Upload
david-edvardsson
View
214
Download
2
Embed Size (px)
Citation preview
The person-centred care of older people with cognitive impairmentin acute care scale (POPAC)
DAVID EDVARDSSON R N , P h D1,2, ANITA NILSSON R N , P h D
3, DEIRDRE FETHERSTONHAUGH R N , P h D4,
RHONDA NAY R N , P h D 5 and SHANE CROWE R N6
1Associate Professor/Director of Nursing Research, School of Nursing and Midwifery, La Trobe University,Melbourne, Victoria, Australia, 2Associate Professor, Department of Nursing, Umea University, Umea, Sweden,3PhD Candidate, Department of Nursing, Umea University, Umea, Sweden, 4Deputy Director/Senior ResearchFellow, Australian Centre for Evidence Based Aged Care (ACEBAC), La Trobe University, Melbourne, Victoria,5Professor and Director, Australian Institute for Primary Care and Ageing (AIPCA), La Trobe University, Melbourne,Victoria and 6Deputy Director, Ambulatory and Nursing Services, Austin Health, Heidelberg, Victoria, Australia
Background
Providing cost-effective and high quality care of older
people has been established as one of the major chal-
lenges of this century due to a rapidly increasing ageing
population, and the consequential long-term burden
imposed on health and aged care systems throughout
the Western world through the ageing societies and the
increasing prevalence of older people with dementia,
which has risen globally from an estimated 29 million in
Correspondence
David Edvardsson
School of Nursing and Midwifery
La Trobe University
Level 4
Austin Tower, PO Box 5555
Heidelberg
Victoria 3084
Australia
E-mail:
E D V A R D S S O N D . , N I L S S O N A . , F E T H E R S T O N H A U G H D . , N A Y R . & C R O W E S . (2012) Journal of
Nursing Management
The person-centred care of older people with cognitive impairment in acute carescale (POPAC)
Aim To construct and evaluate psychometric properties of the person-centred careof older people with cognitive impairment in acute care settings (POPAC) scale.
Background Older people with cognitive impairment are admitted frequently to
acute care, with needs not always met through standard practice. Best practice
models have been suggested, but few assessment scales exist.
Methods Psychometric evaluation using statistical estimates of validity and reli-
ability based on an Australian sample of acute care nursing staff (n = 212).
Results The final 15-item questionnaire consists of three subscales, �using cognitive
assessments and care interventions�, �using evidence and cognitive expertise� and
�individualizing care�. Estimates of validity and reliability were highly satisfactory.
Conclusion The POPAC scale makes a valuable contribution by providing valid and
reliable measures of the extent to which acute nursing staff report using best practice
care processes to identify and consider cognitive impairment and to employ nursing
interventions to meet the needs associated with old age and cognitive impairment.
Implications for nursing management The POPAC scale is short, easy to administer
and not time consuming to complete, but still provides clinically relevant infor-
mation. It can be used as a conceptual fundament in developing best practice
nursing care in the acute clinical setting, as well as for nursing research.
Keywords: dementia, evidence-based practice, hospitals, nurse�s practice patterns,
psychometrics
Accepted for publication: 13 March 2012
Journal of Nursing Management, 2012
DOI: 10.1111/j.1365-2834.2012.01422.xª 2012 Blackwell Publishing Ltd 1
2005 to about 34 million in 2009 (Wimo et al. 2010). It
is well known that the rate of hospital use increases
with age, as does the average number of days per stay in
hospitals. In 2004–05, people older than 65 years rep-
resented 35% of all hospital admissions in Australia,
and accounted for 48% of all patient days (Australian
Institute of Health and Welfare 2006), and similar data
have been reported for the United States (DeFrances &
Hall 2007). In addition, older people often present with
co-morbidities and a high prevalence of dementia dis-
orders or other forms of cognitive impairment that add
risks to the hospitalization. Samson et al. (2009)
showed recently that the prevalence of dementia disor-
ders among older people admitted to acute hospital care
was as high as 42.4%, and they further showed that in
comparison with cognitively unimpaired patients, three
times as many patients with dementia and five times as
many with scores on the mini mental state examination
(MMSE) below 15 died during their hospitalization,
even after controlling for age and severity of acute
physical illness (Samson et al. 2009). People with
dementia have also been shown to have a fivefold risk of
developing delirium in hospitals (Royal College of
Psychiatrists 2005), and delirium is rarely detected by
staff even though frequently occurring in older patients
(Joray et al. 2004, Soderqvist 2007). It has been esti-
mated that as much as 30–40% of incident delirium
may be preventable, for example, through screening and
preventive measures (Inouye 2006). Thus, high quality
care of patients with cognitive impairment and delirium
prevention depends to a large extent on the ability of
staff to identify specific patients at risk, to evaluate
cognition and to intervene against potential risk factors
(Dick 1998, Zeleznik 2001, Waszynski 2007).
It has further been described that best practice models
of acute care for older people with cognitive impair-
ment consists of using gerontological expertise, targeted
assessment techniques, discharge planning and en-
hanced interdisciplinary communication (Hickman
et al. 2007). In addition, the development of staff
geriatric expertise, using standardized care protocols,
evidence-based assessments and environmental modifi-
cations have also been identified as central to the
development of best practice models (Moyle et al.
2008). Others have argued that older patients with
cognitive impairment would benefit if acute hospital
wards would increasingly apply person-centred princi-
ples such as a holistic philosophy of care; use interdis-
ciplinary assessments and individualized interventions:
establish and maintain relationships; adapt the envi-
ronment and involve family members (Edvardsson &
Nay 2010). Literature reviews indicate a shortness of
psychometrically sound tools for measuring the extent
to which best practice nursing care of older people with
cognitive impairment is delivered in acute care settings.
This makes it difficult therefore to study the association
between person-centred care practices and health out-
comes in hospitalized older people with cognitive
impairment. Thus, the aim of this study was to con-
struct and evaluate psychometric properties of the per-
son-centred care of older people with cognitive
impairment in acute care settings (POPAC) scale, a self-
report staff questionnaire measuring the extent to which
acute nursing staff report using best practice care pro-
cesses, to identify and consider cognitive impairment
and to employ nursing interventions to meet the needs
associated with old age and cognitive impairment.
Methods
The study utilized methods for scale construction
(Streiner 2008) and a cross-sectional psychometric de-
sign (Nunnally 1994) to evaluate the validity and reli-
ability of a measurement scale for assessing the extent
to which staff report providing person-centred best
practice care of older people with cognitive impairment.
Scale construction
The theoretical foundation of the scale was developed
from the following contemporary nursing research lit-
erature into best practice nursing care of older people
with cognitive impairment in acute hospital settings
(Hickman et al. 2007, Day et al. 2008, Moyle et al.
2008, Edvardsson & Nay 2010). From these articles,
eight dimensions of best practice were identified, and
further used as a base for the development of the tool:
recognizing cognitive impairment, consulting specialist
expertise, using evidence-based care protocols or
guidelines, making environmental adjustments, provid-
ing social enrichments, prioritizing staff continuity and
close interactions, avoiding restraints, and individual-
izing care. These theoretical dimensions were translated
into items to form the preliminary version of the scale.
All items were designed as statements about different
best practice procedures, and participants were asked to
respond how often they estimated these procedures
being used on their ward. The conceptual approach to
this linguistic formation of items was that the scale as a
whole would indicate a set of optimal processes in
providing best practice care of older people with cog-
nitive impairment in acute care settings. Thus, the
scores on the scale indicate the extent to which staff
estimated that such processes generally were happening
D. Edvardsson et al.
ª 2012 Blackwell Publishing Ltd2 Journal of Nursing Management
on the ward, with higher scores representing that such
best practice processes were perceived as occurring
more often. The first draft preliminary tool consisted of
21 items relating to best practice and eight items relat-
ing to staff attitudes towards older people with cogni-
tive impairments. The scaling utilized a six-point Likert-
type scale with the following response options rated as
1–6: never, very rarely, rarely, frequently, very fre-
quently and always.
An international multidisciplinary expert panel
(n = 12) consisting of internationally recognized
researchers within medicine, nursing and sociology with
a strong publication record relating to older people with
cognitive impairment in acute care settings was invited
to comment on the preliminary tool. The expert panel
members were invited to comment on the relevance and
specificity of items in recognizing procedures that
facilitate valid recognition of cognitive impairment, and
in describing best practice care that could meet the
needs associated with cognitive impairment. Responses
were received from two-thirds of the panel (n = 8), and
these comments confirmed the theoretical relevance of
the items, and suggested revisions to increase the clarity
of the aim and instructions to participants; personalize
the phrasing of statements; and enhance the clinical
applicability of the scale. The preliminary scale was
revised according to comments from the panel, which
resulted in a final preliminary pool of 21 items for
further testing.
Sampling
The 21-item preliminary scale was distributed to an
eligible sample of 360 nursing staff at 12 wards in a
metropolitan acute hospital in Melbourne, Australia.
The wards had the following focus: respiratory care (1),
colorectal/urology (1), neurosurgery (1), cardiology (1),
oncology (2), liver/transplants (1), renal (1), orthopae-
dics (1), general medicine (2), neurology (1). All data
were collected between June and August 2011, and in
July 2011, 1345 patients were cared for at these wards,
58% were 65 years or older, and the mean length of
stay for all patients across these wards was 6.6 days.
Ward participation was based on self-reported frequent
admission of older patients (>65 years). Participating
staff had the option of completing the survey online or
paper based. A 59% response rate was achieved, pro-
viding a final total sample of 212 nursing staff. In
addition, a sample of 25 nursing staff from the ortho-
paedic ward participated in test–retest evaluation,
which involved completion of the final version scale at
two time points within a 2-week interval.
Psychometric evaluation
Item performance was evaluated through assessments of
item means and standard deviations, item-total corre-
lations, Cronbach�s alpha if item deleted, and indepen-
dent samples t-test. The cut-off for acceptable
homogeneity contribution on item level was set to item-
total correlations of >0.3 and <0.7, and that the total
scale Cronbach�s alpha would not increase by deleting
any single item. The discriminative ability of items was
investigated by comparing item mean scores for
respondents in the top and bottom quartile of the scale
total scores, and a value of P < 0.05 was considered
appropriate. The cut-off for using explorative factor
analyses to assess construct validity was set to a sig-
nificant Bartlett�s test of sphericity, and a Kaiser-Meyer-
Olkin (KMO) measure of sampling adequacy of >0.7.
Construct validity was then assessed using principal
component analysis with oblimin rotation as factors
were expected to be correlated. Assessment of content
validity was based on literature and the expert panel
review process. Internal consistency reliability was
evaluated using Cronbach�s alpha on total scale and
subscale levels, and the cut-off for acceptable internal
consistency was set to >0.7. Test–retest reliability was
assessed using the paired samples t-test and intraclass
correlation coefficient, and the cut-off for acceptable
test–retest reliability was set to a value of P > 0.05 be-
tween test and retest scores, and/or a correlation coef-
ficient of >0.3. Missing values (<10%) in surveys were
replaced with the item mean so as not to distort the
analyses (Shrive et al. 2006). Item five was negatively
worded and was reversed prior to all calculations.
Higher scores on item, subscale and total scores indicate
a higher level of person-centredness, and all aggregated
scores were calculated using sum scores divided by
number of items. The PASW statistics version 18 (SPSS
Inc., Chicago, IL, USA) was used to analyse all data.
Results
The sample consisted mostly of females (93%), with a
mean age of 35 years ± 11.5, with an average of
10 years ± 9.2 work experience in acute care, and
5 years ± 5.0 on the ward in question. The majority of
participants (93%) had no postgraduate qualifications
in the care of older people, and most participants (85%)
acknowledged that older people with cognitive impair-
ments were frequently or very frequently admitted to
their wards. The mean POPAC total score was
3.77 ± 0.65 (range 1.73–5.80), with a skewness of
)0.12.
Construction and psychometric evaluation of the POPAC
ª 2012 Blackwell Publishing LtdJournal of Nursing Management 3
Item performance
Six items from the preliminary 21-item pool were de-
leted as they failed to meet the cut-off for acceptable
homogeneity (>0.3). As reported in Table 1, the
remaining 15 items showed satisfactory homogeneity
estimates as evidenced by item-total correlations
ranging between 0.40 and 0.67, and by the fact that
the total scale Cronbach�s alpha would not have been
increased by deleting any of these 15 items. All items
showed a discriminative ability through generating
significant mean score differences, with higher item
mean scores in the top quartile group as expected
(P < 0.01).
Validity
The content validity of the POPAC was deemed satis-
factory as the conceptual content of the scale covered
key dimensions of best practice nursing care of older
people with cognitive impairment as described in con-
temporary nursing research journal publications
(Hickman et al. 2007, Day et al. 2008, Moyle et al.
2008, Edvardsson & Nay 2010). In addition, the expert
panel review comments supported the content validity
of the tool. The appropriateness of using explorative
factor analyses to assess construct validity was sup-
ported by a KMO of 0.88 and a significant Bartlett�s test
of sphericity. As reported in Table 2, the explorative
factor analysis resulted in a stable three factor model
with satisfactory explanatory power, and the structure
of the model followed an underlying theoretical logic.
Based on the factor structure, it was decided that the
final 15-item questionnaire consists of three subscales
that were given the following interpretative labels based
on their conceptual content: �using cognitive assess-
ments and care interventions�, �using evidence and
cognitive expertise�, and �individualizing care�.
Table 1Item performance of the final POPAC scale
Item Mean SDCorrected Item-total
correlationCronbach�s Alpha
if item deleted
1. We assess the cognitive status of our older patients onadmission
4.34 1.34 0.50 0.87
2. We make environmental adjustments to avoid over-stimulationin older people with cognitive impairment (e.g. single rooms,noise reductions etc.)
3.87 1.06 0.67 0.86
3. We diagnose symptoms of cognitive impairment (e.g.dementias, delirium etc.)
3.91 1.04 0.53 0.86
4. We spend more time with older patients with cognitiveimpairments compared with cognitively intact patients
3.99 1.11 0.48 0.87
5. We leave older people with cognitive impairments alone in theward
4.52 1.12 0.40 0.87
6. We use evidence-based tools to assess cognitive status ofolder patients (e.g. the MMSE, SPMSQ, CAM etc.)
3.03 1.40 0.48 0.87
7. We consult specialist expertise (e.g. psychologist,gerontologist) if we find that a patient has cognitiveimpairment
3.83 1.20 0.62 0.86
8. We use evidence-based care guidelines in the care of oldercognitively impaired patients
3.69 1.11 0.60 0.86
9. We use biographical information about older patients� (e.g.habits, interests and wishes etc.) to plan their care
3.31 1.01 0.48 0.87
10. We involve family members in the care of older patients withcognitive impairment
4.26 0.95 0.52 0.86
11. We provide staff continuity for older patients with cognitiveimpairments (e.g. the same nurses providing care to thesepatients as often as possible)
3.45 0.96 0.42 0.87
12. We systematically evaluate whether or not older patientswith cognitive impairment receive care that meets their needs
3.69 0.99 0.60 0.86
13. We involve older patients with cognitive impairment in decisionsabout their care (e.g. examinations, treatments etc.)
3.65 0.93 0.54 0.86
14. We ensure that older patients with cognitive impairment havetests/examinations/consultations in the unit rather than havingto go to another department
3.26 0.98 0.49 0.87
15. We discuss ways to meet the complex care needs of peoplewith cognitive impairment
3.80 1.09 0.53 0.86
D. Edvardsson et al.
ª 2012 Blackwell Publishing Ltd4 Journal of Nursing Management
Reliability
Internal consistency reliability was high for the 15-item
POPAC scale, as evidenced through a total Cronbach�salpha of 0.87. The subscales also had satisfactory reli-
ability, with the subscales showing the following
Cronbach�s alpha values: subscale one �using cognitive
assessments and care interventions� scoring 0.74, sub-
scale two �using evidence and cognitive expertise� scor-
ing 0.79, and subscale three �individualizing care�scoring 0.78. Furthermore, the assumption that all
items reliably measure a single underlying construct was
supported by the item-total correlations ranging be-
tween 0.40 and 0.67 as shown in Table 1.
Regarding test–retest reliability, the paired sample
t-test showed that there were no significant mean dif-
ferences between test and retest scores for the total scale
or the subscales, thus indicating a satisfactory temporal
stability as reported in Table 3. In addition, the intra-
class correlation coefficient between scores at test and
retest showed that scores for the total scale and for the
subscale �individualizing care� exceeded the r > 0.3 cut-
off for satisfactory test–retest reliability. However, two
of the subscales, �using cognitive assessments and care
interventions� and �using evidence and expertise in
cognition� did not meet the r > 0.3 cut-off.
Discussion
This study aimed to evaluate the psychometric proper-
ties of the POPAC, and the findings support a valid and
reliable use of the scale based on a sample of Australian
Table 2Factor loadings of final items included in the POPAC (N = 212)
Items in factor
Using cognitiveassessments and careinterventions (factor 1)
Using evidenceand expertise in
cognition (factor 2)Individualizingcare (factor 3)
(1) We assess the cognitive status of our older patients on admission 0.702 )0.271 )0.173(2) We make environmental adjustments to avoid over-stimulation in
older people with cognitive impairment (e.g. single rooms, noisereductions etc.)
0.498 )0.177 0.308
(3) We diagnose symptoms of cognitive impairment (e.g. dementias,delirium etc.)
0.564 )0.168 0.083
(4) We spend more time with older patients with cognitiveimpairments as compared to cognitively intact patients
0.723 0.072 0.066
(5) We leave older people with cognitive impairments alone in theward
0.616 0.076 0.063
(6) We use evidence-based tools to assess cognitive status of olderpatients (e.g. the MMSE, SPMSQ, CAM)
)0.065 )0.859 0.059
(7) We consult specialist expertise (e.g. psychologist, gerontologist)if we find that a patient has cognitive impairment
0.347 )0.714 )0.069
(8) We use evidence-based care guidelines in the care of oldercognitively impaired patients
0.062 )0.681 0.023
(9) We use biographical information about older patients� (e.g. habits,interests and wishes etc.) to plan their care
)0.196 )0.341 0.607
(10) We involve family members in the care of older patients withcognitive impairment
0.345 0.065 0.470
(11) We provide staff continuity for older patients with cognitiveimpairments (e.g. the same nurses providing care to thesepatients as often as possible)
)0.128 )0.097 0.664
(12) We systematically evaluate whether or not older patients withcognitive impairment receive care that meets their needs
0.338 0.013 0.533
(13) We involve older patients with cognitive impairment in decisionsabout their care (e.g. examinations, treatments etc.)
0.162 0.149 0.737
(14) We ensure that older patients with cognitive impairment havetests/examinations/consultations in the unit rather than having togo to another department
0.073 )0.218 0.467
(15) We discuss ways to meet the complex care needs of people withcognitive impairment
0.346 )0.026 0.415
Eigenvalue 5.50 1.40 1.08Variance explained by factor 37% 9% 7%Total variance explained by the factor model 53%
Bold values indicate that items (1)-(5) belong to factor 1, (6)-(8) belong to factor 2 and (9)-(15) belong to factor 3.
Construction and psychometric evaluation of the POPAC
ª 2012 Blackwell Publishing LtdJournal of Nursing Management 5
acute hospital staff. The POPAC scale makes a valuable
contribution to the literature by being the first valid and
reliable scale to measure the extent to which acute
nursing staff report using best practice care processes to
identify and consider cognitive impairment and to em-
ploy nursing interventions to meet the needs associated
with old age and cognitive impairment. The scale can
enable descriptions and comparisons between wards,
correlating care practices and patient outcomes, and/or
evaluating the impact of interventions.
It seems that the POPAC scale can make a contribu-
tion to the literature by enabling studies of the extent to
which various acute care settings are �older people
friendly�, i.e. the extent to which the assessments, pro-
cedures, protocols and routines incorporate aspects
such as valid and reliable screening for cognitive
impairment, consulting adequate expertise and using
individualized and needs-based nursing interventions.
This first subscale of the POPAC capture aspects pre-
viously described as being central to best practice care
of older people with cognitive impairment, such as
making environmental adjustments and cognitive
assessments, diagnosing symptoms, spending time with
and not leaving these patients alone in the ward
(Hickman et al. 2007, Day et al. 2008, Moyle et al.
2008, Edvardsson & Nay 2010). The POPAC contrib-
utes by providing a staff-based measure of the extent to
which these important dimensions actually are in place
in acute care settings.
The second subscale illuminates three aspects relating
to evidence-based practice, and measures the extent to
which acute care staff members use evidence-based
tools for their cognitive assessments, consult specialist
expertise, and use evidence-based guidelines/protocols
for the care of older people with cognitive impairment.
Moyle et al. (2010) recently highlighted a number of
shortcomings in the acute care management of older
people with dementia, and they encouraged the in-
creased use of evidence-based practices relating to older
people with cognitive impairment to become part of
hospital policy. The POPAC can assist in highlighting
the extent to which such evidence-based practices are
reported across acute care wards and settings, and if
used in concordance with measures of patient health
outcomes, it could potentially also provide evidence of
the extent to which such practices are associated with
beneficial health outcomes in older patients. Thus, this
scale would seem to have a central place in nursing
research that test and report on best practice models of
care for this population, and such evidence is urgently
needed (Moyle et al. 2008, Cowdell 2010). The POPAC
as a whole might also be relevant to use as a part in
developing a clinical best practice model of care, as it
incorporates some of the most commonly described
central aspects to consider when providing high quality
care of older people with dementia disorders and other
forms of cognitive impairments (Hickman et al. 2007,
Day et al. 2008, Moyle et al. 2008, Edvardsson & Nay
2010). Thus, the item level content can be used to form
a base for developing best practice interventions, have
reflective discussions, and the whole questionnaire as a
pre–post intervention measure to explore if staff report
any practice change as a result of an intervention.
The third subscale of the POPAC illuminates the ex-
tent to which care of older people with cognitive
impairments is individualized or person-centred. There
is evidence to support that the aspects covered by the
third subscale indicates the use of a person-centred
approach in care, such as using biographical informa-
tion, involving family members and patients in care,
using staff continuity as a way to promote therapeutic
relationships, and continuously evaluating and dis-
cussing how to meet the unique needs of these frail
patients. Person-centred care is a contemporary buzz
concept used to describe best practice nursing care in a
range of settings. The concept has recently been sug-
gested as a national safety and quality health services
standard by Australian Commission on Safety and
Quality in Health Care (2011), and it has influenced
health care policy in several western countries, for
example such as Australia, Norway, and Sweden
(Norwegian Ministry of Health and Care Services 2008,
Table 3Test–retest reliability estimates for the final POPAC (n = 25)
Scale dimensionTest
mean € SDRetest
mean € SDTest–Retestmean € SD P
Intra-classcorrelation (CI)
Total score 4.79 € 0.43 4.90 € 0.47 )0.11 € 0.55 0.33 0.41 ()0.31–0.74)Using cognitive assessments and care interventions 4.96 € 0.42 5.06 € 0.59 )0.10 € 0.72 0.51 0.05 ()1.13–0.58)Using evidence and expertise in cognition 4.99 € 0.68 5.00 € 0.51 )0.01 € 0.83 0.94 0.11 ()0.99–0.61)Individualizing care 4.59 € 0.56 4.75 € 0.61 )0.16 € 0.64 0.23 0.56 (0.01–0.80)
SD, standard deviation; CI, confidence interval.
D. Edvardsson et al.
ª 2012 Blackwell Publishing Ltd6 Journal of Nursing Management
Swedish National Board of Health and Welfare 2010,
Victorian Government Department of Human Services
2011). The concept is most often used to describe a
holistic model care that is based on the patients� prior-
itized needs and wishes, involves the person and their
significant others, promoting beneficial relationships,
and incorporates the use of individually tailored psy-
chosocial and complementary interventions inasmuch
as traditional medical interventions (McCormack &
McCance 2006, Edvardsson et al. 2008). Thus, it seems
that the POPAC can be used as a measure to indicate
the extent to which care not only is evidence-based, but
person-centred, and this supports both the theoretical
and clinical relevance of this scale as these two con-
ceptual dimensions arguably are central to nursing care
quality.
Limitations
There are some limitations to this study that need
consideration. The 59% response rate makes it possible
that the results of the study might have been different if
a larger number of staff had chosen to participate, given
that non-responding staff members could have been
different from the responders. As the study employed
voluntary and anonymous participation for ethics pur-
poses, the reasons for not participating remains un-
known and comparative analyses between responders
and non-responders were not possible. Thus, we can
only speculate to what extent the final study sample is
truly representative of the overall sample. However, this
limitation indicates a need for further application and
evaluation of the POPAC in additional samples. An-
other limitation is the non-randomized, cross-sectional
design of the study, which puts constraints on the extent
to which these finding can be generalized to other set-
tings and samples. These design issues underscore the
value of additional testing of the POPAC to further
confirm or reject the psychometric properties of this
scale in various samples and cultures. In addition, as the
scale makes use of self-reporting on behalf of staff, it
needs to be complemented with other observational
approaches if information is sought about the extent to
which staff reporting corresponds to what actually
happens in the wards.
Furthermore, as reported in Table 1 the test–retest
estimates provided slightly contrasting data for two of
the subscales, �using cognitive assessments and care
interventions� and �using evidence and expertise in
cognition�. The intraclass correlation coefficient did not
exceed the cut-off of r > 0.3, whilst the paired samples
t-test showed that the scores for these subscales was not
significantly different between test and retest. These
findings, together with the fact that test–retest reliability
was based on a limited sub-sample of nursing staff on
one orthopaedic ward, indicate that the extent to which
the POPAC and its subscales have temporal stability in
other settings need further testing.
To conclude, this study presented promising tentative
data to support a valid and reliable use of the POPAC
based on an Australian sample of acute hospital nursing
staff. However, further studies are needed and invited
to explore the psychometric properties and the clinical
applicability of the scale in other samples and settings.
Conclusions and implications for nursingmanagement
The new POPAC scale is relevant to nursing manage-
ment as it is short, easy to administer and not time
consuming to complete, but still provides clinically rel-
evant, valid and reliable information on the extent to
which acute nursing staff report using best practice care
processes to identify and consider cognitive impairment
and to employ nursing interventions to meet the needs
associated with old age and cognitive impairment. The
main contribution to nursing management might be that
the scale can be used as a conceptual fundament in
developing best practice models of nursing care in the
acute setting, as it draws upon and contains most com-
monly described aspects relating to high quality nursing
care of older people with cognitive impairments. Thus, it
can help clinical staff to develop interventions increas-
ingly to apply best practice care of older people with
cognitive impairments. It can also be relevant for nursing
management, by providing a valid and reliable tool for
clinical audits of how �older people friendly� their wards
and/or units are. In a time where population ageing im-
pacts on most acute care settings in the Western world, it
seems reasonable to argue for a high clinical relevance
for a measure providing relevant, valid and reliable data
on the extent to which such acute settings are experi-
enced to incorporate evidence-based, best practice
nursing care of older people with cognitive impairment.
Conflict of interest statement
The authors have no conflicts of interest to declare.
Sources of funding
The study was funded by grants from the Swedish Re-
search Council, and the Gothenburg Centre for Person-
Centred Care.
Construction and psychometric evaluation of the POPAC
ª 2012 Blackwell Publishing LtdJournal of Nursing Management 7
Ethical approval
Ethics approval for the study was obtained from La
Trobe University and the hospital Human Research
Ethics Committees.
References
Australian Commission on Safety and Quality in Health Care
(2011) Patient-centred care: improving quality and safety by
focusing care on patients and consumers. Available at: http://
www.safetyandquality.gov.au/internet/safety/publishing.nsf/
Content/36AB9E5379378EBECA2577B3001D3C2B/$File/
PCCC-DiscussPaper.pdf, accessed 17 January 2012.
Australian Institute of Health and Welfare (2006) Australian
Hospital Statistics 2004–05. AIHW, Canberra.
Cowdell F. (2010) Care of older people with dementia in an acute
hospital setting. Nursing Standard 24, 42–48.
Day J., Higgins I. & Koch T. (2008) Delirium and older people:
what are the constraints to best practice in acute care? Inter-
national Journal of Older People Nursing 3, 170–177.
DeFrances C.J. & Hall M.J. (2007) 2005 National Hospital
Discharge Survey. Advance Data from Vital and Health
Statistics, no 385. National Center for Health Statistics,
Hyattsville, ML.
Dick K.L. (1998) Acute confusion in the elderly hospitalized
patient: an exploration of experienced nursing care. Doctoral
dissertation. University of Rhode Island. ISBN: 978-0-591-
99673. Available at: http://digitalcommons.uri.edu/disserta
tions/AAI9902556, accessed on 17 January 2012.
Edvardsson D. & Nay R. (2010) Acute care and older people:
challenges and ways forward. Australian Journal of Advanced
Nursing 27, 63–69.
Edvardsson D., Winblad B. & Sandman P.O. (2008) Person-
centred care for people with Alzheimer�s disease – current status
and ways forward. Lancet Neurology 7, 362–367.
Hickman L., Newton P., Halcomb E.J., Chang E. & Davidson P.
(2007) Best practice interventions to improve the management
of older people in acute care settings: a literature review.
Journal of Advanced Nursing 60, 113–126.
Inouye S.K. (2006) Delirium in older persons. Review. New
England Journal of Medicine 354, 1157–1165.
Joray S., Wietlisbach V. & Bula C. (2004) Cognitive impairment
in elderly medical inpatients. American Journal of Geriatric
Psychiatry 12, 639–647.
McCormack B. & McCance T.V. (2006) Development of a
framework for person-centred nursing. Journal of Advanced
Nursing 56, 472–479.
Moyle W., Olorenshaw R., Wallis M. & Borbasi S. (2008) Best
practice for the management of older people with dementia in
the acute care setting: a review of the literature. International
Journal of Older People Nursing 3, 121–130.
Moyle W., Borbasi S., Wallis M., Olorenshaw R. & Gracia N.
(2010) Acute care management of older people with dementia:
A qualitative perspective. Journal of Clinical Nursing 20, 420–
428.
Norwegian Ministry of Health and Care Services. (2008)
Dementia plan 2015. Available at: http://www.regjeringen.no/
upload/HOD/Dokumenter%20KTA/DementiaPlan2015.pdf,
accessed on 17 January 2012.
Nunnally J.C. (1994) Psychometric Theory. McGraw, New York,
NY.
Royal College of Psychiatrists (2005) Who Cares Wins. Royal
College of Psychiatrists, London.
Samson E.L., Blanchard M.R., Jones L., Tookman A. & King M.
(2009) Dementia in the acute hospital: prospective cohort study
of prevalence and mortality. British Journal of Psychiatry 195,
61–66.
Shrive F.M., Stuart H., Quan H. & Ghali W.A. (2006) Dealing
with missing data in a multi-question depression scale: com-
parison of imputation methods. BMC Medical Research
Methodology 6, 57.
Soderqvist A. (2007) Assessment of cognitive function in elderly
patients with hip fractures. Doctoral dissertation. The Karo-
linska Institute. ISBN 91-7357-065-6. Available at: http://
hdl.handle.net/10616/38613, accessed 17 January 2012.
Streiner D. (2008) Health Measurement Scales: A Practical Guide
To Their Development and Use. Oxford University Press,
Oxford.
Swedish National Board of Health and Welfare (2010) Nationella
riktlinjer for vard och omsorg vid demenssjukdom 2010 – stod
for styrning och ledning. Available at: http://www.social
styrelsen.se/Lists/Artikelkatalog/Attachments/18012/2010-5-1.pdf,
accessed 17 January 2012.
Victorian Government Department of Human Services (2011)
Improving care for older people. A policy for health services.
Available at: http://docs.health.vic.gov.au/docs/doc/60D29439
F093F9E7CA25792B0077226D/$FILE/improvingcare.pdf, accessed
5 August 2011.
Waszynski C. (2007) How to try this: Detecting delirium.
American Journal of Nursing 107, 50–59.
Wimo A., Winblad B. & Jonsson L. (2010) The worldwide
societal costs of dementia: Estimates for 2009. Alzheimer�s &
Dementia 6, 98–103.
Zeleznik J. (2001) Delirium: Still searching for risk factors and
effective preventive measures. Journal of the American Geri-
atrics Society 49, 1729–1732.
D. Edvardsson et al.
ª 2012 Blackwell Publishing Ltd8 Journal of Nursing Management