision chart.

Am J Optom Physiol Opt 57, 378-387.

Bailey, I. L., and Tuan, K. M. (1998). Prescribing magnifiers for near vision In:

Vision '96, Book 1, International Conference on Low Vision Proceedings.

Madrid: Organización Nacional de Ciegos Españoles, p. 141-145.

Baldasare, J., Watson, G. R., Whittaker, S. G., and Miller-Shaffer, H. (1986). The

development and evaluation of a reading test for low vision individuals with macular

loss. J Vis Impairment Blindness 80, 785-789.

Bane, M. C., Fish, G. E., Spencer, R., and Moushegian, S. E. (1996). MNRead acuity

test reliability in patients with age-related maculopathy In: Vision Science and Its

Applications, 1996 Technical Digest Series. New Mexico: Optical Society of

America, p. 24-27.

Beatty, S., Boulton, M., Henson, D., Koh, H. H., and Murray, I. J. (1999). Macular

pigment and age related macular degeneration. Br J Ophthalmol 83, 867-877.

Beckmann, P. J., and Legge, G. E. (1996). Psychophysics of reading - XIV. The page

navigation problem in using magnifiers. Vision Res 36, 3723-3733.

References

289

Bergink, G. J., Hoyng, C. B., van der Maazen, R. W. M., Vingerling, J. R., van Daal,

W. A., and Deutman, A. F. (1998). A randomized controlled clinical trial on the

efficacy of radiation therapy in the control of subfoveal choroidal neovascularization

in age-related macular degeneration: radiation versus observation. Graefes Arch

Clin Exp Ophthalmol 236, 321-325.

Bird, A. C., Bressler, N. M., Bressler, S. B., Chisholm, I. H., Coscas, G., Davis, M.

D., and et al. (1995). An international classification and grading system for age-

related maculopathy and age-related macular degeneration. The International ARM

Epidemiological Study Group. Surv Ophthalmol 39, 367-374.

Bischoff, P. (1995). Long-term results of low vision rehabilitation in age-related

macular degeneration. Doc Ophthalmol 1995, 305-311.

Bland, J. M., and Altman, D. G. (1986). Statistical methods for assessing agreement

between two methods of clinical measurement. The Lancet 8476, 307-310.

Bowers, A. R. (1998 (a)). Reading performance in age-related macular degeneration:

The relationship between intelligence and comprehension In: Vision '96, Book 1,

International Conference on Low Vision Proceedings. Madrid, Spain:

Organización Nacional de Ciegos Españoles, p. 217-222.

Bowers, A. R. (1998 (b)). Reading Performance in Visual Impairment.

Unpublished PhD Thesis. Glasgow Caledonian University, Glasgow, United

Kingdom.

Bowers, A. R. (2000 (a)). Eye movements and reading with plus-lens magnifiers.

Optom Vis Sci 77, 25-33.

Bowers, A. R. (2000 (b)). Stand magnifiers: page navigation and short-term practice

effects. In: Vision Rehabilitation - Assessment, Intervention and Outcomes (C.

Stuen, A. Arditi, A. Horowitz, M. A. Lang, B. Rosenthal, and K. Seidman, eds.).

Lisse: Swets and Zeitlinger, p. 319-323.

References

290

Bowers, A. R., and Ackerley, R. (1994). Reading characteristics of normal observers

using low vision aids. In: Low Vision - Research and New Developments in

Rehabilitation (A. C. Kooijman, P. L. Looijestijn, J. A. Welling, and G. J. van der

Wildt, eds.). Amsterdam: IOS Press, p. 235-238.

Bowers, A. R., Lovie-Kitchin, J. E., and Cheong, A. M. Y. (2002 (a)). The effect of

passage width on page navigation with optical magnifiers. Invest Ophthalmol Vis

Sci 43, E-Abstract 4786.

Bowers, A. R., Lovie-Kitchin, J. E., and Cheong, A. M. Y. (2002 (b)). Factors

affecting page navigation with optical magnifiers In: The 7th International

Conference on Low Vision: Conference Proceedings. Goteborg, p. 94.

Bowers, A. R., Lovie-Kitchin, J. E., and Woods, R. L. (2001 (b)). Eye movements

and reading with large print and optical magnifiers in macular disease. Optom Vis

Sci 78, 325-334.

Bowers, A. R., Meek, C., and Steward, N. (2001 (a)). Illumination and reading

performance in age-related macular degeneration. Clin Exp Optom 84, 139-147.

Bowers, A. R., and Reid, V. M. (1997). Eye movements and reading with simulated

visual impairment. Ophthal Physiol Opt 17, 392-402.

Bowman, K. J. (1980). The relationship between colour discrimination and visual

acuity in senile macular degeneration. Am J Optom Physiol Opt 57, 145-148.

Bradley-Klug, K. L., Shapiro, E. S., Lutz, J. G., and DuPaul, J. G. (1998). Evaluation

of oral reading rate as a curriculum-based measure within literature-based

curriculum. Vision Res 36, 183-197.

Brady, L., Freire, J. E., Longton, W. A., Miyamoto, C. T., Augsburger, J. J., Brown,

G. C., and et al. (1997). Radiation therapy for macular degeneration: technical

considerations and preliminary results. Int J Radiat Oncol Biol Phys 39, 945-948.

Bressler, N. M., and Bressler, S. B. (1995). Preventative ophthalmology. Age-related

macular degeneration. Ophthalmology 102, 1206-1211.

References

291

Bressler, N. M., Bressler, S. B., and Fine, S. L. (1988). Age-related macular

degeneration. Surv Ophthalmol 32, 375-416.

Bressler, N. M., Bressler, S. B., West, S. K., Fine, S. L., and Taylor, H. R. (1989).

The grading and prevalence of macular degeneration in Chesapeake Bay watermen.

Arch Ophthalmol 107, 847-852.

Bressler, N. M., Munoz, B., Maguire, M. G., Vitale, S. E., Schein, O. D., Taylor, H.

R., and et al. (1995). Five- year incidence and disappearance of drusen and retinal

pigment epithelial abnormalities. Waterman study. Arch Ophthalmol 115, 301-308.

Brilliant, R. L. (1999). Essentials of Low Vision Practice. Boston: Butterworth-

Heinemann.

Brown, B. (1981). Reading performance in low vision patients: relation to contrast

and contrast sensitivity. Am J Optom Physiol Opt 58, 218-226.

Brown, B. (1997). The Low Vision Handbook. Thorofare, NJ, SLACK Inc.

Brown, B., Adams, A. J., Coletta, N. J., and Haegerstrom-Portnoy, G. (1986 (a)).

Dark adaptation in age-related maculopathy. Ophthal Physiol Opt 6, 81-84.

Brown, B., Brabyn, J. A., Welsh, R. L., Haegerstrom-Portnoy, G., and Colenbrander,

A. (1986 (b)). Contribution of vision variables to mobility in age-related

maculopathy patients. Am J Optom Physiol Opt 63, 733-739.

Brown, B., and Garner, L. F. (1983). Effects of luminance on contrast sensitivity

senile macular degeneration. Am J Optom Physiol Opt 60, 188-193.

Brown, B., and Kitchin, J. E. (1983). Dark adaptation and the acuity/luminance

response in senile macular degeneration (SMD). Am J Optom Physiol Opt 60, 645-

650.

Brown, B., Tobin, C., Roche, N., and Wolanowski, A. (1986 (c)). Cone adaptation in

age-related maculopathy. Am J Optom Physiol Opt 63, 450-454.

Bruce, I., McKennell, A., and Walker, E. (1991). Blind and Partially Sighted

Adults in Britain: the RNIB Survey. London: HMSO.

References

292

Buettner, M., Krischer, C. C., and Meissen, R. (1985). Characterization of gliding

text as a reading stimulus. Bull Psychom Society 23, 479-482.

Bullimore, M. A., and Bailey, I. L. (1989). Stand magnifiers: an evaluation of new

optical aids from COIL. Optom Vis Sci 66, 766-773.

Bullimore, M. A., and Bailey, I. L. (1995). Reading and eye movements in age-

related maculopathy. Optom Vis Sci 72, 125-138.

Bullimore, M. A., Bailey, I. L., and Wacker, R. T. (1990). Contrast sensitivity and

functional performance in low vision. Invest Ophthalmol Vis Sci (Suppl) 31, 598.

Carver, R. P. (1983). Is reading rate constant or flexible? Reading Res Quarterly

18, 190-215.

Carver, R. P. (1990). Reading Rate: A Review of Research and Theory. San

Diego: Academic Press.

Chakravarthy, U., Houston, R. F., and Archer, O. B. (1993). Treatment of age-related

subfoveal neovascular membranes by teletherapy: a pilot study. Br J Ophthalmol

77, 265-273.

Cheng, A. S., and Vingrys, A. J. (1993). Visual losses in early age-related

maculopathy. Optom Vis Sci 70, 89-96.

Cheong, A. M. Y., Lovie-Kitchin, J. E., and Bowers, A. R. (2002). Determining

magnification for reading with low vision. Clin Exp Optom 85, 229-237.

Chisholm, I. H. (1996). Macular degeneration: management and prospects for the

future. Optom Today 21, 39-40.

Chung, S. T., Mansfield, J. S., and Legge, G. E. (1998 (a)). Psychophysics of reading

- XVIII. The effect of print size on reading speed in normal peripheral vision. Vision

Res 38, 2949-2962.

Cohen, J. M. (1993). Contrast of common near point reading materials. J Vis Rehab

7, 2-4.

References

293

Cohen, J. M., and Waiss, B. (1991 (a)). Comparison of reading speed in normal

observers through different forms of equivalent power low vision devices. Optom

Vis Sci 68, 127-131.

Cohen, J. M., and Waiss, B. (1991 (b)). Reading speed through different equivalent

power low vision devices with identical field of view. Optom Vis Sci 68, 795-797.

Cole, B. L. (1974). Prescribing light for the aging patient. Aust J Optom 57, 207-

214.

Cole, R. G. (1993). Predicting the low vision reading add. J Am Optom Assoc 64,

19-27.

Colenbrander, A. (1977). Dimensions of visual performance. Trans Am Acad

Ophthalmol Otolaryngol 83, 332-337.

Colenbrander, A. (2000). Quantifying low vision: ranges of vision loss. In:

Principles and Practice of Ophthalmology (D. M. Albert, F. A. Jakobiec, D. T.

Azar, and E. S. Gragoudas, eds.). Pennsylvania: W.B. Saunders Company, p. 5399-

5404.

Colenbrander, A., Liegner, J. T., and Fletcher, D. C. (1999). Enhancing impaired

vision. In: Low Vision Rehabilitation: Caring for the Whole Person (D. C.

Fletcher, ed.). San Francisco: American Academy of Ophthalmology.

Collins, M. J. (1986). Pre-age related maculopathy and the desaturated D-15 color

vision test. Clin Exp Optom 69, 223-227.

Collins, M. J., and Brown, B. (1989). Glare recovery and age-related degeneration.

Clin Vision Sci 4, 145-153.

Collins, M. J., and Carney, L. G. (1987). Patient compliance and its influence on

contact lens wearing problems. Am J Optom Physiol Opt 63, 952-956.

Cooper, R. L. (1990). Blind registrations in Western Australia: a five-year study.

Aust NZ J Ophthalmol 18, 421-426.

References

294

Corn, A., and Ryser, G. (1989). Access to print for students with low vision. J Vis

Impairment Blindness 83, 340-349.

Cornelissen, F. W., Kooijman, A. C., Bootsma, A., and van der Wildt, G. J. (1994).

Light and objects; perceiving "The real thing". In: Low Vision: Research and New

Developments in Rehabilitation (A. C. Kooijman, P. L. Looijestin, J. A. Welling,

and G. J. van der Wildt, eds.). Netherlands: IOS Press, p. 78-84.

Cornelissen, F. W., Kooijman, A. C., Dumbar, G. V., van der Wildt, G. J., and

Nijland, R. (1991). Illumination research as part of a visual assessment of visually

impaired individuals. Doc Ophthalmol 78, 195-203.

Culham, L. E., Fitzke, F. W., Timberlake, G. T., and Marshall, J. (1992). Use of

scrolled text in a scanning laser ophthalmoscope to assess reading performance at

different retinal locations. Ophthal Physiol Opt 12, 281-286.

Cumming, R. W., Whittaker, S. G., and Sinclair, S. (1991). Scotoma characteristics

and reading performance in patients with central and paracentral vision loss. Invest

Ophthalmol Vis Sci (Suppl) 32, 816.

Cummings, R. W., Whittaker, S. G., and Swieson, L. (1989). Individuals with

maculopathies use OKN to scan text during unconstrained reading. Invest

Ophthalmol Vis Sci (Suppl) 30, 398.

Cummings, R. W., Whittaker, S. G., and Watson, G. R. (1988). Macular scotoma and

reading. Optom Vis Sci (Suppl) 65, 126P.

Cummings, R. W., Whittaker, S. G., Watson, G. R., and Budd, J. M. (1985).

Scanning characters and reading with a central scotoma. Am J Optom Physiol Opt

62, 833-843.

Davis, C., Lovie-Kitchin, J., and Thompson, B. (1995). Psychosocial adjustment to

age-related macular degeneration. J Vis Impairment Blindness 89, 16-27.

Den Brinker, B. P. L. M., and Bruggeman, H. (1996). Visual requirements for

reading: the importance of a large field of view in reading with a magnifier. J

Videology 1, 27-38.

References

295

Deruaz, A., Whatham, A. R., Mermoud, C., and Safran, A. B. (2002). Reading with

multiple preferred retinal loci: implications for training a more efficient reading

strategy. Vision Res 42, 2947-2957.

Dickinson, C. (1998). Low Vision: Principles and Practice. Boston: Butterworth-

Heinemann.

Dickinson, C. M., and Fotinakis, V. (2000). The limitations imposed on reading by

low vision aids. Optom Vis Sci 77, 364-372.

Dickinson, C. M., and Rabbitt, P. M. A. (1991). Simulated visual impairment:

Effects on text comprehension and reading rate. Clin Vision Sci 6, 301-308.

Dickman, I. R. (1982). A Vision Impairment of the Later Years: Macular

Degeneration. (Public Affairs Pamphlet No. 610). New York: Public Affairs

Committee.

Doorduyn, K. E., Harper, R. A., Slater, L., and Reeves, B. C. (1998). A pilot study of

a low vision specific questionnaire In: Vision '96, Book 2, International

Conference on Low Vision Proceedings. Madrid: Organización Nacional de Ciegos

Españoles, p. 471-472.

Dosso, A. A., and Bovet, J. (1992). Monozygotic twin brothers with age-related

macular degeneration. Ophthalmologica 205, 24-28.

Duchnicky, R. L., and Kolhners, P. A. (1983). Readability of text scrolled on visual

display terminals as a function of window size. Human Factors 25, 683-692.

Duret, F., Issenhuth, M., and Safran, A. B. (1999). Combined use of several preferred

retinal loci in patients with macular disorders when reading single words. Vision Res

39, 873-879.

Edwards, M. G., Bressler, N. M., and Raja, S. C. (1999). Age-related macular

degeneration. In: Ophthalmology (M. Yanoff and J. S. Duker, eds.). London:

Mosby, p. 28.01-28.10.

References

296

Eisner, A., Fleming, S. A., Klein, M. L., and Mauldin, W. M. (1987). Sensitivities in

older eyes with good acuity: eyes whose fellow eye has exudative AMD. Invest

Ophthalmol Vis Sci 28, 1832-1837.

Eisner, A., Stoumbos, V. D., Klein, M. L., and Fleming, S. A. (1991). Relations

between fundus appearance and function. Eyes whose fellow eye has exudative age-

related macular degeneration. Invest Ophthalmol Vis Sci 32, 8-20.

Elam, J. H. (1997). Analysis of methods for predicting near-magnification power. J

Am Optom Assoc 68, 31-36.

Elam, J. T., Graney, M. J., Beaver, T., El Derwi, D., Applegate, W. B., and Miller, S.

T. (1991). Comparison of subjective ratings of function with observed functional

ability of frail older persons. Am J Public Health 81, 1127-1130.

Eldred, K. B. (1992). Optimal illumination for reading in patients with age-related

maculopathy. Optom Vis Sci 69, 46-50.

Elliott, D. B., Patel, B., and Whitaker, D. (2001). Development of a reading speed

test for potential-vision measurements. Invest Ophthalmol Vis Sci 42, 1945-1949.

Elliott, D. B., Sanderson, K., and Conkey, A. (1990 (b)). The reliability of the Pelli-

Robson contrast sensitivity chart. Ophthal Physiol Opt 10, 21-24.

Elliott, D. B., Trukolo-Ilic, M., Strong, J. G., Pace, R., Plotkin, A., and Bevers, P.

(1997). Demographic characteristics of the vision-disabled elderly. Invest

Ophthalmol Vis Sci 38, 2566-2575.

Elliott, D. B., Whitaker, D., and Bonette, L. (1990 (a)). Differences in the legibility

of letters at contrast threshold using the Pelli-Robson chart. Ophthal Physiol Opt

10, 323-326.

Enoch, J. M. (1998). "Gold Standards" for Vision Testing. Invest Ophthalmol Vis

Sci 39, 1297-1299.

Evans, J. R. (1995). Causes of Blindness and Partial Sight in England and Wales

1990-1991. Studies on Medical and Population Subjects. London: HMSO.

References

297

Evans, J. R. (2001). Risk factors for age-related macular degeneration. Prog Retin

Eye Res 20, 227-253.

Evans, J. R., and Wormald, R. (1996). Is the incidence of registrable age-related

macular degeneration increasing? Br J Ophthalmol 80, 9-14.

Farrall, H. (1991). Optometric Management of Visual Handicap. London:

Blackwell Scientific Publications.

Faye, E. E. (1970). The Low Vision Patient: Clinical Experience with Adults and

Children. New York, Grune & Stratton.

Ferris, F. L. I., Fine, S. L., and Hyman, L. G. (1984). Age-related macular

degeneration. Surv Ophthalmol 102, 1640-1642.

Fine, E. M., Kirschen, M. P., and Peli, E. (1996). The necessary field of view to read

with an optimal stand magnifier. J Am Optom Assoc 67, 382-389.

Fine, E. M., and Peli, E. (1995). Scrolled and rapid serial visual presentation texts are

read at similar rates by the visually impaired. J Opt Soc Am A 12, 2286-2292.

Fine, E. M., and Peli, E. (1996). Visually impaired observers require a larger window

than normally sighted observers to read from a scroll display. J Am Optom Assoc

67, 390-396.

Fine, E. M., and Peli, E. (1998). Benefits of rapid serial visual presentation (RSVP)

over scrolled text vary with letter size. Optom Vis Sci 75, 191-196.

Fine, E. M., and Rubin, G. S. (1999). The effects of simulated cataract on reading

with normal vision and simulated central scotoma. Vision Res 39, 4274-4285.

Fischer, M. (2000). Macular degeneration: functional implications. In: Vision

Rehabilitation - Assessment, Intervention and Outcomes (C. Stuen, A. Arditi, A.

Horowitz, M. A. Lang, B. Rosenthal, and K. Seidman, eds.). Lisse: Swets and

Zeitlinger. p. 32-35.

References

298

Fitz, S. N., Gan, J., Lin, K., Tuan, K. M., and Bailey, I. (2000). Effects of a user

controlled highlighter on reading rates and eye movements in ARMD patients.


Flaxel, C. J. (2002). Use of radiation in the treatment of age-related macular

degeneration. Ophthalmol Clin North Am 15, 437-444.

Fletcher, D. C. (1999). Low Vision Rehabilitation: Caring For The Whole

Person. San Francisco: American Academy of Ophthalmology.

Fletcher, D. C., and Schuchard, R. A. (1997). Preferred retinal loci relationship to

macular scotomas in a low-vision population. Ophthalmology 104, 632-638.

Fletcher, D. C., Schuchard, R. A., and Watson, G. R. (1999). Relative locations of

macular scotomas near the PRL: Effect on low vision reading. J Rehabil Res Dev

36, 356-364.

Flom, R. E., Raasch, T. W., Braudway, S. M., Rubin, G. S., and Massof, R. W.

(1993). A test of Kestenbaum's rule for prescribing low vision reading devices.


Fotinakis, V. P., and Dickinson, C. M. (1994). Reading with hand magnifiers. In:

Low Vision - Research and New Developments in Rehabilitation (A. C.

Kooijman, P. L. Looijestijn, J. A. Welling, and G. J. van der Wildt, eds.).

Amsterdam: IOS Press, p. 259-268.

Franks, W. (2002). Photodynamic therapy for age-related macular degeneration. Age

Ageing 31, 5-6.

Freeman, P. B., and Jose, R. T. (1991). The Art and Practice of Low Vision.

Boston: Butterworth-Heinemann.

Freeman, P. B., and Jose, R. T. (1997). The Art and Practice of Low Vision.

Boston: Butterworth-Heinemann.

Friedman, D. B. (2000). Psychosocial factors in vision rehabilitation. In: Principles

and Practice of Ophthalmology (D. M. Albert, F. A. Jakobiec, D. T. Azar, and E.

S. Gragoudas, eds.). Pennsylvania: W.B. Saunders Company, p. 5404-5408.

References

299

Friedman, D. S., Katz, J., Bressler, N. M., Rahmani, B., and Tielsch, J. M. (1999 (a)).

Racial differences in the prevalence of age-related macular degeneration: the

Baltimore Eye Survey. Ophthalmology 106, 1049-1055.

Friedman, S. M., Munoz, B., Rubin, G. S., West, S. K., Bandeen-Roche, K., and

Fried, L. P. (1999 (b)). Characteristics of discrepancies between self-reported visual

function and measured reading speed. Salisbury Eye Evaluation Project Team.

Invest Ophthalmol Vis Sci 40, 858-864.

Gass, J. D. (1994). Biomicroscopic and histopathologic considerations regarding the

feasibility of surgical excision of subfoveal neovascular membranes. Am J

Ophthalmol 118, 285-298.

Gass, J. D. M. (1997). Stereoscopic Atlas of Macular Diseases; Diagnosis and

Treatment. St Louis: CV Mosby.

Goodrich, G. L., Apple, L. E., Frost, A., Wood, A., Ward, R., and Darling, N. (1976).

A preliminary report on experienced closed-circuit television users. Am J Optom

Physiol Opt 53, 7-15.

Goodrich, G. L., and Kirby, J. (2001). Low vision training for reading: A Goldilocks

effect? Optom Vis Sci (Suppl) 78, 292.

Goodrich, G. L., Kirby, J., Keswick, C., Donald, B., Oros, T., Wagstaff, P., Hazan,

J., and et al. (1998). Low vision reading: effects of field loss, training and device.


Goodrich, G. L., Kirby, J., Keswick, C., Oros, T., Wagstaff, P., Donald, B., and et al.

(2000). Training the patient with low vision to read: Does it significantly improve

function? In: Vision Rehabilitation - Assessment, Intervention and Outcomes (C.

Stuen, A. Arditi, A. Horowitz, M. A. Lang, B. Rosenthal, and K. Seidman, eds.).

Lisse: Swets and Zeitlinger, p. 230-236.

Goodrich, G. L., and Mehr, E. B. (1986). Eccentric viewing training and low vision

aids: current practice and implications of peripheral retinal research. Am J Optom


References

300

Goodrich, G. L., Mehr, E. B., and Darling, N. C. (1980). Parameters in the use of

CCTV's and optical aids. Am J Optom Physiol Opt 57, 881-892.

Goodrich, G. L., Mehr, E. B., Quillman, R. D., Shaw, H. K., and Wiley, J. K. (1977).

Training and practice effects in performance with low-vision aids: a preliminary

study. Am J Optom Physiol Opt 54, 312-318.

Goodrich, G. L., and Quillman, R. D. (1977). Training eccentric viewing. J Vis


Gottfredsdottir, M. S., Sverrisson, T., Musch, D. C., and Stefansson, E. (1999). Age-

related macular degeneration in monozygotic twins and their spouses in Iceland.

Acta Ophthalmol Scand 77, 422-425.

Government Statistical Service. (1994). Registered Blind and Partially-Sighted

People at 31 March 1994 England. London: HMSO.

Graessley, D., and Kirby, J. (1996). Literature review of current programs for

training eccentric viewing. J Vis Rehab 10, 19-21.

Green, S. B., Salkind, N. J., and Akey, T. M. (2000). Using SPSS for Windows:

Analyzing and Understanding Data. London: Prentice Hall.

Gregor, Z., and Joffe, L. (1978). Senile macular changes in the black African. Br J


Grey, R. H. B., Burns-Cox, C. J., and Hughes, A. (1989). Blind and partial sight

registration in Avon. Br J Ophthalmol 73, 88-94.

Grizzard, W. S., and Beck, R. W. (1994). Twin study of age related macular

degeneration. Invest Ophthalmol Vis Sci (Suppl) 35, S1504.

Guez, J., Le Gargasson, J., Rigaudiere, R., and O'Regan, J. K. (1993). Is there any

systematic location for the pseudo-fovea in patients with central scotoma? Vision

Res 33, 1271-1279.

Hall, A., Sacks, S. Z., Dornbusch, H., and Raasch, T. (1987). A preliminary study to

evaluate patients services in a low vision clinic. J Vis Rehab 1, 7-25.

References

301

Hampton, G. R., and Nelsen, P. T. (1992). Age Related Macular Degeneration.

New York: Raven Press Ltd.

Harper, R., Doorduyn, K., Reeves, B., and Slater, L. (1999). Evaluating the outcomes

of low vision rehabilitation. Ophthal Physiol Opt 19, 3-11.

Hart, P. M., Chakravarthy, U., Mackenzie, G., Chisholm, I. H., Bird, A. C.,

Stevenson, M. R., and et al. (2002). Visual outcomes in the subfoveal radiotherapy

study: a randomized controlled trial of teletherapy for age-related macular

degeneration. Arch Ophthalmol 120, 1029-1038.

Hart, W. M. (1987). Acquired dyschromatopsias. Surv Ophthalmol 32, 10-31.

Hassan, S. E., Lovie-Kitchin, J. E., and Woods, R. L. (2002). Vision and mobility

performance of subjects with age-related macular degeneration. Optom Vis Sci 79,

697-707.

Hawkins, B. S., Bird, A., Klein, R., and West, S. K. (1999). Epidemiology of age-

related macular degeneration [Review]. Mol Vis 5, 26.

Haymes, S. A., Johnston, A. W., and Heyes, A. D. (2000). Melbourne low vision

ADL index as an outcome measure. Optom Vis Sci (Suppl) 77S, 122.

Haymes, S. A., Johnston, A. W., and Heyes, A. D. (2001). A weighted-version of the

Melbourne Low-Vision ADL Index: a measure of disability impact. Optom Vis Sci

78, 565-579.

Hazel, C. A., Petre, K. L., Armstrong, R. A., Benson, M. T., and Frost, N. A. (2000).

Visual function and subjective quality of life compared in subjects with acquired

macular disease. Invest Ophthalmol Vis Sci 41, 1309-1315.

Henson, D. B. (1993). Visual Fields. New York: Oxford University Press.

Herreros, M. J. G., Mateus, G. G., Garcia, M. A. M., Ramon, M. P., and Mosquete,

M. J. V. (1998). Characteristics of reading performed by low-vision persons In:

Vision '96, Book 2, International Conference on Low Vision Proceedings.

Madrid: Organización Nacional de Ciegos Españoles, p. 28-35.

References

302

Hickson, L., Lind, C., Worrall, L., Yiu, E., Barnett, H., and Lovie-Kitchin, J. (1999).

Hearing and vision in health older Australians: Objective and self-reported measures.

Advances in Speech-Language Pathology 1, 95-105.

Howe, L. J. (1995). Clinical management of age-related macular degeneration: room

for improvement. Eye 9, 42-46.

Humphrey, R. C., and Thompson, G. M. (1986). Low vision aids - evaluation in a

general eye department. Trans Am Ophthalmol Soc 105, 296-303.

Hyman, L. (1992). Epidemiology of AMD. In: Age-Related Macular

Degeneration: Principles and Practice (G. R. Hampton and P. T. Nelsen, eds.).

New York: Raven Press, p. 1-35.

Hyman, L. G., Lilienfeld, A. M., Ferris, F. L., and Fine, S. L. (1983). Senile macular

degeneration: a case-control study. Am J Epidemiol. 118, 213-227.

Ip, M., Kroll, A., and Reichel, E. (1999). Transpupillary thermotherapy. Semin


Johnston, A. W. (1982). Technical note: The relationship between magnification and

field of view of simple magnifiers. Aust J Optom 65, 74-77.

Jose, R. T. (1983). Understanding Low Vision. New York: American Foundation

for the Blind.

Julian, W. G. (1984). Variation in near visual acuity with illuminance for a group of

27 partially-sighted people. Lighting Res Technol 16, 34-41.

Kanski, J. J. (1999). Clinical Ophthalmology: A Systematic Approach. Oxford:

Butterworth-Heinemann.

Karel, I., and Zahlava, J. (2002). Transpupillary thermotherapy in age-related

macular degeneration with choroidal membrane neovascularization. Cesk Slov

Oftalmol 58, 274-277.

References

303

Karel, I., Zahlava, J., Boguszakova, J., Dubska, Z., and Lestak, J. (2002).

Transpupillary thermotherapy in age-related macular degeneration. Preliminary

results. Cesk Slov Oftalmol 58, 215-223.

Katz, J., and Tielsch, J. M. (1996). Visual function and visual acuity in an urban

adult population. J Vis Impairment Blindness 90, 367-377.

Keppel, G. (1991). Design and Analysis: A Researcher's Handbook. Englewood

Cliffs, N.J., Prentice-Hall.

Kestenbaum, A., and Sturman, R. M. (1956). Reading glasses for patients with very

poor vision. Arch Ophthalmol 56, 451-470.

Kitchin, J. E., and Bailey, I. L. (1981). Task complexity and visual acuity in senile

macular degeneration. Aust J Optom 64, 235-242.

Klaver, C. C., Assink, J. J., van Leeuwen, R., Wolfs, R. C., Vingerling, J. R., Stijnen,

T., and et al. (2001). Incidence and progression rates of age-related maculopathy: the

Rotterdam Study. Invest Ophthalmol Vis Sci 42, 2237-2241.

Klaver, C. C. W., Wolfs, R. C. W., Vingerling, J. R., Hofman, A., and De Jong, P. T.

V. M. (1998). Age-specific prevalence and causes of blindness and visual

impairment in an older population: The Rotterdam Study. Arch Ophthalmol 116,

653-658.

Klein, B. E., and Klein, R. (1982). Cataracts and macular degeneration in older

Americans. Arch Ophthalmol 100, 571-573.

Klein, R. (1999). Epidemiology. In: Age Related Macular Degeneration (J. W.

Berger, S. L. Fine, and M. G. Maguire, eds.). St Louis: Mosby, p. 31-56.

Klein, R., Klein, B. E., Jensen, S. C., and Meuer, S. M. (1997). The 5 year incidence

and progression of age related maculopathy: the Beaver Dam Eye Study.

Ophthalmology 104, 7-21.

Klein, R., Klein, B. E., Lee, K. E., Cruickshanks, K. J., and Chappell, R. J. (2001).

Changes in visual acuity in a population over a 10-year period : The Beaver Dam

Eye Study. Ophthalmology 108, 1757-1766.

References

304

Klein, R., Klein, B. E., and Linton, K. L. (1992). Prevalence of age-related

maculopathy. The Beaver Dam Eye Study. Ophthalmology 99, 933-943.

Klein, R., Klein, B. E. K., Linton, K. L., and DeMets, D. L. (1991). The Beaver Dam

Eye Study: Visual acuity. Ophthalmology 98, 1310-1315.

Klein, R., Wang, Q., Klein, B. E., Moss, S. E., and Meuer, S. M. (1995). The

relationship of age-related maculopathy, cataract, and glaucoma to visual acuity.


Kleiner, R. C., Enger, C., Alexander, M. F., and Fine, S. L. (1988). Contrast

sensivitiy in age-related macular degeneration. Arch Ophthalmol 106, 55-57.

Knave, B., Tengroth, B., and Voss, M. (1984). Age and sex distribution of some

retinal macular diseases: senile and presenile macular degeneration and central

serous retinitis. Acta Ophthalmol 61, 95-103.

Kocur, I., and Resnikoff, S. (2002). Visual impairment and blindness in Europe and

their prevention. Br J Ophthalmol 86, 716-722.

Koenig, A. J., Layton, C. A., and Ross, D. B. (1992). The relative effectiveness of

reading in large print and with low vision devices for students with low vision. J Vis


Krieger, A. A. (1967). The partially sighted patient: a study of 917 cases. Trans Am

Ophthalmol Soc 65, 544-590.

Krischer, C. C., Nat, R., Meissen, R., and Ing, D. R. (1983). Reading speed under

real and simulated visual impairment. J Vis Impairment Blindness 77, 386-388.

Krischer, C. C., Stein-Arsic, M., Meissen, R., and Zihl, J. (1985). Visual

performance and reading capacity of partially sighted persons in a rehabilitation

center. Am J Optom Physiol Opt 62, 52-58.

Kuyk, T., Elliott, J. L., and Fuhr, P. S. (1998 (c)). Visual correlates of mobility in

real world settings in older adults with low vision. Optom Vis Sci 75, 538-547.

References

305

Kuyk, T., Elliott, J. L., and Fuhr, P. S. (1998 (b)). Visual correlates of obstacle

avoidance in adults with low vision. Optom Vis Sci 75, 174-182.

Kuyk, T., Elliott, J. L., and James, J. C. (1998 (a)). Drifting text with a reading stand.

J Vis Impairment Blindness 92, 669-674.

La Heij, E. C., Liem, A. T., and Hendrikse, F. (2001). Age-related macular

degeneration; therapeutic options. Ned Tijschr Geneeskd 145, 1390-1397.

LaGrow, S. J. (1986). Assessing optimal illumination for visual response accuracy in

visually impaired adults. J Vis Impairment Blindness 83, 888-895.

Lambert, H. M., Capone, A. J., Aaberg, T. M., Sternberg, P. J., Mandell, B. A., and

Lopez, P. F. (1992). Surgical excision of subfoveal neovascular membranes in age-

related macular degeneration. Am J Ophthalmol 113, 257-262.

Langmann, A., Lindner, S., and Kollegger, E. (1994). Low-Vision Training For

Better Usage of Magnifying Visual Aids. Ophthalmologica 208, 92-94.

Lanzetta, P., Michieletto, P., Pirracchio, A., and Bandello, F. (2002). Early vascular

changes induced by transpupillary thermotherapy of choroidal neovascularization.


Layton, C. A., and Koenig, A. J. (1998). Increasing reading fluency in elementary

students with low vision through repeated readings. J Vis Impairment Blindness

92, 276-292.

Leat, S. J., Fryer, A., and Rumney, N. J. (1994). Outcome of low vision aid

provision: the effectiveness of a low vision clinic. Optom Vis Sci 71, 199-206.

Leat, S. J., Legge, G. E., and Bullimore, M. A. (1999). What is low vision? A re-

evaluation of definitions. Optom Vis Sci 76, 198-211.

Leat, S. J., and Rumney, N. J. (1990). The experience of a university-based low

vision clinic. Ophthal Physiol Opt 10, 8-15.

Leat, S. J., and Woo, G. C. (1997). The validity of current clinical tests of contrast

sensitivity and their ability to predict reading speed in low vision. Eye 11, 893-899.

References

306

Leat, S. J., and Woodhouse, J. M. (1993). Reading performance with low vision aids:

relationship with contrast sensitivity. Ophthal Physiol Opt 13, 9-16.

Legge, G. E. (1991). Glenn A. Fry Award Lecture 1990: Three perspectives on low

vision reading. Optom Vis Sci 68, 763-769.

Legge, G. E., Ahn, S. J., Klitz, T. S., and Luebker, A. (1997). Psychophysics of

reading - XVI. The visual span in normal and low vision. Vision Res 37, 1999-2010.

Legge, G. E., Pelli, D. G., Rubin, G. S., and Schleske, M. M. (1985 (a)).

Psychophysics of reading - I. Normal vision. Vision Res 25, 239-252.

Legge, G. E., Ross, J. A., Isenberg, L. M., and LaMay, J. M. (1992). Psychophysics

of reading. Clinical predictors of low-vision reading speed. Invest Ophthalmol Vis

Sci 33, 677-687.

Legge, G. E., Ross, J. A., Luebker, A., and LaMay, J. M. (1989 (a)). Psychophysics

of reading - VIII. The Minnesota Low-Vision Reading Test. Optom Vis Sci 66, 843-

853.

Legge, G. E., Ross, J. A., Maxwell, K. T., and Luebker, A. (1989 (b)). Psychophysics

of reading - VII. Comprehension in normal and low vision. Clin Vision Sci 4, 51-60.

Legge, G. E., Rubin, G. S., and Luebker, A. (1987). Psychophysics of reading - V.

The role of contrast in normal vision. Vision Res 27, 1165-1177.

Legge, G. E., Rubin, G. S., Pelli, D. G., and Schleske, M. M. (1985 (b)).

Psychophysics of reading - II. Low vision. Vision Res 25, 253-265.

Legge, G. E., Rubin, G. S., Schleske, M. M., Luebker, A., and Ross, J. A. (1988).

Understanding low vision reading. J Vis Impairment Blindness 82, 54-59.

Lei, H., and Schuchard, R. A. (1997). Using two preferred retinal loci for different

lighting conditions in patients with central scotomas. Invest Ophthalmol Vis Sci 38,

1812-1818.

Leonard, R. (1999). Statistics on Vision Impairment: A Resource Manual. Arlene

R. Gordon Research Institute of Lighthouse International, New York.

References

307

Lewis, M. L. (1992). Clinical presentations of AMD. In: Age-Related Macular

Degeneration: Principles and Practice (G. R. Hampton and P. T. Nelsen, eds.).

New York: Raven Press, Ltd., p. 137-181.

Lie, I. (1977). Relation of visual acuity to illumination, contrast and distance in the

partially sighted. Am J Optom Physiol Opt 54, 528-536.

Liu, J. H., Cheng, C. Y., Chen, S. J., and Lee, F. L. (2001). Visual impairment in a

Taiwanese population: prevalence, causes, and socioeconomic factors. Ophthal

Epidemiol 8, 339-350.

Lott, L. A., Schneck, M. E., Haegerstrom-Portnoy, G., Brabyn, J. A., Gildengorin, G.

L., and West, C. G. (2001). Reading performance in older adults with good acuity.


Lovie-Kitchin, J., Mainstone, J., Robinson, J., and Brown, B. (1990). What areas of

the visual field are important for mobility in low vision patients? Clin Vision Sci 4,

249-263.

Lovie-Kitchin, J. E. (1988). Validity and reliability of visual acuity measurements.

Ophthal Physiol Opt 8, 363-370.

Lovie-Kitchin, J. E. (1993). Reading performance of people with low vision. J Vis

Rehab 7, 3.

Lovie-Kitchin, J. E. (1996). Reading Performance of Adults with Low Vision.

Unpublished PhD thesis. Queensland University of Technology, Australia, Brisbane

Lovie-Kitchin, J. E., Bowers, A. R., and Woods, R. L. (2000 (a)). Oral and silent

reading performance with macular degeneration. Ophthal Physiol Opt 20, 360-370.

Lovie-Kitchin, J. E., and Bowman, K. J. (1985). Senile Macular Degeneration -

Management and Rehabilitation. Boston: Butterworths.

Lovie-Kitchin, J. E., and Brown, B. (1986). Reaction time in age-related

maculopathy. Am J Optom Physiol Opt 63, 366-371.

References

308

Lovie-Kitchin, J. E., Oliver, J. N., Bruce, A., Leighton, S. M., and Leighton, K. W.

(1994). The effect of print size on reading rate for adults and children. Clin Exp

Optom 77, 2-7.

Lovie-Kitchin, J. E., and Whittaker, S. G. (1998 (a)). Relative-size magnification

versus relative-distance magnification: effect on the reading performance of adults

with normal and low vision. J Vis Impairment Blindness 92, 433-446.

Lovie-Kitchin, J. E., and Whittaker, S. G. (1998 (b)). Low vision assessment for

reading rehabilitation: indications for visual field assessment In: Vision '96, Book 1,

International Conference on Low Vision Proceedings. Madrid: Organización

Nacional de Ciegos Españoles, p. 268-275.

Lovie-Kitchin, J. E., and Whittaker, S. G. (1999 (a)). Interaction between acuity

reserve and window size for scrolled reading rates in adults with normal and low

vision. Invest Ophthalmol Vis Sci (Suppl) 40, 34.

Lovie-Kitchin, J. E., and Whittaker, S. G. (1999 (b)). Prescribing near magnification

for low vision patients. Clin Exp Optom 82, 214-224.

Lovie-Kitchin, J. E., and Whittaker, S. G. (2000). Prescribing magnification for

reading with low vision: what are the criteria? In: Vision Rehabilitation -

Assessment, Intervention and Outcomes (C. Stuen, A. Arditi, A. Horowitz, M. A.

Lang, B. Rosenthal, and K. Seidman, eds.). Lisse: Swets and Zeitlinger, p. 314-318.

Lovie-Kitchin, J. E., and Woo, G. C. (1987). Effect of magnification and field of

view on reading speed using a CCTV. In: Low Vision: Principles and Application

(G. C. Woo, ed.), New York: Springer-Verlag, p. 308-322.

Lowe, J. B., and Drasdo, N. (1990). Efficiency in reading with closed-circuit

television for low vision. Ophthal Physiol Opt 10, 225-233.

Machemer, R., and Steinhorst, U. H. (1993 (a)). Retinal separation, retinotomy and

macular relocation. I: experimental studies in rabbit eyes. Graefes Arch Clin Exp


References

309

Machemer, R., and Steinhorst, U. H. (1993 (b)). Retinal separation, retinotomy and

macular relocation. II: a surgical approach for age related macular degeneration.

Graefes Arch Clin Exp Ophthalmol 231, 635-641.

Macular Photocoagulation Study Group. (1986). Argon laser photocoagulation for

neovascular maculopathy. Three-year results from randomized clinical trials. Arch


Macular Photocoagulation Study Group. (1991). Argon laser photocoagulation for

neovascular maculopathy. Five-year results from randomized clinical trials. Arch

Ophthalmol 109, 1109-1114.

Macular Photocoagulation Study Group. (1994 (a)). Laser photocoagulation for

juxtafoveal choroidal neovascularization. Five-year results from randomized clinical

trials. Arch Ophthalmol 112, 500-509.

Macular Photocoagulation Study Group. (1994 (b)). Persistent and recurrent

choroidal neovascularization after laser photocoagulation for subfoveal choroidal

neovascularization of AMD. Arch Ophthalmol 112, 489-499.

Macular Photocoagulation Study Group. (1994 (c)). Visual outcome after laser

photocoagulation for subfoveal choroidal neovascularization secondary to age-

related macular degeneration. The influence of initial lesion size and initial visual

acuity. Arch Ophthalmol 112, 480-488.

Macular Photocoagulation Study Group. (1996). Occult choroidal

neovascularization. Influence on visual outcome in patients with age-related macular

degeneration. Arch Ophthalmol 114, 400-412.

Mancil, G. L., and Nowakowski, R. (1986). Evaluation of reading speed with four

low vision aids. Am J Optom Physiol Opt 63, 708-713.

Mangione, C. M., Berry, S., Spritzer, K., Janz, N. K., Klein, R., Owsley, C., and et

al. (1998). Identifying the content area for the 51-item National Eye Institute Visual

Function Questionnaire: results from focus groups with visually impaired persons.

Arch Ophthalmol 116, 227-233.

References

310

Mangione, C. M., Lee, P. P., Berry, S. H., Hays, R. D., Janz, N. K., Klein, R., and

Owsley, C. (1995). Development of a questionnaire for the assessment of vision

function. Invest Ophthalmol Vis Sci (Suppl) 95, 428.

Mangione, C. M., Lee, P. P., Gutierrez, P. R., Spritzer, K., Berry, S., Hays, R. D.,

and National Eye Institute Visual Function Questionnaire Field Test Investigators.

(2001). Development of the 25-item National Eye Institute Visual Function

Questionnaire. Arch Ophthalmol 119, 1050-1058.

Mangione, C. M., Phillips, R. S., Seddon, J. M., Lawrence, M. G., Cook, E. F.,

Dailey, R., and et al. (1992). Development of the 'Activities of Daily Vision Scale'. A

measure of visual functional status. Med Care 30, 1111-1126.

Mansfield, J. S., Ahn, S. J., Legge, G. E., and Leubker, A. (1993). A new reading -

acuity chart for normal and low vision In: Noninvasive Assessment of the Visual

System Technical Digest. Washington D. C.: Optical Society of America, p. 232-

235.

Mansfield, J. S., Legge, G. E., and Bane, M. C. (1996). Psychophysics of reading -

XV. Font effects in normal and low vision. Invest Ophthalmol Vis Sci 37, 1492-

1501.

Mansfield, J. S., Legge, G. E., Leubker, A., and Cunningham, K. (1994). MNRead

Acuity Charts. Lighthouse Low Vision Products, New York.

Marshall, J. (1985). Radiation and the ageing eye. Ophthal Physiol Opt 5, 241-261.

Martin, R., Coco, B., Ruben, C., Isabel, A., Mayo, A., Beatriz, M., and et al. (2002).

Study of the efficacy of a new electronic device for the ARMD: a new therapeutic

option. Invest Ophthalmol Vis Sci 43, E-Abstract 1205.

Massof, R. W., and Rubin, G. S. (2001). Visual function assessment questionnaires.

Surv Ophthalmol 45, 531-548.

McClure, M. E., Hart, P. M., Jackson, A. J., Stevenson, M. R., and Chakravarthy, U.

(2000). Macular degeneration: do conventional measurements of impaired visual

function equate with visual disability. Br J Ophthalmol 84, 244-250.

References

311

McIlwaine, G. G., Bell, J. A., and Dutton, G. N. (1991). Low vision aids - is our

service cost effective? Eye 5, 607-611.

McMahon, T. T., Hansen, M., and Viana, M. (1991). Fixation characteristics in

macular disease. Relationship between saccadic frequency, sequencing, and reading

rate. Invest Ophthalmol Vis Sci 32, 567-574.

McMahon, T. T., and Spigelman, V. (1989). Reading with a stand magnifier: 1

Effect of text configuration and experience on normal subjects. J Vis Rehab 3, 19-

23.

Mehr, E. B., and Freid, A. N. (1975). Low Vision Care. Chicago, Professional Press.

Meyers, S. M., Greene, T., and Gutman, F. A. (1995). A twin study of age-related

macular degeneration. Am J Ophthalmol 120, 757-766.

Meyers, S. M., and Zachary, A. A. (1988). Monozygotic twins with age-related

macular degeneration. Arch Ophthalmol 106, 651-653.

Midena, E., Degli Angeli, C., Blarzino, M. C., Valenti, M., and Segato, T. (1997).

Macular function impairment in eyes with early age-related macular degeneration.


Mitchell, P. (1991). Editorial - Changing patterns of blindness in Australia. Aust NZ

J Ophthalmol 19, 265-266.

Mitchell, S. M., Smith, W., Attebo, K., and Wang, J. J. (1995). Prevalence of age-

related maculopathy in Australia. The Blue Mountains Eye Study. Ophthalmology

102, 1450-1460.

Mohammed, Z., and Dickinson, C. M. (2000). The inter-relationship between

magnification, field of view and contrast reserve: the effect on reading performance.

Ophthal Physiol Opt 20, 464-472.

Munoz, B., and West, S. K. (2002). Blindness and visual impairment in the Americas

and the Caribbean. Br J Ophthalmol 86, 498-504.

References

312

Murray, W., and Scheme, L. K. W. R (2000). Maclure Bar Reading Book for

Children. London, Clement Clarke International Ltd. (http://www.clement-clarke-

webserver.co.uk/cci/cc_home.asp?pageID=cci_home).

Myrberg, M., Backman, O., and Lennerstrand, G. (1996). Reading proficiency of

elderly visually impaired persons after rehabilitation. J Vis Impairment Blindness

90, 341-351.

Nasrallah, F. P., Jalkh, A. E., Friedman, G. R., Trempe, C. L., McMeel, J. W., and

Schepens, C. L. (1988). Visual results with low-vision aids in age-related macular

degeneration. Am J Ophthalmol 106, 730-734.

National Advisory Eye Council. (1993). Vision Research - A National Plan: 1994-

1998. National Eye Institute, US Department of Health and Human Services.

National Institute on Disability and Rehabilitation Research. (1993). Consensus

Statement: Protocols for Choosing Low Vision Devices. Washington D.C., US

Department of Education.

National Research Council. (2002). Visual Impairments: Determining Eligibility

For Social Security Benefits. Washington D.C., National Academy Press.

Neale, M. D., McKay, M. F., and Barnard, J. (1999). Neale Analysis of Reading

Ability. Camberwell, Vic., Australian Council for Educational Research.

Neve, J. J. (1989 (a)). Reading with hand-held magnifiers. J Med Eng Technol 13,

68-75.

Neve, J. J. (1989 (b)). On the use of hand-held magnifiers during reading. Optom

Vis Sci 66, 440-449.

Nilsson, S. E. G., and Nilsson, U. L. (1994). Educational training in the use of aids

and residual vision is essential in rehabilitation of patients with severe age-related

macular degeneration. In: Low Vision: Research and New Developments in



References

313

Nilsson, U. L. (1986). Visual rehabilitation of patients with advanced diabetic

retinopathy. A follow-up study at the Low Vision Clinic, Department of

Ophthalmology, University of Linkoping. Doc Ophthalmol 62, 369-382.

Nilsson, U. L. (1988). Visual rehabilitation of patients with advanced stages of

glaucoma, optic atrophy, myopia or retinitis pigmentosa. Doc Ophthalmol 70, 363-

383.

Nilsson, U. L. (1990). Visual rehabilitation with and without educational training in

the use of optical aids and residual vision. A prospective study of patients with

advanced age-related macular degeneration. Clin Vision Sci 6, 3-10.

Nilsson, U. L., Frennesson, C., and Nilsson, S. E. (1998). Location and stability of a

newly established eccentric retinal locus suitable for reading, achieved through

training of patients with a dense central scotoma. Optom Vis Sci 75, 873-878.

Nilsson, U. L., and Nilsson, S. E. (1986). Rehabilitation of the visually handicapped

with advanced macular degeneration. A follow-up study at the Low Vision Clinic,

Department of Ophthalmology, University of Linkoping. Doc Ophthalmol 62, 345-

367.

Nork, T. M. (1995). Selective loss of blue cones and rods in human retinal

detachment. Arch Ophthalmol 113, 1066-1073.

Nowakowski, R. W. (1994). Primary Low Vision Care. Connecticut: Appleton &

Lange.

Oppenheim, A. N. (1992). Questionnaire Design, Interviewing and Attitude

Measurement. London: Printers Publishers Ltd.

Ortiz, A., Chung, S. T., Legge, G. E., and Jobling, J. T. (1999). Reading with a head-

mounted video magnifier. Optom Vis Sci 76, 755-763.

Oxford Progressive English Readers Grade 3. (1995(a)). Hong Kong, Oxford

University Press.

Oxford Progressive English Readers Grade 4. (1995(b)). Hong Kong, Oxford

University Press.

References

314

Owsley, C., Jackson, G. R., Cideciyan, A. V., Huang, Y., Fine, S. L., Ho, A. C., and

et al. (2000). Psychophysical evidence for rod vulnerability in age-related macular

degeneration. Invest Ophthalmol Vis Sci 41, 267-273.

Pallant, J. (2001). SPSS Survival Manual: A Step-By-Step Guide To Data

Analysis Using SPSS for Windows (Version 10). Buckingham: Open University

Press.

Patel, B., Elliott, D. B., and Whitaker, D. (2001). Optimal reading speed in simulated

cataract: development of a potential vision test. Ophthal Physiol Opt 21, 272-276.

Pelli, D. G., Robson, J. G., and Wilkins, A. J. (1988). The design of a new letter chart

for measuring contrast sensitivity. Clin Vision Sci 2, 187-199.

Percy, M., and Hindle, R. (1989). New method for the non-invasive three-

dimensional measurement of human back movement. Clin Biomech 4, 73-79.

Pertile, G., and Claes, C. (2002). Macular translocation with 360 degree retinotomy

for management of age-related macular degeneration with subfoveal choroidal

neovascularization. Am J Ophthalmol 134, 560-565.

Plass, R., and Yager, D. (1995). Presentation of video-based text to low vision

readers: Rapid serial visual presentation vs full-page reading. Invest Ophthalmol

Vis Sci (Suppl) 36, 71.

Polhemus 3SPACE Isotrak User's Manual. (1987). Vermont, OPM3704-001.

Postgens, H., Bodanowitz, S., and Kroll, P. (1997). Low-dose radiation therapy for

age-related macular degeneration. Graefes Arch Clin Exp Ophthalmol 235, 656-

661.

Prevent Blindness America. (1998). [Online]. Available:

http://www.preventblindness.org/eye_problems.html.

Raasch, T. W., Leat, S. J., Kleinstein, R. N., Bullimore, M. A., and Cutter, G. R.

(1997). Evaluating the value of low-vision services. J Am Optom Assoc 68, 287-

295.

References

315

Raasch, T. W., and Rubin, G. S. (1993). Reading with low vision. J Am Optom

Assoc 64, 15-18.

Rayner, K. (1979). Reading without a fovea. Science 206, 468-469.

Rayner, K., Well, A. D., Pollatsek, A., and Bertera, J. H. (1982). The availability of

useful information to the right of the fixation in reading. Percept Psychophys 31,

537-550.

Rechtman, E., Ciulla, T. A., Criswell, M. H., Pollack, A., and Harris, A. (2002). An

update on photodynamic therapy in age-related macular degeneration. Expert Opin

Pharmacother 3, 931-938.

Richer, S. (1996). Multicenter ophthalmic and nutritional age-related macular

degeneration study - Part 2: Antioxidant intervention and conclusions. J Am Optom

Assoc 67, 30-49.

Richer, S. (1999). A protocol for the evaluation and treatment of atrophic age-related

macular degeneration. J Am Optom Assoc 70, 13-23.

Robbins, H. G. (1981). Low vision care for the over 80's. Aust J Optom 64, 243-

251.

Robbins, H. G., and Murray, N. E. (1988). Psychological and visual factors in low

vision rehabilitation of patients with age related maculopathy. J Vis Rehab 2, 11-21.

Rothman, K. J. (1990). No adjustments are needed for multiple comparisons.

Epidemiology 1, 43-46.

Royal National Institute for the Blind. (1997). Office of National Statistics mid

1996 population Estimates: estimates for 1996 of visually impaired people (i.e.

registrable) and the number of people registered as a blind and partially sighted

as at 31 March 1997 in United Kingdom.

http://www.rnib.org.uk/wesupply/fcsheet/authuk.htm (updated to December, 2000).

Rubin, G. S. (1987). Predicting reading performance in low vision observers with

age related maculopathy (ARM). In: Low Vision: Principles and Application (G.

C. Woo, ed.). New York: Springer-Verlag, p. 323-333.

References

316

Rubin, G. S., and Legge, G. E. (1985). Predicting low-vision reading rates from

measures of contrast sensitivity. In: Digest of Topical Meeting on Non-invasive

Assessment of Visual Function, Paper WB5. Washington D.C: Optical Society of

America, p. 1-4.

Rubin, G. S., and Legge, G. E. (1989). Psychophysics of reading - VI. The role of

contrast in low vision. Vision Res 29, 79-91.

Rubin, G. S., Roche, K. B., Prasada-Rao, P., and Fried, L. P. (1994). Visual

impairment and disability in older adults. Optom Vis Sci 71, 750-760.

Rubin, G. S., and Turano, K. A. (1992). Reading without saccadic eye movements.

Vision Res 32, 895-902.

Rubin, G. S., and Turano, K. A. (1994). Low vision reading with sequential word

presentation. Vision Res 34, 1723-1733.

Rumney, N. J. (1995). Using visual thresholds to establish low vision performance.

Ophthal Physiol Opt (Suppl) 15, S18-24.

Rumney, N. J., and Leat, S. J. (1994). Why do low vision patients still read slowly

with a low vision aid? In: Low Vision - Research and New Developments in



Safran, A. B., Duret, F., Issenhuth, C., and Mermoud, C. (1999). Full text reading

with central scotoma: pseudo regression and pseudo line losses. Br J Ophthalmol

83, 1341-1347.

Salive, M. E., Guralnik, J., Christen, W., Glynn, R. J., Colsher, P., and Ostfeld, A. M.

(1992). Functional blindness and visual impairment in older adults from three

communities. Ophthalmology 99, 1840-1847.

Samuels, S. J. (1979). The method of repeated readings. The Reading Teacher 33,

403-408.

Schuchard, R. A. (1995). Adaptation to macular scotomas in persons with low vision.

Am J Occup Ther 49, 870-876.

References

317

Schuchard, R. A., and Fletcher, D. C. (2000). Preferred retinal locus and the scanning

laser ophthalmoscope. In: Principles and Practice of Ophthalmology (D. M.

Albert, F. A. Jakobiec, D. T. Azar, and E. S. Gragoudas, eds.). Pennsylvania: W.B.

Saunders Company, p. 5438-5443.

Schuchard, R. A., Naseer, S., and de Castro, K. (1999). Characteristics of AMD

patients with low vision receiving visual rehabilitation. J Rehabil Res Dev 36, 294-

302.

Scupola, A., Coscas, G. J., Soubrane, G., and Balestrazzi, E. (1999). Natural history

of macular subretinal hemorrhage in age-related macular degeneration.

Ophthalmologica 213, 97-102.

Shankweiler, D., and Liberman, I. (1972). Mis-reading: A search for causes. In:

Language By Ear and Eye (J. Cavanagh and I. Mattingly, eds.). Cambridge: M.I.T.

Press.

Shuttleworth, G. N., Dunlop, A., Collins, J. K., and James, C. R. (1995). How

effective is an integrated approach to low vision rehabilitation? Two year follow up

results from south Devon. Br J Ophthalmol 79, 719-723.

Silvestri, G., Johnston, P. B., and Hughes, A. E. (1994). Is genetic predisposition an

important risk factor in age-related macular degeneration? Eye 8, 564-568.

Sloan, L. L., and Brown, D. J. (1962). Progressive retinal degeneration with selective

involvement of the cone mechanism. Am J Ophthalmol 54, 620-641.

Sloan, L. L., and Habel, A. (1965). Reading aids for the partially blind. Am J Optom


Sloan, L. L., Habel, A., and Feiock, K. (1973). High illumination as an auxiliary

reading aid in diseases of the macula. Am J Ophthalmol 76, 745-757.

Sloan, L. S., and Habel, A. (1973). Reading speeds with textbooks in large and in

standard print. The Sight-Saving Review 43, 107-111.

Smith, F. (1978). Understanding Reading. New York: Cambridge University Press.

References

318

Smith, W., Assink, J., Klein, R., Mitchell, P., Klaver, C. C., Klein, B. E., and et al.

(2001). Risk factors for age-related macular degeneration: Pooled findings from three

continents. Ophthalmology 108, 697-704.

Smith, W., and Mitchell, P. (1998). Family history and age-related maculopathy: the

Blue Mountains Eye Study. Aust N Z J Ophthalmol 26, 203-206.

Solan, H. A., Feldman, J., and Tujak, L. (1995). Developing visual and reading

efficiency in older adults. Optom Vis Sci 72, 139-145.

Song, S., White, J. M., and Jose, R. T. (1996). Monocular versus binocular reading

speed in low vision patients. J Videology 1, 21-26.

Spitzberg, L., Jose, R. T., and Kuether, C. (1989). A new ergonomically designed

prism stand magnifier. J Vis Rehab 3, 47-51.

Stalmans, P., Leys, A., and Van Limbergen, E. (1997). External beam radiotherapy

(20 Gy, 2 Gy fractions) fails to control the growth of choroidal neovascularization in

age-related macular degeneration: a review of 111 cases. Retina 17, 481-492.

Starr, C. E., Guyer, D. R., and Yannuzzi, L. A. (1998). Age-related macular

degeneration. Can we stem this worldwide public health crisis? Postgrad Med 103,

153-156.

Steinmetz, R. L., Haimovici, R., Jubb, C., Fitzke, F. W., and Bird, A. C. (1993).

Symptomatic abnormalities of dark adaptation in patients with age-related Bruch's

membrane change. Br J Ophthalmol 77, 549-554.

Stelmack, J., Reda, D., Ahlers, S., Bainbridge, L., and McCray, J. (1991). Reading

performance of geriatric patients post exudative maculopathy. J Am Optom Assoc

62, 53-57.

Sterns, G. (1994). Low vision rehabilitation in the elderly. Ophthalmol Clin North

Am 7, 231-236.

Stoll, S., Sarma, S., and Hoeft, W. W. (1995). Low vision aids training in the home.

J Am Optom Assoc 66, 32-38.

References

319

Sunness, J. S. (1999). The natural history of geographic atrophy, the advanced

atrophic form of age-related macular degeneration. Mol Vis 5, 25.

Sunness, J. S., Applegate, C. A., Haselwood, D., and Rubin, G. S. (1996). Fixation

patterns and reading rates in eyes with central scotomas from advanced atrophic age-

related macular degeneration and Stargardt disease. Ophthalmology 103, 1458-

1466.

Sunness, J. S., Bressler, N. M., Applegate, C. A., and Chukwueke, U. (2002). Long-

term progression of visual acuity in age-related geographic atrophy from AMD.

Invest Ophthalmol Vis Sci 43, E-Abstract 2905.

Sunness, J. S., Gonzalez-Baron, J., Applegate, C. A., Bressler, N. M., Tian, Y.,

Hawkins, B., and et al. (1999). Enlargement of atrophy and visual acuity loss in the

geographic atrophy form of age-related macular degeneration. Ophthalmology 106,

1768-1779.

Sunness, J. S., Rubin, G. S., Applegate, C. A., Bressler, N. M., Marsh, M. J.,

Hawkins, B. S., and et al. (1997). Visual function abnormalities and prognosis in

eyes with age-related geographic atrophy of the macula and good visual acuity.


Swann, P. G., and Lovie-Kitchin, J. E. (1990). Age-related maculopathy. I: A review

of its morphology and effects on visual function. Ophthal Physiol Opt 10, 149-158.

Taylor, H. R. (2001). World blindness: 21st century perspective. Br J Ophthalmol

85, 261-266.

Taylor, H. R., Livingston, P. M., Stanislavsky, Y. L., and McCarty, C. A. (1997).

Visual impairment in Australia: Distance visual acuity, near vision and visual field

findings of the Melbourne Visual Impairment Project. Am J Ophthalmol 123, 328-

337.

Taylor, H. R., McCarty, C. A., and Nanjan, M. B. (2000). Vision impairment predicts

five-year mortality. Trans Am Ophthalmol Soc 98, 91-99.

References

320

Temel, A. (1989). Low vision aids (evaluation of 185 patients). Ophthal Physiol

Opt 9, 327-331.

The Lighthouse Inc. (1995). The Lighthouse National Survey on Vision Loss.

http://www.lighthouse.org/research_nationalsurvey.htm.(March, 2003).

Thomas, M. A., and Kaplan, H. J. (1991). Surgical removal of subfoveal

neovascularization in the presumed ocular histoplasmosis syndrome. Am J


Thompson, J. R., Du, L., and Rosenthal, A. R. (1989). Recent trends in the

registration of blindness and partial sight in Leicestershire. Br J Ophthalmol 73, 95-

99.

Thylefors, B., Negrel, A. D., Pararajasegaram, R., and Dadzie, K. Y. (1995). Global

data on blindness. Bull World Health Organisation 73, 115-121.

Tielsch, J. M., Sommer, A., Witt, K. A., Katz, J., and Royall, R. M. (1990).

Blindness and visual impairment in an American urban population. The Baltimore

Eye Survey. Arch Ophthalmol 108, 286-290.

Timberlake, G. T., Blakey, P., Bertera, J. H., and Peli, E. (1989). Oculomotor control

of retinal text position by foveal and extra-foveal readers. Invest Ophthalmol Vis

Sci (Suppl) 30, 398.

Timberlake, G. T., Mainster, M. A., Peli, E., Augliere, R. A., Essock, E. A., and

Arend, L. E. (1986). Reading with a macular scotoma. I. Retinal location of scotoma

and fixation area. Invest Ophthalmol Vis Sci 27, 1137-1147.

Toth, C. A., and Machemer, R. (1999). Macular translocation. In: Age-Related

Macular Degeneration (J. W. Berger, S. L. Fine, and M. G. Maguire, eds.),

Philadelphia: Mosby Inc, p. 353-362.

Trauzettel-Klosinski, S., Teschner, C., Tornow, R. P., and Zrenner, E. (1994).

Reading strategies in normal subjects and in patients with macular scotoma assessed

by two new methods of registration. Neuro-ophthalmology 14, 15-30.

References

321

Trauzettel-Klosinski, S., Tornow, R. P., Teschner, C., and Zrenner, E. (1995).

Fixation behavior and reading in patients with macular scotoma assessed by SLO and

Tubingen manual perimetry. Invest Ophthalmol Vis Sci (Suppl) 36, S674.

Trott, P. H., Pearcy, M. J., Ruston, S. A., Fulton, I., and Brien, C. (1996). Three-

dimensional analysis of active cervical motion - the effect of age and gender. Clin

Biomech 11, 201-206.

van der Wildt, G. J., Kooijman, A. C., van Roode, F. W. J., and Cornelissen, F. W.

(1994). A simulator for studying the lighting needs of the visually impaired people.

In: Low Vision - Research and New Developments in Rehabilitation (A. C.

Kooijman, P. L. Looijestin, J. A. Welling, and G. J. van der Wildt, eds.).

Netherlands: IOS Press, p. 227-231.

Van Newkirk, M. R., Nanjan, M. B., Wang, J. J., Mitchell, P., Taylor, H. R., and

McCarty, C. A. (2000). The prevalence of age-related maculopathy. The Visual

Impairment Project. Ophthalmology 107, 1593-1600.

Van Newkirk, M. R., Weih, L. M., McCarty, C. A., and Taylor, H. R. (2001). Cause-

specific prevalence of bilateral visual impairment in Victoria, Australia. The Visual


Van Rens, G. H., Chmielowski, R. J., and Lemmens, W. A. (1991). Results obtained

with low vision aids. A retrospective study. Doc Ophthalmol 78, 205-210.

VanNewkirk, M. R., Nanjan, M. B., Wang, J. J., Mitchell, P., Taylor, H. R., and

McCarty, C. A. (2000). The prevalence of age-related maculopathy. The Visual


Varmus, H. (1997). Age-related Macular Degeneration: Status of Research.

National Eye Institute (http://www.nei.nih.gov/news/varmus.htm: updated to April

2001).

Verteporfin in Photodynamic Therapy Study Group. (2001 (a)). Photodynamic

therapy of subfoveal choroidal neovascularization in pathologic myopia with

verteporfin. 1-year results of a randomized clinical trial - VIP report no. 1.


References

322

Verteporfin In Photodynamic Therapy Study Group. (2001 (b)). Verteporfin therapy

of subfoveal choroidal neovascularization in age-related macular degeneration: two-

year results of a randomized clinical trial including lesions with occult with no

classic choroidal neovascularization - verteporfin in photodynamic therapy report 2.

Am J Ophthalmol 131, 541-560.

Vinding, T. (1995). Age-related macular degeneration. An epidemiological study of

1000 elderly individuals. With reference to prevalence, funduscopic findings, visual

impairment and risk factors. Acta Ophthalmol Scand 217, 1-32.

Vingerling, J. R., Dielemans, I., Hofman, A., Grobbee, D. E., Hijmering, M.,

Kramer, C. F., and et al. (1995). The prevalence of age-related maculopathy in the

Rotterdam Study. Ophthalmology 102, 202-210.

Virtanen, P., and Laatikainen, L. (1991). Primary success with low vision aids in

age-related macular degeneration. Acta Ophthalmol 69, 484-490.

Virtanen, P., and Laatikainen, L. (1993). Low vision aids in age-related macular

degeneration. Curr Opin Ophthalmol 4, 33-35.

von Noorden, G. K., and Mackensen, G. (1962). Phenomenology of eccentric

fixation. Am J Ophthalmol 53, 642-661.

Waiss, B., and Cohen, J. M. (1991). Modification of common low vision devices for

uncommon needs. J Am Optom Assoc 62, 65-68.

Watson, G., Baldasare, J., and Whittaker, S. (1990). The validity and clinical uses of

the Pepper Visual Skills for reading test. J Vis Impairment Blindness 84, 119-123.

Watson, G. R., De l'Aune, W., Long, S., Maino, J., and Stelmack, J. (1997).

Veterans' use of low vision devices for reading. Optom Vis Sci 74, 260-265.

Watson, G. R., and Whittaker, S. G. (1994). Rate and comprehension reading skills

in individuals with macular degeneration. In: Low Vision - Research and New

Developments in Rehabilitation (A. C. Kooijman, P. L. Looijestijn, J. A. Welling,

and G. J. van der Wildt, eds.). Amsterdam: IOS Press, p. 279-282.

References

323

Watson, G. R., Wright, V., and De l'Aune, W. (1992). The efficacy of

comprehension training and reading practice for print readers with macular loss. J

Vis Impairment Blindness 96, 37-43.

Weale, R. A. (1963). The Aging Eye. London: HK Lewis & Co.

Weih, L. M., VanNewkirk, M. R., McCarty, C. A., and Taylor, H. R. (2000). Age-

specific causes of bilateral visual impairment. Arch Ophthalmol 118, 264-269.

Weleber, R. G., and Tobler, W. R. (1986). Computerized quantitative analysis of

kinetic visual fields. Am J Optom Physiol Optom 101, 461-468.

Whittaker, S. G., Budd, J., and Cummings, R. W. (1988). Eccentric fixation with

macular scotoma. Invest Ophthalmol Vis Sci 29, 268-278.

Whittaker, S. G., and Cummings, R. (1990). Eccentric eye movements with loss of

central vision In: Proceedings of the International Conference on Low Vision

1990: Low Vision Ahead. Melbourne: Association for the Blind, p. 67-73.

Whittaker, S. G., Cummings, R. W., and Swieson, L. (1991). Saccade control

without a fovea. Vision Res 31, 2209-2218.

Whittaker, S. G., and Lovie-Kitchin, J. E. (1993). Visual requirements for reading.


Whittaker, S. G., and Lovie-Kitchin, J. E. (1994). The assessment of contrast

sensitivity and contrast reserve for reading rehabilitation. In: Low Vision - Research

and New Developments in Rehabilitation (A. C. Kooijman, P. L. Looijestijn, J. A.

Welling, and G. J. van der Wildt, eds.). Amsterdam: IOS Press, p. 88-92.

Williams, D. R. (1999). Nonoptical and accessory devices. In: Essentials of Low

Vision Practice (R. L. Brilliant, ed.). Boston: Butterworth Heinemann, p. 269-299.

Wolffsohn, J. S., and Cochrane, A. L. (1999). The changing face of the visually

impaired: the Kooyong low vision clinic's past, present, and future. Optom Vis Sci

76, 747-754.

References

324

Woo, G. C. (1985). Contrast sensitivity function as a diagnostic tool in low vision.

Am J Optom Physiol Opt 62, 648-651.

Woo, G. C. (2001). The term magnification. Clin Exp Optom 84, 113-119.

Woods, R. L., and Lovie-Kitchin, J. E. (1995). The reliability of visual performance

measures in low vision. Vision Science and Its Applications 1, 246-249.

Woods, R. L., and Woods, J. M. (1995). The role of contrast sensitivity charts and

contrast letter charts in clinical practice. Clin Exp Optom 78, 43-57.

World Health Organization. (1980 (a)). The International Classification of

Diseases, 9th revision, Clinical Modification: ICD-9-CM. Washington, D.C., US

Government Printing Office.

World Health Organization. (1980 (b)). In: International Classification of

Impairments, Disabilities and Handicaps: A Manual of Classification Relating

to the Consequences of Disease. Geneva: World Health Organisation:

World Health Organization. (1997). Global Initiative for The Prevention of

Avoidable Blindness. Geneva: World Health Organization.

World Health Organization. (2001 (a)). ICF : International Classification of

Functioning, Disability and Health. Geneva, World Health Organization.

World Health Organization. (2001 (b)). Prevention of Blindness. Geneva, World

Health Organization. (http://www.who.int /pbd/pbl/pbl_home.htm).

Yager, D., Aquilante, K., and Plass, R. (1998). High and low luminance letters,

acuity reserve, and font effects on reading speed. Vision Res 38, 2527-2531.

Yoshida, A., Yoshida, M., Yoshida, S., Shiose, S., Hiroishi, G., and Ishibashi, T.

(2000). Familial cases with age-related macular degeneration. Jpn J Ophthalmol 44,

290-295.

Zainal, M., Ismail, S. M., Ropilah, A. R., Elias, H., Arumugam, G., Alias, D., and et

al. (2002). Prevalence of blindness and low vision in Malaysian population: results

from the National Eye Survey 1996. Br J Ophthalmol 86, 951-956.

References

325

Zimmerman, G. J. (1992). Orientation and mobility training: Enhancing the

employment prospects for persons with blindness and visual impairments. Journal

of Vocational Rehabilitation 2, 66-72.

Zimmerman, G. J. (1996). Optics and low vision devices. In: Foundations of Low

Vision: Clinical and Functional Perspectives (A. L. Corn and A. J. Koeing, eds.).

New York: AFB Press, p. 115-142.

Zumber, B. L., and Wetzel, P. A. (1981). Eye movement determinants of reading

rate. In: Models of Oculomotor Behavior and Control (B. L. Zuber, ed.). Boca

Raton: CRC Press, p. 193-2308.

Appendices

326

APPENDICES

Appendix 1 Examples of Bailey-Lovie reading charts and

passages for different studies

Appendix 2 Sample size (power)

Appendix 3 Calculation of visual field loss

Appendix 4 Measurement of optical parameters of stand

magnifiers

Appendix 5 Questionnaires for Chapters 4 and 5

Appendix 6 Calibration of Isotrak (Chapter 5)

Appendix 1

327

Appendix 1 Examples of Bailey-Lovie text reading charts and passages for different studies

Figure A1.1 Bailey-Lovie text reading chart

5M

My father takes me to school every day in his big green car

4M Everyone wanted to go outside when the rain finally stopped

3.2M They were not able to finish playing the game before dinner

2.5M My father asked me to help the two men carry the box inside

2M The three elephants in the circus walked around very slowly

1.6M We could not guess what was inside the big box on the table

1.25M The two friends did not know what time the play would start

BAILEY-LOVIE READING CHART MN Text version T3

1.0M You must make the cake with four eggs and a little milk

0.8M Her father said that he was as

hungry as nine hundred bears

0.63M He wants to play in the new boat before we swim in the lake

0.50M The baby laughs as his mother splashes water on his big toe

0.4 M My teacher told my brother to study the big words carefully

0.32M Most of my friends came to the surprise birthday party here

0.25M Mother said she did not know why little boys cannot be nice

Appendix 1

328

Figure A1.2 Example of a passage used in Chapter 3 to measure reading rate with magnifiers Figure A1.3 Example of a passage used in Chapter 4 to measure reading rate on large print without stand magnifiers

But they are mortified when they find themselves at an evening party with many men in

bow ties and discover that their bow tie is the worst. One way of solving the bow tie

problem is to look for a girl who can tie a bow tie and request a demonstration. Print size: N12 Number of characters: 252

He moved forward a few steps; the house was so dark behind him, the world so dim and uncertain in front of him, that for a moment his heart failed him. He might have to search the whole garden for the dog. Then he heard a sniff, felt something wet against his leg.

Print size: N16 Number of characters: 265

Appendix 1

329

Figure A1.4 Example of a passage used in Chapter 4 to measure reading rate with magnifiers

It is a sunny summer day. The children are away at the sea for the day. The sea is not far from where they live. Here they are on the rocks by the sea. They have two friends with them. The two girls sit on the rocks while they watch the boys. Peter and Bob are going to swim under the water. They are not going far under the water, so there is little danger. The water is not deep. Daddy is on the sands not far away. He watches them while he sits in the sun. Fred is out in his boat to catch some fish and he watches the children also. Peter goes under first, while the others watch from the rocks. He wants to explore by himself, and swims about for some time. Then he comes up to tell them about it. Now Peter watches while Bob is doing the same. When he comes up he says, "I saw the fish, but it soon hid by the rocks again." The sun is in the sky and it is a warm day. There is an island nearby and the children plan to explore it. Before they do so, the girls say they will have a swim. Print size: N10 Number of characters: 993

Appendix 1

330

Figure A1.5 Example of a passage used in Chapter 5 to measure reading rate with magnifiers

It was not long before David returned to the fields. The harvest was over now, so they beat the grain and planted the fields with seed for the winter crops. They worked at this all day, and their little son lay asleep on the ground beside them. Winter came, but they were ready for it. The harvests had been the best Peter had ever known, and the small house was so full of grain and food that it was almost bursting. There were dried onions and garlic hanging from the roof, and great jars of wheat and rice. Most of this food would be sold later, but there was no hurry. David did not waste his money. He was careful and did not need to sell his food until the prices were good. He liked to save it until the snow came at New Year, when people would pay well for food. Many farmers had to sell their grain as soon as the harvest was over because they had wasted their money and had none left. David's uncle often had to sell his crops before they were even ripe. Sometimes he even sold them when they were still standing in the field. Print size: N12 Number of characters: 1036

Appendix 2

331

Appendix 2. Statistical power of experiments reported in each

chapter

Appendix 2a Statistical power of experiments reported in Chapter 2

In the calculation statistical power, determination of α- and β-values are necessary.

The α-value, or probability of a Type I error, is the acceptable probability of finding

a significant difference in the sample being tested, when in fact there is no difference

in the overall populations. To be conservative, an α-value for a two-sided test (0.05)

has been recommended. The β-value, or probability of a Type II error, is the

acceptable risk of missing a real difference in the overall populations. The power of

the study is the probability of correctly finding a ‘significant’ difference in the

samples being measured, when the true mean difference is equal to a minimum

difference of the reading rates. 95% power (β) is usually considered desirable for

statistical calculations (De Land and Chase, 1992). The sample size of subjects that

would be required for sufficient statistical power was calculated based on the

standard deviations of reading rates from previous studies (Bowers, 1998 (b)) and the

minimum difference in reading rate that was considered to be significant by a

program called GraphPad InStat - version 2.

Table A2.1 Results of power analysis: minimum number of subjects per

group for various levels of alpha and beta (for Chapter 2) (a) Log reading rate

Minimum difference to be detected as significant 0.05 log wpm Estimated standard deviation of the population 0.08 log wpm

α 0.1 0.05 0.02 0.01 Power β

80 0.2 16 21 26 30 90 0.1 22 27 34 39 95 0.05 28 34 41 46 Minimum difference to be detected as significant 0.07 log wpm Estimated standard deviation of the population 0.08 log wpm

α 0.1 0.05 0.02 0.01 Power β

80 0.2 9 11 14 16 90 0.1 12 14 18 20 95 0.05 15 17 21 24

Appendix 2

332

(b) Reading rate

Minimum difference to be detected as significant 25 wpm Estimated standard deviation of the population 38 wpm

α 0.1 0.05 0.02 0.01 Power β

80 0.2 15 19 24 27 90 0.1 20 25 31 35 95 0.05 26 31 37 42

Minimum difference to be detected as significant 35 wpm Estimated standard deviation of the population 38 wpm

α 0.1 0.05 0.02 0.01 Power β

80 0.2 8 10 12 14 90 0.1 11 13 16 18 95 0.05 13 16 19 22

A sample size of eighteen subjects being recruited in Chapter 2 was sufficient to give

a power (β) of 95% for a two-sided test (α=0.05) if 0.07 log wpm was considered as

the minimum difference in reading rate for a significant finding.

Appendix 2b Statistical power of experiments reported in Chapter 4

Reading rate varies greatly especially among people with AMD (Sunness et al.,

1996; Lovie-Kitchin et al., 2000 (a); Bowers et al., 2001 (b); Martin et al., 2002).

The reading rates measured in this study were no exception. A large standard

deviation was found, which was almost half the mean of each measure. For example,

Lovie-Kitchin et al. (2000 (a)) reported that mean reading rate on passages for

subjects with macular degeneration was 79 ± 35 wpm. Similarly, the mean reading

rate for subjects in Chapter 4 at week 2 was 87.1 ± 43.8 wpm (1.87 ± 0.26 log wpm).

Figures A2.1 to A2.4 show the reading rates with and without stand magnifiers for

subjects who participated in each group - control, practice and clinical - in the

longitudinal study described in Chapter 4.

Because of this wide variation in reading rates between subjects, a large sample size

would be needed to give statistically significant results. The sample size of subjects

that would be required for sufficient statistical power was calculated based on the

standard deviations of reading rates in low vision populations from previous studies

Appendix 2

333

(Lovie-Kitchin et al., 2000 (a); Cheong et al., 2002) and the minimum difference in

reading rate that was considered to be significant by a program called GraphPad

InStat - version 2.

The mean and standard deviation of the reading rate without stand magnifier in

Chapter 4 was very similar to the means and standard deviations of the reading rates

found in Chapter 3 and Lovie-Kitchin et al. (2000 (a)), therefore using data from

either of these earlier studies was a good basis for the power calculation to determine

the preferred sample size for the Chapter 4 study. For the subjects who participated

in the study described in Chapter 3, the mean log reading rate on large print without

magnifier was approximately 2.0 log wpm (100 wpm) with a standard deviation of

0.23 log wpm (45 wpm); these values were used in the power calculation for the

Chapter 4 study. The sample size calculated using log reading rate was slightly

different from that using simple reading rate. However the reading rates analysed in

the thesis were transformed to log reading rate for analyses purposes, so sample size

determined by log reading rate was used.

Results of the power calculation indicated that a sample size of 138 subjects per

group was needed to give a power (β) of 95% for a two-sided test (α=0.05) if 0.1 log

wpm was considered as the minimum difference in reading rate for a significant

finding (Table A2.2). A smaller sample of 62 subjects per group would be required

to detect a minimum difference of 0.15 log wpm in reading rates with sufficient

power for significant statistical analyses.

Due to difficulties in recruitment, a sample size of only 43 subjects with

approximately 10 subjects in each group were recruited for the study in Chapter 4.

This sample size could only give a power (β) of 95% for a two-sided test (α=0.05) to

detect a minimum difference of 0.35 log wpm as a significant finding. Because of the

insufficient number of participants, a number of the results from Chapter 4 indicated

a trend but failed to reach statistical significance. With a larger subject sample size,

these results might become statistically significant.

Appendix 2

334

Figure A2.1 Reading rates with and without stand magnifiers for the control group (N) As a consequence of substantial vision deterioration for Subject 22 in the control group, reading rates with and without STM could not be measured. Reading rates for this subject at week 20 were not displayed in the graphs. LP = reading on large print STM = reading with stand magnifier 0 = week 0 1 = week 1 2 = week 2 4 = week 4 8 = week 8 20 = week 20

Reading rate without STM (control)

1

1.2

1.4

1.6

1.8

2

2.2

2.4

LP-0 LP-1 LP-2 LP-20

Log

read

ing

rate

(wpm

)

13

14

16

17

18

19

20

21

22

23

Reading rate with STM (control)

1

1.2

1.4

1.6

1.8

2

2.2

2.4


Log

read

ing

rate

(wpm

)

13

14

16

17

18

19

20

21

22

23

Appendix 2

335

Figure A2.2 Reading rates with and without stand magnifiers for the large print practice group (P1) LP = reading on large print STM = reading with stand magnifier 0 = week 0 1 = week 1 2 = week 2 4 = week 4 8 = week 8 20 = week 20

Reading rate without STM (P1)

1

1.2

1.4

1.6

1.8

2

2.2

2.4


Log

read

ing

rate

(wpm

)

24

25

26

27

28

29

30

31

32

33

34


1

1.2

1.4

1.6

1.8

2

2.2

2.4

STM-0 STM-1 STM-2 STM-4 STM-8 STM-20Lo

g re

adin

g ra

te (w

pm)

24

25

26

27

28

29

30

31

32

33

34

Appendix 2

336

Figure A2.3 Reading rates with and without stand magnifiers for the large print with reduced field of view practice group (P2) LP = reading on large print STM = reading with stand magnifier 0 = week 0 1 = week 1 2 = week 2 4 = week 4 8 = week 8 20 = week 20

Reading rate without STM (P2)

1

1.2

1.4

1.6

1.8

2

2.2

2.4


Log

read

ing

rate

(wpm

)

1

2

3

4

5

6

7

9

10

11

12


1

1.2

1.4

1.6

1.8

2

2.2

2.4

STM-0 STM-1 STM-2 STM-4 STM-8 STM-20L

og

rea

din

g r

ate

(wp

m)

1

2

3

4

5

6

7

9

10

11

12

Appendix 2

337

Figure A2.4 Reading rates with and without stand magnifiers for the clinical group (C) LP-2 = reading on large print at week 2 LP-20 = reading on large print at week 20 STM = reading with stand magnifier 2 = week 2 4 = week 4 8 = week 8 20 = week 20

Reading rate without STM (clinical)

1

1.2

1.4

1.6

1.8

2

2.2

LP-2 LP-20

Log

read

ing

rate

(wpm

)

3536373940414243444546

Reading rate with STM (clinical)

1

1.2

1.4

1.6

1.8

2

2.2

STM-2 STM-4 STM-8 STM-20Lo

g re

adin

g ra

te (w

pm)

3536373940414243444546

Appendix 2

338

Table A2.2 Results of power analysis: minimum number of subjects per group for various levels of alpha and beta (for Chapter 4)

(a) Log reading rate

Minimum difference to be detected as significant 0.1 log wpm Estimated standard deviation of the population 0.23 log wpm

α 0.1 0.05 0.02 0.01 Power β

80 0.2 66 84 107 124 90 0.1 91 112 138 158 95 0.05 115 138 167 89 Minimum difference to be detected as significant 0.15 log wpm Estimated standard deviation of the population 0.23 log wpm

α 0.1 0.05 0.02 0.01 Power β

80 0.2 30 37 48 55 90 0.1 41 50 62 71 95 0.05 51 62 75 84

(b) Reading rate

Minimum difference wish to detect as significant 20 wpm Estimated standard deviation of the population 45 wpm

α 0.1 0.05 0.02 0.01 Power β

80 0.2 63 80 102 119 90 0.1 87 107 132 151 95 0.05 110 132 160 181

Minimum difference wish to detect as significant 30 wpm Estimated standard deviation of the population 45 wpm

α 0.1 0.05 0.02 0.01 Power β

80 0.2 28 36 46 53 90 0.1 39 48 59 67 95 0.05 49 59 72 81

Due to difficulties in recruitment, a sample size of only 43 subjects with

approximately 10 subjects in each group (22 for the combined practice group) were

recruited for the study in Chapter 4. Because of the insufficient number of

participants, a number of the results from Chapter 4 indicated a trend but failed to

reach statistical significance. With a larger subject sample size, these results might

become statistically significant.

Appendix 3

339

Appendix 3 Calculation of visual field loss

Figure A3.1 Example of calculation of visual field loss for subject 40

The solid angle subtended by the scotoma (the visual field loss) was calculated by

entering the following data point into the program “Solangle” (Weleber and Tobler,

1986; Lovie-Kitchin et al., 1990). The total solid angle of this scotoma subtended by

subject 40 was –0.053 (in ster-radians) while the visual field loss covered 0.42% of a

sphere.

Radius Meridian Radius Meridian Radius Meridian

1 0 8 270 10.5 195 7.5 345 9 255 10.2 180 8.5 330 10 240 9 165 7.5 315 10.5 225 5 165 7.5 300 11.5 218 1 180 7.5 285 11 210

Scotoma

Appendix 4

340

Appendix 4 Measurement of optical parameters of stand

magnifiers

As a fixed focus stand magnifier (STM) is a type of magnifier where a positive lens

is mounted at a fixed position on a stand, which is designed to rest on the paper

during reading (Figure A4.1). The position of the lens system is such that the object

plane is normally placed within the focal length of the lens. As a consequence, the

image is not located at infinity but at some other image distance (l’) and the image

vergence (L’) is divergent. Therefore the magnifier user needs to have sufficient

accommodative amplitude, or an appropriate reading correction, to focus the image.

In addition, because the light emerging from the STM is diverging, the effective

power and magnification effects of the lens varies as the eye-to-lens distance is

changed. In general, as the eye-to-lens distance decreases, the effective power

increases and magnification achieved increases. Magnifying power may be expressed

in terms of equivalent viewing distance (EVD) (Bailey, 1984 (a); Bullimore and

Bailey, 1989). The EVD is the distance at which the object of regard subtends an

angle equals to the angle that the virtual enlarged image (formed by the magnifying

lens) subtends at the observer’s eye. Definition of equivalent viewing power (EVP) is

an inverse of EVD (Figure A4.1). Therefore measuring the optical parameters of a

STM – the eye-to-image distance and the EVD for different eye-to-lens distances -

are important to ensure that a STM with sufficient magnification and position for a

clear image is prescribed to the patient and that the patient has an appropriate reading

correction for the eye-to-lens distance used.

Procedures

An optical bench was set up to measure two optical parameters for each of the STMs

used in this research: image distance (l') and equivalent power (Fe). An average of

five values were taken for each parameter.

1. Vergence measurement of a stand magnifier

Two methods were used in this study to measure the image distance for each STM.

The two methods were compared.

Appendix 4

341

l = object distance l’ = image distance L = object vergence L’ = image vergence z = eye-lens distance -l’+ z = eye-image distance Fe = equivalent power of stand magnifier

zl'-1 eyeat vergenceL2+

==

Figure A4.1 Ray diagram of a stand magnifier

Object (text on page)

Stand magnifier (STM)

Image

Eye

z

l f l’

L‘

L Fe

Appendix 4

342

Conjugate object-image method

By the use of a focusable telescope (with an additional near cap for very short

working distance), a beam splitter, an object (fine grid lines) and a measurement

scale, the image distance of each STM was measured using the conjugate object-

image method (Bailey, 1981 (c); Chung and Johnston, 1989) (Figure A4.2).

An object (fine grid lines) was located at the base of the STM. The magnified image

from the STM was viewed through a beam splitter, which was placed on the lens

surface of the STM on an optical bench. The image was then focused by a 8x

focusable telescope. With the use of the beam splitter, both the magnified image and

a scale on the other side of the optical bench were simultaneously visible to the

observer (Figure A4.2). The image distance generated by the STM was equal to the

distance between the beam splitter and the scale on the optical bench. As such, the

image distance (l') of the STM was determined by adjusting the distance of the scale

until a sharp image was formed. Image vergence (L’) was the reciprocal of the image

distance.

Figure A4.2 Conjugate object-image method

Beam splitter

Object (Fine grid lines)

Eye

Telescope

Scale Flash light

Image

l'

l' = image distance

Optical bench

Appendix 4

343

Real image method

The principle of the real image method is similar to that of conjugate object-image

method but it uses a converging lens rather than a telescope. The purpose of using a

converging lens was to re-converge the diverging light that emerged from the STM

such that the image could be projected onto a screen (Figure A4.3a). Once a sharp

real image was formed on the screen, the STM was removed. Without changing the

location of the converging lens or the screen, the object (the paper with fine grid

lines) was moved backwards until a sharp image was again formed on the screen

(Figure A4.3b). The distance between the original location of the STM and the new

position of the object plane was equal to the image distance (l’) of the STM. Image

vergence (L’) was the reciprocal of the image distance.

a.

b.

Figure A4.3 Real image method

Eye Screen Condensing lens (+10D)

Object (Fine grid lines)

Object grid

l' = image distance

Optical bench

Adjust the condensing lens and the screen until a clear image was formed

Remove STM and move the grid (object plane) until a sharp image of the grid was again focussed on the screen

Eye Screen Condensing lens (+10D)

Stand magnifier

Appendix 4

344

2. Measurement of equivalent power of STM

In measuring the equivalent power of the STM, two methods were used and

compared. One of the methods used distant objects while the other used near objects.

Distant object method

Two light sources, which were 1m apart, were placed 5 m away from the STM to be

measured. As the object distance was reasonably long, the incident rays directed to

the magnifier were close to parallel. This light, which was emitted by the light

sources (object) was focused by the positive lens of the STM to form a minified

(real) inverted image in a plane behind the STM (Figure A4.4). The two image spots

(light spots) were then marked and the image size (or distance between the two light

spots) was measured by a travelling telescope to within 0.1 mm. By comparing the

ratio of the image size and object size, image distance could be calculated.

distanceObject size)(Object size) (Imagedistance Image ×=

As the object was at a far distance, the image distance was assumed to be the

equivalent focal length of the STM. The reciprocal of the image distance was the

equivalent power (Fe) of the magnifying lens.

h = object height h’ = image height l = object distance l’ = image distance fm = focal length of the stand magnifier Figure A4.4 Determination of the dioptric power of a stand magnifier

h'

h=1m

l=5m l'≈fe

Appendix 4

345

Two conjugate method

This method relies on the measurement of the transverse magnification (M) of the

optical system for two positions of conjugate planes and image planes (Smith and

Atchison, 1997). The change of magnification and object position between the two

sets of conjugate planes are relative measures, which do not require measurements to

be made with respect to the principal planes of the lens.

Similar to the distant-object method described previously, two light sources 10 cm

apart were used and placed at 50 cm from the lens surface of the STM. The image

height was measured and the magnification (M1) was calculated. The procedure was

repeated by placing the two light sources further away at 100 cm from the STM and

the magnification (M2) was calculated.

1height Object 1height ImageM1 = and

2height Object 2height ImageM2 =

By comparing magnification calculated at the two object distances, the equivalent

power of the STM (Fe) was determined using the following equation (Smith and

Atchison, 1997).

2) distance objective - 1 distance(Object )M-(MF

-12

-11

e =

3. Calculation of EVD for different eye-to-lens distances

Given the image distance (l’) and the equivalent power (Fe) of the magnifier, the

equivalent viewing distance (EVD), equivalent viewing power (F*) and the eye-to-

image distance (Figure A4.4) were calculated (Smith and Atchison, 1997). This was

done for eye-lens distances (z) of 2.5, 5, 10, 15, 20 and 25 cm.

)zL' - (1)L' - (FF* e

=

*F1EVD =

Eye-to-image distance = -l' + z

Appendix 4

346

Results

The measured optical parameters – vergence distance (l') and equivalent power (Fe) -

for the Eschenbach illuminated STM used in this study are summarised in Table

A4.1. In general, there was no significant difference between the conjugate object-

image method and the true image method for measuring the image distance (paired t-

test, t=2.19, df=9, p=0.06). However, there was a tendency towards a greater

difference in the image distance measured by these two methods for high-powered

STMs. As the magnification of the STM increased, the image distance increased. The

virtual image produced by the lens system with the telescope (conjugate object-

image method) increased the difficulty for the observers to focus the image since it

was located further away. For this reason, the difference in the image distance

between the two methods increased as the magnification increased. Between the two

methods, the image formed with the real image method was real and erect, which

made it easier for the observer to determine the position of a sharp image. Unlike the

conjugate object-image method, the observer's accommodation is not a factor

affecting the judgment, as no focusing procedure was required. Therefore, the errors

induced by a change in accommodation were minimal. For these reasons, image

distance measured by the real image method was used in the calculation of EVD for

each eye-to-lens distance (Table A4.2).

For the equivalent power, the Bland and Altman analysis showed that the mean

difference of Fe determined by the distant object method and the two-conjugate

planes method was 0.05 D with limits of agreement between 0.77 and –0.88 D. The

mean difference was small and the limits of agreement in Fe measured by the two

methods covered a small range which would be acceptable for clinical or research

measures. In addition, the difference in Fe between methods was not significantly

correlated with the mean Fe (r=-0.25, p=0.48). As such, it is reasonable to suggest

that the measurement of the equivalent power by either distant object method or two-

conjugate planes method is interchangeable. In this appendix, the results from the

two conjugate planes method was used to calculate the EVD of the STM for each

eye-to-lens distance (Table A4.2). Comparatively, the measured values of Fe were

slightly different from the dioptric power quoted by the manufacturer. Except for the

strongest magnified STM (model 15577), the dioptric power of the other STMs

Appendix 4

347

between the measured values of Fe and those quoted (or labelled) by the

manufacturer were close.

The optical parameters (calculated from the equivalent power and image distance

measurements for each eye-to-lens distance (z)) for each STM used in Chapter 4 are

given in Table A4.2. From this information, the near spectacles prescribed for

subjects to achieve a clear images and the effective magnification power (F*) at a

particular viewing distances from the lens could be determined.

Table A4.1 Results of the optical parameter measurements for the stand magnifiers used in this study

Descriptions Vergence distance (l’) in centimetres

Equivalent Power (Fe) in dioptre

Model Labelled dioptric

power (D)

Labelled magnification

(X)

Method 1 (Conjugate

object image method)

Method 2 (True image)

Method 1 (Distant objects)

Method 2 (Two-

Conjugate)

Biconvex lenses 15878 5 2.2 8.4 10.00 5.38 5.71

Biconvex lenses 15788 10 2.5 9.46 8.50 9.33 10.14

105 x 80 mm 15849 7 2.8 10.66 11.56 6.32 5.85 Aspheric

60mm round 15589 12 3 13.9 13.96 11.29 11.18

Aspheric 70mm round 15549 16 4 19.86 19.90 14.75 14.70





Aspheric 35mm round 15577 50 12.5 39.52 48.72 46.55 46.04

Appendix 4

348

Table A4.2 Optical parameters of the stand magnifiers used in this study, calculated from the measured values for equivalent power (Fe) and image distance (l’)

Descriptions z=2.5 (cm) z=5 (cm) z=10 (cm) z=15 (cm) z=20 (cm) z=25 (cm)

Dioptric Power

Fe (D)

-l' (cm)

F* (D)

EVD (cm)

ey/im(cm)

F* (D)

EVD(cm)

ey/im(cm)

F* (D)

EVD(cm)

ey/im(cm)

F* (D)

EVD(cm)

ey/im(cm)

F* (D)

EVD(cm)

ey/im(cm)

F* (D)

EVD(cm)

ey/im(cm)

Biconvex lenses 5 5.71 10 12.57 7.96 12.5 10.48 9.55 15 7.86 12.73 20 6.29 15.91 25 5.24 19.09 30 4.49 22.27 35

Biconvex lenses 10 10.14 8.5 16.93 5.91 11 13.79 7.25 13.5 10.07 9.94 18.5 7.92 12.62 23.5 6.53 15.31 28.5 5.56 17.99 33.5

105 x 80 mm 7 5.85 11.56 11.92 8.39 14.06 10.12 9.88 16.56 7.78 12.86 21.56 6.31 15.84 26.56 5.31 18.82 31.56 4.59 21.81 36.56

Aspheric 60mm round 12 11.18 13.96 15.55 6.43 16.46 13.5 7.41 18.96 10.68 9.36 23.96 8.84 11.31 28.96 7.54 13.27 33.96 6.57 15.22 38.96







Fe = Equivalent power of stand magnifier (D) F* = Effective power according to different Z values EVD = equivalent viewing distance (cm) l’ = image distance(cm) ey/im = eye-to-image distance (= -l’+z)

Appendix 5

349

Appendix 5 Questionnaires used in Chapters 4 and 5

Questionnaires for participants in Chapter 4

Before the prescription of stand magnifier

1. Frequency of reading

1. > 1/day for >10 mins 2. Regularly (1/day for >10 mins)

3. Sometimes (once per week) 4. Rarely (< once every 2 weeks)

5. Never

2. Duration of reading at each time

>1 hour 10 mins-1 hour <10 mins sometimes rarely

3 (a) What sort of reading materials do you read?

Mail/ letters Newspapers Magazines Leaflets Books

None Others: ________________________

3 (b) What sort of reading materials would you like to read?


None Others: ________________________

After the prescription of stand magnifier

1. Did you find the stand magnifier helped you in reading?

Excellent Good Average Little Useless


1. > 1/day for >10 mins 2. Regularly (1/day for >10 mins)

3. Sometimes (once per week) 4. Rarely (< once every 2 weeks)

5. Never

3.(a) Duration of reading with magnifiers at each time


Appendix 5

350

3.(b) Duration of reading without magnifiers at each time


4(a) What sort of reading materials do you read with magnifier?


None Others: ________________________

4(b) What sort of reading materials do you read without magnifier?


None Others: ________________________

Questionnaires for participants in Chapter 5

Usage of stand magnifiers

1. How often do you use your stand magnifier?

1. Frequently (> 1/day) 2. Regularly (2-3 days)

3. Sometimes (1/week) 4. Rarely (1/ 2 weeks) 5. Never

2. How long do you read at any one time with your stand magnifier?

>30 mins 10 -30 mins <10 mins sometimes rarely

3. What sort of reading materials do you read with your magnifier?

Mails/ letters Newspapers Magazines Leaflets Books

None Others: ________________________

4. When reading, how much difficulty do you have in moving your magnifier

along the line?

1. No difficulty 2. Slightly difficult 3. Moderate difficult

4. Significantly difficult 5. Extremely difficult

5. When reading, how much difficulty do you have in moving your magnifier

back to find the start of the next line?

1. No difficulty 2. Slightly difficult 3. Moderate difficult

4. Significantly difficult 5. Extremely difficult

Appendix 5

351

6. When reading, how often do you loose your place or miss a line?

1. Very frequently (many times in one reading)

2. Frequently (a few times in one reading)

3. Sometimes 4. Rarely 5. Never

7. Do you use a reading stand to read with your magnifier?

Yes / No

8. Do you use piles of books to raise the reading task during reading?

Yes / No

Preference of the reading assistance on stand magnifiers

9. Did you find the line guide useful?

Yes / No

Reason: _________________

10. Did you find the black fixation spot useful?

Yes/ No

Reason: __________________

Other comments: _____________________________________________________

Analysis of the questionnaires

As the scales used in the questionnaires to classify the subjective responses were all

ordinal variables rather than continuous variables, repeated measures on the

responses of the frequency and duration of reading and usefulness of the stand

magnifier for subjects participating in Chapter 4 could not be analysed. For this

reason, the scales of these variables were recoded into continuous variables by time-

weighted factors of the responses. This recoding system was determined based on an

assumption that the scale approximately reflected the actual meaning of the scales.

Appendix 5

352


Original variable New variable

Scale Description Times of reading per month 1 > 1 day 40 2 1 day 30

3 Sometimes (once per week) 12

4 Rarely (< once times per 2 weeks) 8

5 Never 1

2. Duration of reading with and without STM

Original variable New variable

Scale Description Times of reading at each time in terms of minutes

1 > 1 hour 60 2 10 minutes to 1 hour 40 3 < 10 minutes 8 4 Sometimes 2 5 Rarely 1

3. Usefulness of STM

Original variable New variable Scale Description Scores of the usefulness of

the STM over 100 1 Excellent 90 2 Good 70 3 Average 50 4 Little 30 5 Useless 0

Appendix 6

353

Appendix 6 Calibration of 3SPACE Isotrak system (Chapter

5)

In order to assess the accuracy, linearity and sensitivity of the Isotrak system in

measuring the movement of the magnifiers, calibration of the system was conducted

before experimental data were collected. The Isotrak system had the capability to

measure magnifier movement in 3 dimensions (x, y and z); however, subjects were

instructed to put their stand magnifiers on the reading stand for all experimental

measures. Consequently, the movement in the z-direction was assumed to be

constant and was not measured. Calibration of the instrument was therefore only

assessed for movement in the x and y planes.

Procedures

For the calibration data collection, the Isotrak was set up in the same configuration as

that used for the experimental data collection. Linearity and sensitivity were assessed

across the total extent of the area over which subjects typically moved their

magnifiers during the experiment. The reading stand was larger than an A4 paper

sheet with a dimension of 60 x 44.5 cm and the reading passages were always placed

in the middle of the reading stand. The area typically used by subjects was within the

following boundaries: 15 cm in from left and right margins of the reading stand, 5

cm below the top and 10 cm above the bottom of the stand (Figure A6.1). The source

was securely attached to the reading stand and the sensor was held by the

experimenter and moved across a piece of A4 paper (in horizontal and vertical

dimension to calibrate for both planes). The sensor was moved both in step intervals

using marks printed at regular intervals and in smooth continuous movements across

complete lines to simulate the actual forward and retrace movements made when

reading. Movements of the position of the sensor were analysed using Excel and

Matlab programs.

Appendix 6

354

Figure A6.1 Area of reading stand across which calibration measurements were performed

1. Measurement of movement by step intervals

Assessment of the Isotrak by movement in step intervals was conducted in the x-

plane across three horizontal lines, starting at 15 cm in from the left margin of the

board and extending for a total width of 28 cm. The vertical positions for each of the

lines were at 5, 17 and 30 cm from the top edge of the reading board. At each

location where the paper was attached, the sensor was moved along a horizontal

(dashed) line with markings at 2 cm intervals. To assess the sensitivity of the system,

these measures were repeated with the marks at 1, 0.5, 0.2 and 0.1 cm steps for the

horizontal line across the centre of the reading board (i.e. vertical position 17 cm

below top of board). Figure A6.2 is a schematic diagram of the set up of the paper

and location of the calibration along x-plane at 17 cm below the top edge, and Figure

A6.3 shows an example of the output of the recorded data in Excel.

Calibration measures were performed across the area used by subjects when reading.

Dimension of the reading stand (60 x 44.5 cm)

Appendix 6

355

Figure A6.2 Layout of the page to measure the sensitivity and linearity of the

Isotrak in x-plane by step interval movements (e.g. 2 cm), at 17 cm below top of reading stand

For the step interval assessment in the y-plane, the sensor was moved along 3 lines in

the vertical direction, starting at 5 cm below the upper edge of the board. The 3

vertical lines started at x-positions of 15 cm, 27 cm and 40 cm in from the left

margin of the board. At each location where the paper was attached, the sensor was

moved along a vertical (dashed) line with markings at 2 cm intervals. To assess the

sensitivity of the system, these measures were repeated with the marks at 1.0, 0.5, 0.2

and 0.1 cm steps for the vertical line along the centre of the reading board (i.e.

horizontal position 27 cm in from left margin of board; Figures A6.4 and A6.5).


Dimension of an A4 paper (28 cm)

15 cm

17 cm2 cm

The location where the sensor was placed and then jumped to the next interval along the horizontal line

Appendix 6

356

Figure A6.3 An example of a plot of sensor position along x-plane for interval

step movements of 2 cm. This plot demonstrates good linearity across the full horizontal extent assessed. All measurements of sensor and magnifier movements were made with respect to the source, which was placed on the right hand side of the reading stand, towards the upper corner. The x-distance values decreased in numerical value (e.g. from 41 cm to 11 cm on this figure) as the sensor was moved from left to right across the board towards the source and the y-distance values decreased as the sensor was moved from the top to bottom of the stand (Figure A6.5) away from the source.

11

16

21

26

31

36

41

Time (seconds)

Horiz

onta

l pos

ition

of t

he s

enso

r (cm

)

0 10 20 30 40 50 60

LEFT

RIGHT

Appendix 6

357

Figure A6.4 Layout of the page to measure the sensitivity and linearity of the Isotrak in y-plane by step interval movements (e.g. 2 cm), at 27 cm in from the right side of the reading stand

Figure A6.5 An example of a plot of sensor position in the y-plane for interval step movements of 2 cm. Plot demonstrates good linearity across the full vertical extent assessed


27 cm

2 cm

The location where the sensor was placed and then jumped to the next interval along the vertical line

5 cm

-25

-20

-15

-10

-5

0

5

10

Time (seconds)

Vert

ical

pos

ition

of t

he s

enso

r (cm

)

0 10 20 30 40 50 60 70

Up

Down

Appendix 6

358

2. Measurement of movement across continuous lines

The magnifier movements recorded in this study were continuous movements.

Therefore, it was essential to assess the accuracy of the instrument when it generated

continuous movement in 2 dimensions. Apart from using step intervals, the

calibration was also conducted on a page in “portrait orientation” with 8 lines

printed. The length for each line was 14.725 cm (representative of maximum

horizontal distance a magnifier was moved in the experiment) whilst the spacing

between lines was 3.0 cm. The sheet of paper was placed at two vertical positions on

the stand, with the top of the sheet aligned to the upper edge and then with the

bottom of the sheet aligned with the lower edge of the reading stand. The sensor was

placed at the first point of a horizontal line for roughly 5 seconds before it was

moved along a horizontal line. When it reached the last point of the line, the sensor

was held steady for another 5 seconds before it was retraced to the next line. The

reason for the pauses at the beginning and the end of the line was to clearly identify

the location where the line started and finished before the sensor was moved on to

the next line. Two trials were taken for each measurement condition and a mean

value was used to compare to the exact dimension of the lines.

By measuring the continuous movement of the sensor along xy-planes, not only the

total distance for horizontal and vertical movement could be measured but also the

change of the vertical orientation of the sensor during the horizontal movement

(along a line) was analysed (Figure A6.6). Horizontal distance per line was

calculated as the difference in the x-coordinate at the starting point and the x-

coordinate at ending point of a line. By comparing the difference in y-coordinate at

the starting points of two consecutive lines, the vertical distance between lines could

be computed. Difference in y position between start and end point of a line was also

determined.

Appendix 6

359

Figure A6.6 An example of an x-y plot of sensor position for continuous

movements across a page

Results

1. Accuracy of movement in step intervals for x and y-planes

Table A6.1 shows the mean distances moved by the sensor between each interval

marker for each interval step size in the x- and y-planes. The measured movement

distances were very close to the actual distances between interval markers for all step

sizes at all positions in both planes.

-20

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

2

4

6

151719212325272931

Horizontal position of the sensor (cm)

Vert

ical

pos

ition

of t

he s

enso

r (cm

)

Appendix 6

360

Table A6.1 The mean distance (cm) moved by the sensor between interval markers along x-and y-planes

Location of the line Measurement along the x-plane

X-coordinate

Y-coordinate 2 cm step 1 cm step 0.5 cm step 0.2 cm step 0.1 cm step

5 cm down 2.14 ± 0.11

17 cm down 2.08 ± 0.17 1.08 ± 0.1 0.53 ± 0.07 0.19 ± 0.04 0.10 ± 0.03

15 cm in from the top left corner 30 cm

down 2.10 ± 0.15

Percentage errora 7% 8% 6% 5% 0%

Location of the line Measurement along the y-plane

X-coordinate

Y-coordinate 2 cm step 1 cm step 0.5 cm step 0.2 cm step 0.1 cm step

15 cm in 2.02 ± 0.10

27 cm in 2.01 ± 0.08 0.97 ± 0.08 0.51 ± 0.05 0.20 ± 0.05 0.11 ± 0.03

40 cm in

5 cm below the

upper border

2.06 ± 0.11

Percentage error 3% 3% 2% 0% 10%

2. Linearity in a single dimension

Linearity plots show how accuracy varied across the range of distance that the

magnifier was moved. Figure A6.7 shows the linearity plot of measured position

against actual position of the sensor in the X and Y-plane separately. The plots

demonstrate excellent linearity.

a Percentage error was the largest difference made for that particular measure (i.e. 2 cm step intervals) compared with the actual value calculated in terms of percentage.

Appendix 6

361

y = 0.9978x + 8.0276R2 = 0.9999

y = 1.0296x - 0.3204R2 = 0.9997

0

5

10

15

20

25

30

35

40

45

0 10 20 30 40

Position of interval marker (cm)

Mea

sure

d Is

otra

k po

sitio

n (c

m)

2 cm intervals

1 cm intervals

Figure A6.7a Isotrak position as a function of marker position (step intervals)

on the paper – X plane movement

y = 1.014x + 0.4309R2 = 0.9999

y = 1.005x - 2.3275R2 = 1

-25

-20

-15

-10

-5

0

5

10

-25 -15 -5 5 15

Position of the intervals' marker (cm)

Mes

aure

d Is

otra

k po

sitio

n (c

m)

2 cm intervals

1 cm intervals

Figure A6.7b Isotrak position as a function of marker position (step intervals)

on the paper – Y plane movement

Appendix 6

362

3. Sensitivity of the Isotrak

According to the manual (Polhemus 3SPACE Isotrak User's Manual, 1987) the

smallest resolution of the sensor is 0.22 cm; however it was possible to consistently

distinguish step interval movements of 0.1 cm in X- and Y-planes (Table A6.1) in

the experimental set up used for the study in Chapter 5. The sensitivity determined

from the calibration recordings was better than that suggested in the handbook.

4. Accuracy of continuous movement across XY plane

Table A6.2 summarises the results of the continuous movement recordings when the

sheet of paper was placed with the top of the sheet aligned to the upper edge of the

stand (“top” position) and with the bottom of the sheet aligned with the lower edge of

the reading stand (“bottom” position). For each position, the distance moved along

the x-plane (horizontal distance) and the vertical distance between lines is the mean

of the values measured for the 8 lines on the page when the page was in that position.

Again the accuracy is good, with a mean difference between the actual horizontal

distance and the measured horizontal movement distance of 0.1 to 0.2 cm for the

14.7 cm line (about 1.4% error) and a mean difference of 0.05 cm between actual and

measured vertical separation of lines 3 cm apart (about 1.7% error). The difference in

Y-position of the sensor between the start and end of each line averaged about 0.2

cm, with the Y-position of the sensor being about 0.2 cm lower at the end of the line

than the start of the line (Figure A6.6).

Table A6.2 The mean distance (cm) measured along x-and y-planes using continuous lines

Location of the paper

Actual value (cm) Top (cm) Bottom (cm) Percentage

error

Horizontal distance along the line (x) 14.725 14.808 ± 0.16 14.935 ± 0.14 1.4%

Vertical difference between lines (y) 3.0 3.05 ± 0.10 3.044 ± 0.08 1.7%

Difference in Y-coordinate within a

line

Presume to be 0 0.2 ± 0.04 0.19 ± 0.085

Appendix 6

363

Discussion

In comparing the results from this study and the data given in the Isotrak handbook,

the sensitivity and accuracy of the sensor movement recordings reported in this

calibration study were better than that that suggested in the handbook and more than

sufficient for the purposes of recording magnifier movements. The linearity in the X

and Y-planes was also excellent.

The fact that a difference was found in the y-position of the sensor between the start

and end of the line, as the sensor was moved horizontally across the board, merits

further consideration. The magnitude of this inaccuracy was approximately 0.2 cm.

Care was taken to ensure that the page was aligned so that the horizontal lines ran

parallel to the top or bottom of the reading stand and the source was aligned so that

the X-Y planes were in line with those of the reading stand. Therefore this

inaccuracy was not a result of careless alignment of paper, reading stand and/or

source. In the analysis of magnifier movements (see section 5.3.4.3), the strategy of

movement that a subject used in the forward and retrace phases of reading was

categorised. During the categorisation of the magnifier movement, any change in y-

position between the start and end of a line, which was greater than 0.3 cm, was

classified as a non-straight movement. The reason for the selection of 0.3 cm as the

criterion to differentiate a straight or non-straight movement was based both on the

line separation (which was 0.5 cm for the actual experiment) and on the measured

inaccuracy in y position between the start and end of a line found during calibration.

If the average inaccuracy in y position between start and end of a 14.7 cm line was

0.2 cm and the average x distance that subjects moved their magnifiers in the

experiment was 12.6 cm (see results in Chapter 5), the average inaccuracy in y-

position across 12.6 cm would be about 0.17 cm. Therefore the 0.3 cm criterion was

considered to be reasonable as it was roughly twice the expected inaccuracy in y-

position and was approximately half the vertical separation between adjacent lines of

the reading passages used in the experiment (text of 12-point print with single line

spacing). Therefore, any movement which had a change in vertical position along the

line greater than this criterion was classified as a non-straight movement, either an

uphill or a downhill movement.

References

364

References

Bailey, I. L. (1981 (c)). Verifying Near Vision Magnifiers - Part 2. Optom Monthly

72, 34-38.

Bailey, I. L. (1984 (a)). Equivalent viewing power or magnification. Which is

fundamental? Optician 188, 32-35.

Bowers, A. R. (1998 (b)). Reading Performance in Visual Impairment. Glasgow

Caledonian University, Glasgow, United Kingdom

Bowers, A. R., Lovie-Kitchin, J. E., and Woods, R. L. (2001 (b)). Eye movements

and reading with large print and optical magnifiers in macular disease.


Bullimore, M. A., and Bailey, I. L. (1989). Stand magnifiers: an evaluation of new

optical aids from COIL. Optom Vis Sci 66, 766-773.

Cheong, A. M. Y., Lovie-Kitchin, J. E., and Bowers, A. R. (2002). Determining

magnification for reading with low vision. Clin Exp Optom 85, 229-237.

Chung, S. T. L., and Johnston, A. W. (1989). Practical options for magnifications

and field of view of stand magnifiers. Clin Exp Optom 72, 140-147.

De Land, P. N., and Chase, W. W. (1992). Statistics notebook: entry III.G: power of

a test. Optom Vis Sci 69, 164-167.

Lovie-Kitchin, J., Mainstone, J., Robinson, J., and Brown, B. (1990). What areas of

the visual field are important for mobility in low vision patients? Clin Vision

Sci 4, 249-263.

Lovie-Kitchin, J. E., Bowers, A. R., and Woods, R. L. (2000 (a)). Oral and silent

reading performance with macular degeneration. Ophthalmic Physiol Opt

20, 360-370.

References

365

Martin, R., Coco, B., Ruben, C., Isabel, A., Mayo, A., Beatriz, M., and et al. (2002).

Study of the efficacy of a new electronic device for the ARMD: a new

therapeutic option. Invest Ophthalmol Vis Sci 43, E-Abstract 1205.

Polhemus 3SPACE Isotrak User's Manual. (1987). Vermont, OPM3704-001.

Smith, G., and Atchison, D. A. (1997). The Eye And Visual Optical Instruments.

Cambridge, Cambridge University Press.

Sunness, J. S., Applegate, C. A., Haselwood, D., and Rubin, G. S. (1996). Fixation

patterns and reading rates in eyes with central scotomas from advanced

atrophic age-related macular degeneration and Stargardt disease.


Weleber, R. G., and Tobler, W. R. (1986). Computerized quantitative analysis of

kinetic visual fields. Am J Optom Physiol Opto 101, 461-468.

Documents

READING PERFORMANCE WITH STAND MAGNIFIERS IN AGE-RELATED … · Chapter 1 Literature review 1 1.1 Low vision 3 1.2 Age-related macular degeneration 10 1.3 Reading with age-related