This article was downloaded by: [Bangor University]On: 20 December 2014, At: 07:59Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Psychology & HealthPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/gpsh20

Illness perceptions and distress inwomen at increased risk of breastcancerG. Rees a , A. Fry b , A. Cull c & S. Sutton da Department of Psychology and Disability Studies , RMITUniversity , PO BOX 71, Bundoora, Victoria, Melbourne, Australiab Cancer Research UK, Edinburgh Oncology Unit, Western GeneralHospital , Crewe Road, Edinburgh EH4 2XR, Scotlandc Director of Training for Psychology Services , NHS Educationfor Scotland , 2nd Floor Hanover Buildings, 66 Rose Street,Edinburgh, EH2 2NN, Scotlandd Institute of Public Health, University of Cambridge , RobinsonWay, Cambridge, CB2 2SR, EnglandPublished online: 01 Feb 2007.

To cite this article: G. Rees , A. Fry , A. Cull & S. Sutton (2004) Illness perceptions anddistress in women at increased risk of breast cancer, Psychology & Health, 19:6, 749-765, DOI:10.1080/08870440412331279764

To link to this article: http://dx.doi.org/10.1080/08870440412331279764

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,

systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

Psychology and HealthDecember 2004, Vol. 19, No. 6, pp. 749–765

ILLNESS PERCEPTIONS AND DISTRESS

IN WOMEN AT INCREASED RISK OF

BREAST CANCER

G. REESa,*, A. FRYb, A. CULLc and S. SUTTONd

aDepartment of Psychology and Disability Studies, RMIT University, PO BOX 71, Bundoora,Victoria, Melbourne, Australia; bCancer Research UK, Edinburgh Oncology Unit, WesternGeneral Hospital, Crewe Road, Edinburgh EH4 2XR, Scotland; cDirector of Training for

Psychology Services, NHS Education for Scotland, 2nd Floor Hanover Buildings, 66 Rose Street,Edinburgh, EH2 2NN, Scotland; dInstitute of Public Health, University of Cambridge, Robinson

Way, Cambridge, CB2 2SR, England

(Received 26 August 2003; In final form 8 June 2004)

Variation in the levels of distress in women at increased risk of breast cancer has been reported, yet there islimited understanding of the factors that are associated with heightened distress in this population. This studytook a theoretical approach using Leventhal’s Self Regulatory Model (SRM) to understand variation in dis-tress levels. The study examined the associations between perceptions of breast cancer and distress in womenat increased risk of breast cancer, and a comparison sample with no experience of the disease in their socialenvironment. Questionnaire data from 117 women at increased risk of breast cancer and 100 comparisonwomen were analysed. Women at increased risk of breast cancer showed comparable levels of general distressbut significantly higher levels of cancer specific distress than the comparison group. There were few differencesin illness perceptions between the samples, although a number of cognitive perceptions of breast cancer wererelated to both general and cancer specific distress in the increased risk sample, but not in the comparisonsample. The results suggest that the SRM provides a useful framework to explore the psychological responseto genetic risk. Further research is required in this population to examine illness perceptions in more detail,validate quantitative measures of illness perceptions, and examine interactions between risk perception andthe SRM constructs.

Keywords: Breast cancer; Psychological distress; Cancer risk; Illness perceptions; Self-regulatory model;Genetic risk

INTRODUCTION

Familial Breast Cancer

A family history of breast cancer has been recognised as a risk factor for the disease formany years. It has been estimated that about 5–10% of breast cancer cases are causedby hereditary factors. Recent research has identified autosomal dominant genes thatwhen mutated predispose the carrier to developing breast cancer (Futreal et al., 1994;

*Corresponding author. Tel.: þ61 3 99256525. Fax: þ61 3 99257695. E-mail: Gwyneth.Rees@rmit.edu.au

ISSN 0887-0446 print: ISSN 1476-8321 online � 2004 Taylor & Francis Ltd

DOI: 10.1080/08870440412331279764

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

Miki et al., 1994; Wooster et al., 1995). These genes do not show 100% penetrance, thelifetime risks associated with a mutation estimated as being between 60–80% (Collins,1996; Struewing et al., 1997). Women identified as carrying the BRCA1 mutation, alsohave an increased risk of ovarian cancer (Ford et al., 1998). The role of environmentalfactors in phenotypic expression of risk has been unclear (Evans et al., 1994).

Although many women will have a relative who has suffered from breast cancer, onlya small proportion of them are likely to be at increased risk of the disease (Mettlin,1994). The risk criteria to identify women at increased risk of breast cancer includethe number of affected relatives, relationship to these relatives and importantly, theage of the relatives at diagnosis (familial breast cancer occurs at a younger age thansporadic breast cancer) (e.g. Scottish Intercollegiate Guidelines Network [SIGN],1998). Assessment of family history in order to obtain a genetic risk estimate can bedifficult due to inaccuracy of self reported family history and medical records, andinsufficient female family members to allow expression of the condition (Richards,1999). Familial breast cancer clinics have been established in order to provide riskassessment and risk management for those women who are deemed to have a significantfamily history. There are no proven preventative options for breast cancer and there-fore early detection is paramount in reducing mortality. The women at increased riskof breast cancer are offered regular mammography from when they are five yearsyounger than the youngest age at which any relative was diagnosed with breastcancer, but at no younger than 35 years or older than 40 years (Scottish Executive,2001). However, the effectiveness of screening younger women is controversial, anddata are only beginning to emerge to support the effectiveness among youngerwomen selected for their familial risk (Macmillan, 2000; Tilanus-Linthorst, Bartels,Obdeijn and Oudkerk, 2000). High risk women may be offered prophylactic bilateralmastectomy which has been found to be effective in reducing the incidence of breastcancer in BRCA1 or BRCA2 mutation carriers (Meijers-Heijboer et al., 2001).Chemoprevention is currently being offered within the context of a clinical trial andlikely to produce distressing and debilitating side effects (Emery, Murphy andLucassen, 2000). Women with a family history of breast cancer therefore face muchuncertainty regarding if and when cancer would develop, and how they shouldmanage that risk.

Distress in Women at Increased Risk of Breast Cancer

Given the ambiguous nature of breast cancer risk, information and uncertain manage-ment options, it is not surprising that high levels of psychological morbidity in womenat increased risk of breast cancer have been reported (Kash, Holland, Halper andMiller, 1992; Lerman and Schwartz, 1993; Gagnon et al., 1996). However, other studieshave reported that women with at least one first degree relative who has suffered frombreast cancer show levels of distress comparable to controls without a family history ofthe disease (Wellisch, Gritz, Schain, Wang and Siau, 1991; Lerman, Kash and Stefanek,1994; Lloyd et al., 1996; Zakowski et al., 1997). Prospective studies assessing the impactof genetic risk counselling for breast cancer on levels of general and cancer specific dis-tress have also been inconsistent (Hopwood et al., 1998; Cull et al., 1999; Watson et al.,1999). The methodological differences between studies in terms of sampling and assess-ment methods may account for variation in the level of distress. However, it is clearthat individual differences in the levels of distress exist and that subgroups of women

750 G. REES et al.

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

maintain high levels of anxiety and worries about breast cancer (Appleton, Fry, Rees,Rush and Cull, 2000).

There is limited understanding of the cause of variation in distress in women atincreased risk of breast cancer and studies have tended to be exploratory and descrip-tive. A number of studies have been focused on the accuracy of risk perception and theassociations between risk perception and the psychological response to risk (Lermanet al., 1994; Lloyd et al., 1996; Hopwood et al., 1998, 2001; Cull et al., 1999; Watsonet al., 1999; Kent, Howie, Fletcher, Newbury-Ecob and Hosie, 2000). These studieshave suggested that risk perception is positively associated with cancer specific distress.Other factors found to be associated with distress include, age (Lerman et al., 1994; Cullet al., 1999), optimism (Audrain et al., 1997), and coping style (Lerman et al., 1996;Audrain et al., 1997). A number of qualitative studies have also identified thatwomen’s experiences of breast cancer in their family is strongly associated with an emo-tional response to their own risk (Wellisch et al., 1991, 1992, 1996; Chalmers andThomson, 1996). A few theoretically driven studies aimed at predicting psychosocialresponse to risk have also been conducted. Further research is needed to explorethe cause of distress in a theoretical manner in order to establish and examine causalassociations that can identify factors amenable to intervention (Rees, Fry and Cull,2001).

A Theoretical Understanding

One model from health psychology that can help put a theoretical light on this issue isLeventhal’s self regulatory model (SRM). This model proposes that individuals activelygenerate cognitive and emotional representations of health threats and that these repre-sentations guide and regulate behaviour (Leventhal et al., 1997). The model indicatesthat internal stimuli (e.g. the experience of symptoms) as well as stimuli from the envi-ronment (e.g. risk information, witnessing a relative’s illness) may trigger cognitive andemotional representations. Based on these representations individuals derive an actionplan to cope with the threat they perceive. The success of a particular coping strategy isappraised and feeds back into both the representation and the action plan, which maybe modified accordingly. The SRM, therefore, provides a relevant framework fromwhich to understand women’s response to breast cancer risk for a number of reasons.Crucially, the SRM enables a greater understanding of the meaning of risk by focusingon what individuals perceive they are at risk of. It enables the exploration of both thecognitive and the emotive representations in determining outcome and also allows afocus on the emotional outcomes rather than the behavioural outcomes that are thefocus of the Health Belief Model (Rosenstock, 1966; Becker, 1974) and Theory ofReasoned Action (Ajzen and Fishbein, 1980).

Initial work on the SRM suggested that cognitive illness representations were organ-ised around five dimensions: identity (symptoms associated with the illness); time line(beliefs about the duration of the illness); consequences (beliefs about the effects);control/cure (beliefs about its controllability and recovery); and cause (perceivedcauses of the illness) (Leventhal et al., 1997). A quantitative measure, the IllnessPerception Questionnaire (IPQ) was developed and revised to assess these dimensions(Weinman, Petrie, Moss-Morris and Horne, 1996; Moss-Morris et al., 2002).

Work on the SRM has mainly focused on the cognitive component of the model, anda number of qualitative and quantitative studies across a range of patient population

ILLNESS PERCEPTIONS AND BREAST CANCER 751

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

and studies have demonstrated associations between illness perceptions and emotional,cognitive and behavioural responses to illness (e.g. Petrie, Weinman, Sharpe andBuckley, 1996; Scharloo et al., 1998; Heijmans, 1999; Jessop and Rutter, 2003;Llewellyn, Miners, Lee, Harrington and Weinman, 2003). For example, illness percep-tions have been found to predict psychological well-being in patients with ChronicFatigue Syndrome and Rheumatoid Arthritis (Heijmans and De Ridder, 1998;Murphy, Dickens, Creed and Bernstein, 1999); self-management of diabetes andosteoarthritis (Hampson, Glasgow and Toobert, 1990; Hampson, Glasgow and Zeiss,1994); adherence to medication among patients with asthma and haemophilia (Jessopand Rutter, 2003; Llewellyn et al., 2003) and return to work after a MyocardialInfarcation (Petrie et al., 1996). A recent meta-analysis of the studies is available(Hagger and Orbell, 2003).

In relation to breast cancer, Buick (1997) found that illness perceptions were import-ant predictors of psychosocial response to treatment, independent of objective illnessseverity. Causal beliefs about breast cancer have also been associated with patients’adjustment to illness (Taylor, Lichtman and Wood, 1984). Early work by Leventhalon breast cancer patients found that the patients’ representations of breast cancervaried as a reflection of their experience, including variations in type of carcinoma,natural history of the disease and treatment type (Leventhal, Easterling, Coon,Luchterhand and Love, 1986). Buick (1997) also found that women’s recommendedtreatment for breast cancer had a large impact on their perceptions of the disease.The women recommended for chemotherapy perceived breast cancer to be longer induration and more severe than those patients recommended for radiation treatment.

Although the SRM has primarily been applied to understanding psychologicalresponse in physically ill patients, illness perceptions are also likely be important predic-tors of healthy individuals’ response to health-threats such as a genetic predisposition tocancer (Decruyenaere, Evers-Kiebooms, Welkenhuysen, Denayer and Claes, 2000).Individuals with a family history of breast cancer are likely to have witnessed andbeen affected by the experience of relatives who have suffered from the disease.Some studies have described women at increased risk of breast cancer as having livedvicariously through their relative’s breast cancer and having shared the illness experi-ence (Chalmers and Thomson, 1996). These experiences are likely to generate strong ill-ness perceptions that guide women’s emotional and behavioural response to their owngenetic risk (Rees et al., 2001). The SRM would suggest that women with differentexperiences of breast cancer will construct different representations of the diseaseand choose different coping strategies for dealing with risk (Rees et al., 2001; Rees,2003). Using the SRM in this context may help us to understand and predict thelevels of distress, the screening behaviour and the decisions about preventative optionssuch as prophylactic surgery.

This study used the SRM as a framework to examine emotional adjustment inwomen at increased risk of breast cancer. The study focused on the cognitive represen-tation component of the model and had three main objectives. Firstly, to compare ill-ness perceptions of breast cancer in women at increased risk due to their family historyand a comparison group of women with no familial experience of the disease. Secondly,to examine associations between illness perceptions and distress in women at increasedrisk of breast cancer and the comparison sample. Thirdly, to examine the predictivevalue of illness perceptions in determining level of distress in women at increased riskof breast cancer.

752 G. REES et al.

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

METHOD

Participants

The study design was a cross-sectional postal questionnaire.

Increased Risk Sample

The participants were 200 women, randomly selected, from the database held by theSouth East Scotland Familial Breast Cancer Clinic. These women were deemed to beat least ‘moderate’ increased risk of breast cancer by a clinician at the South EastScotland Familial Breast Cancer Clinic (this means two to three times the general popu-lation risk). These women had received genetic counselling, and were on an annual pro-gramme of mammography and clinical breast examination. Prior to the selection,women on the database were excluded for the following reasons to ensure that thesample was as homogeneous as possible in respect to factors that may influence percep-tions of breast cancer:

. Ovarian family history;

. Participation in chemo-prevention;

. Received genetic testing;

. Received preventative surgery.

Comparison Sample

The participants were 592 women who were identified from the Community HealthIndex at Lothian Health Board. This is a register of all individuals registered withGPs in the Lothian region of South East Scotland. These women were selected onthe basis of their age and postcode sector in order that the sample was comparableto the increased risk sample on socio-demographic factors (e.g. age, education and eth-nicity). These factors may influence illness perceptions (Klonoff and Landrine, 1994).For each woman at increased risk of breast cancer, three women with comparableage and postcode details were sought.

Women who self reported any experiences of breast cancer in their family, friends,work or ‘other’ experiences of breast cancer were excluded, so that the remainingsample reflected a sample of women in the general population without any experienceof breast cancer in their social environment.

Additional Exclusion Criteria

In both the samples, prior to contacting the participants, the GPs of those women werecontacted. The GPs were provided with a study information sheet and were asked tocontact the researcher, if they considered any patient to be unsuitable for the study.The GPs were specifically asked to check the following exclusion criteria and toinform the researcher if the patient:

. Had a previous diagnosis of cancer in the past;

. Was currently undergoing investigation for cancer;

. Was currently suffering from any other serious illness;

. Was currently suffering from alcoholism, schizophrenia or organic brain damage.

ILLNESS PERCEPTIONS AND BREAST CANCER 753

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

Participants who were eligible for the study were sent an information sheet and a con-sent form. The women who consented were sent a questionnaire. A reminder was sent,if the questionnaire was not returned within 3 weeks.

Test Retest

The women in the increased risk sample who had returned their questionnaire withouta reminder were followed up within 3 months, with an additional questionnaire,in order to obtain a test–retest reliability data.

Measures

Illness Perceptions

The Revised Illness Perception Questionnaire (IPQ-R) (Moss-Morris et al., 2002) isa theoretically derived measure, designed to assess dimensions of illness perceptions(identity, time line, consequences, control and cause) across a range of illnesses(Weinman et al., 1996; Moss-Morris et al., 2002). The measure was recently revisedto include measures of perceptions of duration of illness (‘time acute/chronic’) andfluctuation in illness over time (‘time line cyclical’). In addition, the revised versionalso distinguishes perceptions of ‘treatment control’ and ‘personal control’ overillness. The revised version includes a new measures of ‘illness coherence’ – howclear and comprehensive an individual feels her illness to be. A new measureof ‘emotional representations’ is also included (Moss-Morris et al., 2002). Theemotional representations subscale was not included in this analysis, since theaim of the study is to determine the contribution of cognitive representations ofbreast cancer on emotional adjustment. Including a measure of emotional represen-tations as an independent variable would confound the outcome variables. The50-item version of the IPQ-R was available during the design phase of this studyand was adapted and reworded to make it appropriate for the healthy womenat increased risk of breast cancer. Participants were asked to report theirpersonal views about breast cancer rather than referring to their perceptions ofan illness personally affecting them. For example: ‘My illness has serious financialconsequences’ was replaced with ‘Breast cancer has serious financial consequences’;‘My illness will last for a long time’ was replaced with ‘Breast cancer lasts for along time’.

Identity is assessed using a 17-item symptom checklist. Following previous researchon perceptions of breast cancer patients (Buick, 1996), discussions with breast cancerconsultants and women at increased risk, breast cancer specific items were generatedand added to the existing checklist. These included (‘hard or tender growths inbody’, ‘soreness in body’, ‘skin changes’). Causal beliefs are assessed individually.Participants endorse 19 items, referring to the causes of breast cancer on a 5-pointscale ranging from ‘strongly disagree’ to ‘strongly agree’. The item ‘hormonal’ wasadded to the cause subscale since oestrogen has been associated with breast cancerdevelopment (Vogel, 2000). The items for the remaining subscales are presented ina random order. Responses are rated on a 5-point scale from ‘strongly disagree’ to‘strongly agree’. Subscale scores are the mean of items (after reverse scoring as neces-sary). Subscales with missing items are not computed. The number of items, internal

754 G. REES et al.

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

reliability (Cronbach’s alpha) and test–retest reliability for each subscale in theincreased risk sample is outlined in Table I. The adapted IPQ-R was piloted on 11women at increased risk of breast cancer to ensure that all items were relevant andunderstood appropriately. A copy of the adapted questionnaire is available from theauthors.

Psychological Well-being

General distress The General Health Questionnaire 30-item version (GHQ-30)(Goldberg and Williams, 1998) is a well-validated self-administered screening testaimed at detecting psychiatric disorder in community and non-psychiatric clinical set-tings. The participants respond to thirty statements about their general health overthe past few weeks on a four-point scale. The GHQ-30 was scored using both, theLikert method (0, 1, 2, 3) to investigate individual differences in levels of distress andthe ‘alternative’ scoring system (0, 0, 1, 1) using a cut off score of >5 to determine pre-valence of ‘case level’ of distress that warrants further clinical assessment (Goldbergand Williams 1998).

Specific cancer related distress The Cancer Worry Scale (Watson et al., 1999) is a6-item scale (adapted from four single items, Lerman et al., 1991a,b), that assessesthe frequency of cancer worries and the degree to which these worries affect moodand daily activities. Items are scored from 1–4 and summed for a total score. Thescale has been used in populations at increased risk of cancer and shows good internalreliability (Brain, Norman, Gray and Mansel, 1999; Hopwood et al., 2001). Cronbachalpha in this study was 0.83, and test–retest reliability was r¼ 0.74, p<0.001.

Risk Perception

A single item ‘How likely do you feel it is that you will develop breast cancer in yourlifetime?’ rated on a 5-point Likert scale (1, ‘very unlikely’; 2, ‘unlikely’; 3, ‘likely’; 4,‘very likely’; 5, ‘extremely likely’).

Statistical Analysis

Chi-square test was used to test for differences in categorical variables (GHQ ‘caseness’)and t-tests were used to test for differences on continuous variables (GHQ Likert scoreand cancer worry score).

TABLE I Reliability of IPQ-R subscales in women at increased risk of breast cancer

Subscale Number ofitems

Cronbach’salpha

Test-retest reliability(Pearson’s correlation

coefficient)

Timeline acute/chronic 5 0.61 0.68***Timeline cyclicala 6 0.35 0.33**Consequences 11 0.61 0.77***Personal control 9 0.82 0.74***Treatment control 6 0.72 0.77***Illness coherence 5 0.81 0.68***

aBecause of the low internal reliability of this subscale, it was omitted from subsequentanalysis;*p<0.05; **p<0.01; ***p<0.001

ILLNESS PERCEPTIONS AND BREAST CANCER 755

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

RESULTS

Response Rates and Background Demographics

In the increased risk sample, 16 women were excluded by the GP before questionnaireswere administered. One hundred and eighty-four women were invited to take part in thestudy and 117 women returned their questionnaire, giving a response rate of 64%.Ninety-nine of these women returned the questionnaire without a reminder and weresent a follow-up questionnaire to obtain test–retest data. Seventy-four of thesewomen (75%) returned this questionnaire.

In the general population sample, 45 women were excluded by the GP. Replacementswere found for 15 of these women. Therefore, 562 women were invited to take part inthe study and 258 returned the questionnaires giving a response rate of 46%. Two ofthese questionnaires were excluded due to protocol violation (they reported to haveexperienced cancer themselves). From the 256 responses that were retained for analysis,100 did not report any experience of breast cancer in family, friends or at work. These100 women formed the comparison sample.

Almost 100% of women in both samples were Caucasian. Both samples were com-parable on socio-demographic variables (age, education levels, marital status, andmaternity). The characteristics of the sample are provided in Table II. Both samplesconsisted of well-educated, middle aged women, a majority of whom were marriedwith children.

Women in the increased risk sample were significantly more likely to believe they willdevelop breast cancer in their lifetime than the comparison sample (t¼ 7.48, df¼ 211,p<0.001). In the increased risk sample 77% believed they were ‘likely’, ‘very likely’or ‘extremely likely’ to develop breast cancer in their lifetime compared with 33% inthe comparison sample.

Differences Between the Samples: Distress

Table III shows a descriptive data for the psychological well-being measures in eachsample and the differences between the samples. The results show that the samples

TABLE II Sample characteristics

Variable Increased risk sample(n¼ 117)

Comparison sample(n¼ 100)

Differencetest (df)

Age (years) Mean¼ 40.4 (7.09) Mean¼ 40.9 (7.31) t¼ 1.19 (215)% educated beyond 18 44.5% 39% �2¼ 0.49 (1)% married or living with a partner 88% 76% �2¼ 0.18 (1)% with at least one child 71% 80% �2¼ 2.37 (1)

*p<0.05; **p<0.01; ***p<0.001

TABLE III Levels of distress in each sample

Distress measure Increased risk sample(n¼ 117). Mean (SD)

Comparison sample(n¼ 100). Mean (SD)

Differencetest (df)

GHQ score 27.0 (11.86) 27.4 (10.72) t¼ 0.24 (212)Cancer worry scale 10.7 (2.58) 8.9 (2.0) t¼ 5.56 (213)***

*p<0.05;**p<0.01; ***p<0.001

756 G. REES et al.

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

did not show different levels of general distress ( p<0.05). There was also no differencein the proportion of GHQ cases in each sample (�2¼ 0.009, df¼ 1, p¼ 0.93). However,the increased risk sample reported significantly higher levels of cancer specific distressthan the comparison sample ( p<0.001).

Differences Between the Samples: Illness Perceptions

Table IV shows the mean scores for each of the IPQ-R subscales and the causal itemsfor each sample and differences between the samples. The results indicate that womenat increased risk of breast cancer had a more coherent understanding of breast cancerthan the comparison group ( p<0.001). The results also suggest that women atincreased risk of breast cancer viewed the consequences of breast cancer to be moresevere than the comparison group ( p¼ 0.05). However, the actual difference betweenscores on the consequences measure was small. The samples differed on a number ofcausal beliefs. Women at increased risk of breast cancer were more likely to believethat ‘hereditary’, ‘ageing’ and ‘hormonal factors’ were the causes of breast cancer,and less likely to believe that ‘poor medical care in the past’ or ‘a germ or virus’were causes of breast cancer ( p<0.05).

TABLE IV Illness perceptions in each sample

IPQ-R subscale Increased risk sample(n¼ 117). Mean (SD)

Comparison sample(n¼ 100). Mean (SD)

Differencetest (df)

Identity 4.44 (2.25) 4.98 (2.88) t¼�1.56 (215)Timeline acute 3.27 (0.53) 3.19 (0.49) t¼ 0.99 (204)Consequences 3.94 (0.38) 3.83 (0.41) t¼ 1.97 (201)*Personal control 3.25 (0.60) 3.25 (0.43) t¼ 0.078 (200)Treatment control 3.62 (0.52) 3.62 (0.36) t¼�0.088 (205)Illness coherence 2.45 (0.69) 3.07 (0.64) t¼�6.65 (203)***

Causal items:Stress or worry 3.16 (0.99) 3.19 (0.91) t¼�0.25 (211)Hereditary-it runs in the family 4.56 (0.49) 4.24 (0.64) t¼ 4.45 (df 214)***A germ or virus 2.12 (0.93) 2.36 (0.83) t¼�1.98 (210)*Diet or eating habits 3.31 (0.98) 3.08 (0.93) t¼ 1.75 (210)Chance or bad luck 3.03 (1.12) 2.96 (1.04) t¼ 0.51 (212)Poor medical care in the past 2.26 (0.92) 2.60 (0.84) t¼�2.85 (211)**Pollution in the environment 2.97 (0.91) 3.07 (0.83) t¼�0.84 (211)Patient’s own behaviour 2.53 (0.95) 2.59 (.87) t¼�0.502 (211)Patient’s mental attitude, e.g. thinking

about life negatively2.54 (0.93) 2.53 (0.83) t¼ 0.059 (211)

Family problems or worries causesbreast cancer

2.52 (0.87) 2.54 (0.83) t¼�0.12 (210)

Overwork 2.37 (0.75) 2.47 (0.73) t¼�1.04 (212)Emotional state e.g. feeling down,

lonely, anxious, empty2.5 (0.85) 2.5 (0.84) t¼�0.044 (211)

Ageing 3.28 (0.93) 3.00 (0.91) t¼ 2.18 (212)*Alcohol 2.68 (0.91) 2.77 (0.84) t¼�0.72 (209)Smoking 3.41 (0.99) 3.49 (0.87) t¼�0.66 (210)Accident or injury 2.61 (0.95) 2.76 (0.96) t¼�1.15 (209)Patient’s personality 2.21 (0.78) 2.16 (0.66) t¼ 0.4 (212)Altered immunity 2.89 (1.05) 3.01 (0.83) t¼�0.86 (213)Hormonal 3.69 (0.75) 3.46 (0.81) t¼ 2.19 (213)*

*p<0.05;**p<0.01;***p<0.001

ILLNESS PERCEPTIONS AND BREAST CANCER 757

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

Associations Between Illness Perceptions and Distress

Table V shows the Pearson correlations between illness perceptions, age, risk percep-tions and distress measures in both the samples. The results indicate that both generaldistress and cancer worry are associated with higher scores on the identity, timelineacute and consequences subscales in the increased risk sample ( p<0.05). This suggeststhat women at increased risk of breast cancer with higher levels of distress, view breastcancer as more chronic, more severe, and associate more symptoms with the disease. Inthe comparison sample, no subscales were associated with general distress and only theidentity dimension was positively associated with cancer worry ( p<0.001). Due to thelarge number of causal items a more stringent significance level ( p<0.01) was adopted.No causal items were found to be significantly correlated distress measures in eithersample ( p<0.01).

Predicting Distress

Simultaneous multiple regression models were computed in order to examine the pre-dictive value of illness perception and risk perception on general distress and cancerworry in both samples. The models had two steps. In the first step, age was enteredas a control variable and the IPQ-R variables were entered. In the second step, risk per-ception was added in order to determine its remaining contribution. Tables VI and VIIsummarise the results.

The results showed that the models were not significant in predicting general distressin either sample ( p<0.05). The proportion of variance accounted for by illnessperception variables was low in both samples and risk perception was not a significantpredictor. The only significant cognitive variables were the identity dimension in theincreased risk sample and the illness coherence dimension in the comparison sample.This suggests that identifying more symptoms of breast cancer was predictive of greatergeneral distress in the increased risk sample and that having a less clear understandingof breast cancer was predictive of general distress in the comparison sample. In thecomparison sample age was also a predictor of general distress. Increased age signifi-cantly predicted heightened distress in this sample.

The models were significant in predicting cancer worry. The final model in theincreased risk sample accounted for 26% of the variance in cancer worry ( p<0.001).

TABLE V Pearson’s correlations between distress measures and other measures in each sample

Increased risk sample (n¼ 117) Comparison sample (n¼ 100)

GHQ score Cancer worry GHQ score Cancer worry

Age 0.015 �0.17 0.266** 0.007Risk perception 0.18* 0.43*** 0.137 0.222*Identity 0.296*** 0.187* 0.013 0.394***Time line acute 0.209* 0.253** 0.134 0.108Consequences 0.203* 0.200* 0.122 0.169Personal control �0.027 �0.139 �0.157 0.092Treatment control �0.110 �0.135 �0.137 �0.115Illness coherence �0.013 �0.154 �0.201 �0.123

*p<0.05; **p<0.01; ***p<0.001

758 G. REES et al.

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

However, risk perception was found to be the only significant predictive cognitive vari-able. Higher perceived risk was predictive of higher levels of cancer worry. Age was alsoa significant predictor of cancer worry in this sample. Younger age was predictive ofheightened cancer worry. In the comparison sample the final model accounted for14% of variance in cancer worry ( p<0.05). However, in this sample risk perceptionwas not found to be a significant predictor. The only cognitive variable that was pre-dictive of cancer worry in this sample was the identity subscale. Identifying more symp-toms of breast cancer was predictive of greater cancer worry in the comparison sample.

DISCUSSION

This study was theoretically driven to utilise the SRM as a framework to understandvariation in the psychological response to genetic risk of breast cancer. The studyfocused on one element of the model-illness perceptions, and aimed to examine associa-tions between illness perceptions and distress in women at increased risk of breastcancer, and a comparison sample with no experience of breast cancer in their socialenvironment.

TABLE VII Summary of results from multiple regression predicting Cancer worry scores

Predictor Increased risk sample Comparison sample

Step 1 Step 2 Step 1 Step 2

Age �0.198* �0.200* 0.120 0.112Identity 0.173 0.064 0.355** 0.324Time line acute/chronic 0.180 0.156 0.069 0.106Consequences 0.098 0.120 0.123 0.093Personal control �0.118 �0.026 0.079 0.083Treatment control 0.083 0.051 �0.05 �0.046Illness coherence 0.086 0.145 �0.097 �0.081Risk perception – 0.410** – 0.187

Adj R2¼ 0.103 Adj R2

¼ 0.257 Adj R2¼ 0.112 Adj R2

¼ 0.135F¼ 2.647* F¼ 5.328*** F¼ 2.493* F¼ 2.623*

*p<0.05; **p<0.001; ***p<0.001

TABLE VI Summary of results from multiple regression predicting general distress

Predictor Increased risk sample Comparison sample

Step 1 Step 2 Step 1 Step 2

Age �0.023 �0.022 0.248* 0.244*Identity 0.305** 0.275* �0.078 �0.092Time line acute/chronic 0.129 0.121 0.160 0.176Consequences 0.067 0.074 0.072 0.059Personal control �0.013 0.013 �0.011 �0.009Treatment control �0.007 �0.017 �0.094 �0.092Illness coherence �0.121 �0.103 �0.214* �0.207Risk perception – 0.114 – 0.082

Adj R2¼ 0.07 Adj R2

¼ 0.073 Adj R2¼ 0.08 Adj R2

¼ 0.074F¼ 2.067 F¼ 1.968 F¼ 2.027 F¼ 1.831

*p<0.05; **p<0.001; ***p<0.001

ILLNESS PERCEPTIONS AND BREAST CANCER 759

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

The results are in line with previous research which indicates that women attendingUK familial cancer clinics show levels of general distress comparable to women in thegeneral population (Cull et al., 1999; Cull, Fry, Rush and Steel, 2001). This may verifyprevious suggestions that global measures of distress do not capture the specific sourcesof distress in women at increased risk of breast cancer and that measures of cancer spe-cific related anxiety are more informative (Hopwood et al., 1998; Kent et al., 2000;Thewes, Meiser and Hickie, 2001). However, the clinical significance of levels ofcancer specific distress remains unknown and normative data are not yet available.There are concerns that measures of cancer worry reflect a realistic response to thesituation rather than morbid anxiety (Coyne, Benazon, Gaba, Calzone and Weber,2000; Hopwood, Shenton, Lalloo, Evans and Howell, 2001).

This is the first study to have attempted to measure and examine the impact ofillness perceptions in a sample at increased risk of cancer. Although a number ofsubscales of the IPQ-R showed good internal reliability, for some this was not ade-quate. One subscale in particular (timeline cyclical) was removed from the analysisdue to low internal reliability. This subscale was designed to assess perceptions ofconstancy in rapidly changing illnesses such as menstrual disorders (Moss-Morriset al., 2002). This concept may not be relevant to the timeframe of breast cancer.Beliefs regarding longer term issues such as recurrence are needed to capture this con-struct in this context. The identity, consequences and timeline acute/chronic subscalesalso showed only moderate internal consistency and may benefit from further refine-ment in this population. This may be best achieved by qualitative exploration of illnessperceptions in ‘at risk populations’. This would provide a greater insight into the extentto which illness perceptions are developed and the relative importance of dimensions inthis population as well as provide data with which to validate the use of the IPQ-R inthis population.

Few differences in illness perceptions were identified between the groups. The findingthat women at increased risk had a more coherent understanding of breast cancer andwere more likely to agree that medically proven risk factors were causes of the diseasethan the comparison sample, suggest that genetic counselling may be effective in help-ing women to consolidate their representations of breast cancer and correctly identifythe causes of the disease. However, there may be a number of factors that accountfor the differences between the samples, including difference in risk status, awarenessof the risk and experience of breast cancer in the family. Further research is neededto clarify these issues, for example a controlled prospective study examining illnessperceptions before and after genetic counselling.

Another limitation of the study that should be highlighted is that the sample wasrecruited from a single clinic and caution is needed to generalise findings. Both sampleswere predominately white, Caucasian and of high educational level as seen in otherstudies of screening populations (Kash et al., 1992; Codori, Hanson and Brandt,1994; Rimer, Schildkraut, Lerman, Lin and Audrain, 1996). Individuals with ahigher level of education are more likely to hold beliefs that are compatible with scien-tific and medical approaches (Bowling, 1989). The results of this study may not general-ise to women of lower educational level or ethnic groups whose beliefs about breastcancer are likely to be diverse (Klonoff and Landrine, 1994). The poor response ratein the comparison sample was also problematic. This sample may have been biasedtowards women particularly interested in breast cancer and health issues. It is also poss-ible that the samples were biased in terms of levels of distress. Women with high levels

760 G. REES et al.

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

of distress may avoid cues about their risk such as participating in research studies oreven attending familial breast cancer clinics (Cull et al., 2001).

Although limited differences between illness perceptions were identified betweenthe samples, what is interesting is that a number of illness perception subscales wereassociated with levels of distress in women at increased risk of breast cancer but notthe comparison sample. The pattern of correlations in the increased risk sample repli-cates those found in patient samples including breast cancer patients (Weinman et al.,1996; Buick, 1997; Heijmans and De Ridder, 1998; Heijmans, 1999; Murphy et al.,1999; Scharloo et al., 2000). This may reflect the personal relevance of breast cancerrepresentations when an individual is aware of their own genetic risk and indicatesthat the association between illness perceptions and distress may be moderated by anindividual’s risk perception. However, the multiple regression analysis did not findany of the illness perception variables to be strong predictors of distress. It is possiblethat inter-correlations between the subscales may have created problems of collinearityin the multiple regression. This problem could be reduced by further refinement ofmeasures in this population. Previous research using measures of illness perceptionsshowing high levels of reliability has found a range of cognitive perceptions to signifi-cantly predict emotional response to illness (e.g. Jopson and Moss-Morris, 2003). Theresults of this study provide tentative evidence that both illness perceptions and riskperception contribute independently to different aspects of psychological well-beingin individuals at risk of disease. Research investigating psychological response to riskwill benefit from further examination of the meaning of risk in terms of illness percep-tions and refinement of measures, as well as investigating the interrelations betweenrisk perception and illness perceptions.

The findings support the notion that identity is an important component of illnessrepresentations for healthy individuals (Bishop, Briede, Cavazos, Grotzinger andMcMahon, 1987). Bishop et al. (1987) proposed that lay conceptions of illness arebased primarily around symptoms in order to aid detection of disease. However, thesymptoms outlined in the identity subscale were mainly generic and raise the problemthat high scores on this and other measures may reflect a response bias. Although thisstudy was based on the premise that illness perceptions influence adjustment to risk, thecross sectional nature of this study and many others in the literature mean that anumber of alternative interpretations can be provided for the correlational patterns dis-covered. For example, negative affect may influence both levels of distress and illnessperceptions. Individuals showing greater negative affect may be more prone to develop-ing negative views of breast cancer. This is an area that needs to be examined. Causalrelations are difficult to establish, although a few longitudinal studies have suggestedthat cognitive representations precede illness outcomes (Petrie et al., 1996; Orbell,Johnston, Rowley, Espley and Davy, 1998; Schiaffino, Shawaryn and Blum, 1998;Scharloo et al., 2000). Further intervention studies influencing cognitions are neededto provide causal evidence (e.g. Petrie, Cameron, Ellis, Buick and Weinman, 2002).

It was surprising that perceptions of control of breast cancer were not associatedwith distress in the increased risk sample since perceptions of control are found to beone of the most important determinants of adaptation in patient samples (Scharlooand Kaptein, 1997). Negative associations between beliefs over control of breastcancer and distress have been shown in breast cancer patients (Taylor et al., 1984;Buick, 1997). In addition, Cull et al. (2001) identified that low internal locus of controlwas associated with case level distress in women attending a familial ovarian cancer

ILLNESS PERCEPTIONS AND BREAST CANCER 761

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

clinic. However, the control subscales of the IPQ-R reflect perceptions of control overthe prognosis and the recovery of disease. It is likely that different control perceptionsare more relevant to this population, such as beliefs about control of risk, prevention ofbreast cancer and efficacy of detection methods. Audrain et al. (1997) used a single itemto assess beliefs about control over developing breast cancer in women attendinggenetic counselling. Women who were categorised as having at least moderate perceivedcontrol over developing breast cancer, were found to show lower levels of general andcancer specific distress than those with no, or little, perceived control. The meaning ofcontrol in this population needs to be explored and examined in more detail.

In common with many studies guided by the SRM, this study focused on the cogni-tive representations component of the model. Further research is needed to investigatethe coping and appraisal elements and interactions between these constructs. Researchon coping in women at increased risk of breast cancer has tended to focus on the copingstyle rather than the coping strategies. Although qualitative studies have investigatedhow women cope with an increased risk of breast cancer, no specific measures are cur-rently available (Appleton et al., 2000). Further research is required to investigate andassess specific coping strategies in this population, as well as satisfaction and appraisalof these strategies. Given the need to explore in greater detail of the type of illness per-ceptions formed by women at increased risk of breast cancer, qualitative studies explor-ing illness perceptions, coping and adjustment to risk in women at increased risk ofbreast cancer are required.

This study was guided by the SRM because the model was felt to examine in detailthe meaning of risk for the individual. However, since the SRM was designed to pri-marily understand a patient’s response to illness, constructs reflecting perceived riskhave not been explicitly incorporated into the model. It would be useful to determinehow to incorporate beliefs about risk into the model. In order to achieve this, furtherresearch is required to examine the interactions between risk perception, illness percep-tions, coping, appraisal, and the psychological response to risk.

Acknowledgements

This research was funded by a studentship from Cancer Research UK.

References

Ajzen, I. and Fishbein, M. (1980). Understanding Attitudes and Predicting Social Behaviour. Prentice Hall, NJ.Appleton, S., Fry, A., Rees, G., Rush, R. and Cull, A. (2000). Psychosocial effects of living with an increased

risk of breast cancer: an exploratory study using telephone focus groups. Psycho-oncology, 9, 511–521.Audrain, J., Schwartz, M., Lerman, C., Hughes, C., Peshkin, B. and Biesecker, B. (1997). Psychological dis-

tress in women seeking genetic counseling for breast-ovarian cancer risk: the contributions of personalityand appraisal. Annals of Behavioral Medicine, 19, 370–377.

Becker, M.H. (1974). The health belief model and personal health behavior. Health Education Monographs,2, 324–508.

Bishop, G.D., Briede, C., Cavazos, L., Grotzinger, R. and McMahon, S. (1987). Processing illness informa-tion: the role of disease prototypes. Basic and Applied Social Psychology, 8, 21–43.

Bowling, A. (1989). Implications of preventative health behaviour for cervical and breast cancer screeningprogrammes – a review. Family Practice, 6, 224–231.

Brain, K., Norman, P., Gray, J. and Mansel, R. (1999). Anxiety and adherence to breast self-examination inwomen with a family history of breast cancer. Psychosomatic Medicine, 61, 181–187.

Buick, D.L. (1996). Psychological Adjustment to Radiation and Chemotherapy: The Role of Illness Perceptions.University of Auckland School of Medicine. Unpublished Doctoral Dissertation.

762 G. REES et al.

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

Buick, D.L. (1997). Illness representations and breast cancer: Coping with radiation and chemotherapy.In: Petrie, K.J. and Weinman, J.A. (Eds.), Perceptions of Health and Illness, pp. 379–409. HarwoodAcademic Publishers, Singapore.

Chalmers, K. and Thomson, K. (1996). Coming to terms with risk of breast cancer: perceptions of womenwith primary relatives with breast cancer. Qualitative Health Research, 6, 256–282.

Codori, A.M., Hanson, R. and Brandt, J. (1994). Self selection in predictive testing for Huntington’s disease.American Journal of Medical Genetics, 54, 167–173.

Collins, F.S. (1996). BRCA1 – lots of mutations, lots of dilemmas. The New England Journal of Medicine, 334,186–188.

Coyne, J.C., Benazon, N.R., Gaba, C.G., Calzone, K. and Weber, B.L. (2000). Distress and psychiatricmorbidity among women from high-risk breast and ovarian cancer families. Journal of Consulting andClinical Psychology, 68, 864–874.

Cull, A., Anderson, E.D.C., Campbell, S., Mackay, J., Smyth, E. and Steel, M. (1999). The impact of geneticcounselling about breast cancer risk on women’s risk perceptions and levels of distress. British Journalof Cancer, 79, 501–508.

Cull, A., Fry, A., Rush, R. and Steel, C.M. (2001). Cancer risk perceptions and distress among women attend-ing a familial ovarian cancer clinic. British Journal of Cancer, 84, 594–599.

Decruyenaere, M., Evers-Kiebooms, G., Welkenhuysen, M., Denayer, L. and Claes, E. (2000). Cognitiverepresentations of breast cancer. Emotional distress and preventive health behaviour: A theoreticalperspective. Psycho-oncology, 9, 528–536.

Emery, J., Murphy, M. and Lucassen, A. (2000). Hereditary cancer – the evidence for current recommendedmanagement. The Lancet Oncology, May preview, 9–16.

Evans, D.G.R., Fentiman, I.S., McPherson, K., Asbury, D., Ponder, B.A.J. and Howell, A. (1994). Familialbreast cancer. British Medical Journal, 308, 183–187.

Ford, D., Easton, D.F., Stratton, M., Narod, S., Goldgar, D., Devilee, P. et al. (1998). Genetic heterogeneityand penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. American Journalof Human Genetics, 62, 676–689.

Futreal, P.A., Liu, Q., Shattuck-Eidens, D., Cochran, C., Harshman, K., Tavtigian, S. et al. (1994). BRCA1mutations in primary breast and ovarian carcinomas. Science, 266, 120–122.

Gagnon, P., Massie, M.J., Kash, K.M., Gronert, M., Heerdt, A.S., Brown, K. et al. (1996). Perception ofbreast cancer risk and psychosocial distress in women attending a surveillance program. Psycho-oncology,5, 259–269.

Goldberg, D. and Williams, P. (1988). A Users Guide to the General Health Questionnaire. Nfer-Nelson Ltd.,Berkshire.

Hagger, M.S. and Orbell, S. (2003). A meta-analytic review of the common-sense model of illness representa-tions. Psychology & Health, 18, 141–184.

Hampson, S., Glasgow, R. and Toobert, D. (1990). Personal models of diabetes and their relations to self-careactivities. Health Psychology, 9, 632–646.

Hampson, S.E., Glasgow, R.E. and Zeiss, A.M. (1994). Personal models of osteoarthritis and their relationto self management activities and quality of life. Journal of Behavioral Medicine, 17, 143–158.

Heijmans, M. (1999). The role of patients’ illness representations in coping and functioning with Addison’sdisease. British Journal of Health Psychology, 4, 137–150.

Heijmans, M. and de Ridder, D. (1998). Assessing illness representations of chronic illness: explorations oftheir disease-specific nature. Journal of Behavioral Medicine, 21, 487–503.

Hopwood, P., Keeling, F., Long, A., Pool, C., Evans, G. and Howell, A. (1998). Psychological supportneeds for women at high genetic risk of breast cancer: some preliminary indicators. Psycho-oncology,7, 402–412.

Hopwood, P., Shenton, A., Lalloo, F., Evans, D.G.R. and Howell, A. (2001). Risk perceptions and cancerworry: an exploratory study of the impact of genetic risk counselling in women with a family historyof cancer. Journal of Medical Genetics, 38, 139–142.

Jessop, D.C. and Rutter, D.R. (2003). Adherence to asthma medication: the role of illness representations.Psychology & Health, 18, 595–612.

Jopson, N.M. and Moss-Morris, R. (2003). The role of illness severity and illness representations in adjustingto multiple sclerosis. Journal of Psychosomatic Research, 54, 503–511.

Kash, K.M., Holland, J.C., Halper, M.S. and Miller, D.G. (1992). Psychological distress and surveillancebehaviours of women with a family history of breast cancer. Journal of the National Cancer Institute,84, 24–30.

Kent, G., Howie, H., Fletcher, M., Newbury-Ecob, R. and Hosie, K. (2000). The relationship between per-ceived risk, thought intrusiveness and emotional well-being in women receiving counselling for breastcancer risk in a family history clinic. British Journal of Health Psychology, 5, 15–26.

Klonoff, E.A. and Landrine, H. (1994). Culture and gender diversity in commonsense beliefs about the causesof six illnesses. Journal of Behavioral Medicine, 17, 407–418.

Lerman, C., Trock, B. and Rimer, B.K. (1991a). Psychological side effects of breast cancer screening. HealthPsychology, 10, 259–267.

ILLNESS PERCEPTIONS AND BREAST CANCER 763

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

Lerman, C., Trock, B., Rimer, B.K., Boyce, A., Jepson, C. and Engstrom, P.F. (1991b). Psychological andbehavioural implications of abnormal mammograms. Annals of Internal Medicine, 114, 657–661.

Lerman, C. and Schwartz, M. (1993). Adherence and psychological adjustment among women at high risk forbreast cancer. Breast Cancer Research and Treatment, 28, 144–155.

Lerman, C., Kash, K. and Stefanek, M. (1994). Younger women at increased risk for breast cancer: perceivedrisk, psychological well being and surveillance behavior. Monographs of the National Cancer Institute,16, 171–176.

Lerman, C., Narod, S., Schulman, K., Hughes, C., Gomez-Caminero, A., Bonney, G. et al. (1996). BRAC1testing in families with hereditary breast-ovarian cancer: prospective study of patient decision makingand outcomes. Journal of the American Medical Association, 275, 1885–1892.

Leventhal, H., Easterling, D.V., Coons, H.L., Luchterhand, C.M. and Love, R.R. (1986). Adaptation tochemotherapy treatments. In: Anderson, B.L. (Ed.), Women with Cancer – Psychological Perspectives,pp. 172–203. Springer-Verlag, New York.

Leventhal, H., Benyamini, Y., Brownlee, S., Diefenbach, M., Leventhal, E., Patrick-Miller, L. et al. (1997).Illness representations: theoretical foundations. In: Petrie, K.J. and Weinman, J.A. (Eds.), Perceptionsof Health and Illness, pp. 19–46. Harwood Academic Publishers, Singapore.

Llewellyn, C.D., Miners, A.H., Lee, C.A., Harrington, C. and Weinman, J. (2003). The illness perceptions andtreatment beliefs of individuals with severe haemophilia and their role in adherence to home treatment.Psychology & Health, 18, 185–200.

Lloyd, S., Watson, M., Waites, B., Meyer, L., Eeles, R., Ebbs, S., et al. (1996). Familial breast cancer: a con-trolled study of risk perception, psychological morbidity and health beliefs in women attending forgenetic counselling. British Journal of Cancer, 74, 482–487.

Macmillan, R.D. (2000). Screening women with a family history of breast cancer – results from the BritishFamilial Breast Cancer Group. European Journal of Surgical Oncology, 26, 149–152.

Meijers-Heijboer, H., van Geel, B., van Putten, W.L.J., Henzen-Logmans, S.C., Seynaeve, C., Menke-Pluymers, M.B.E. et al. (2001). Breast cancer after prophylactic bilateral mastectomy in women witha BRCA1 or BRCA2 mutation. New England Journal of Medicine, 345, 159–164.

Mettlin, C. (1994). The relationship of breast cancer epidemiology to screening recommendations. Cancer,74, 228–230.

Miki, Y., Swenson, J., Shattucj-Eidens, D., Futreal, P.A., Harshman, K., Tavtigian, S., Liu, Q. et al.(1994). A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science, 266,66–71.

Moss-Morris, R., Weinman, J., Petrie, K.J., Horne, R., Cameron, L.D. and Buick, D. (2002). The revisedillness perception questionnaire (IPQ-R). Psychology & Health, 17, 1–16.

Murphy, H., Dickens, C., Creed, F. and Bernstein, R. (1999). Depression, illness perception and coping inrheumatoid arthritis. Journal of Psychosomatic Research, 46, 155–164.

Orbell, S., Johnston, M., Rowley, D., Espley, A. and Davey, P. (1998). Cognitive representations ofillness and functional and affective adjustment following surgery for osteoarthritis. Social Science andMedicine, 47, 93–102.

Petrie, K.J., Weinman, J., Sharpe, N. and Buckley, J. (1996). Role of patients’ view of their illness in predict-ing return to work and functioning after myocardial infarction: longitudinal study. British MedicalJournal, 312, 1191–1194.

Petrie, K.J., Cameron, L.D., Ellis, C.J., Buick, D. and Weinman, J. (2002). Changing illness perceptions fol-lowing myocardial infarction: an early intervention randomized controlled trial. Psychosomatic Medicine,64, 580–586.

Rees, G. (2003). Understanding distress in women at increased risk of breast cancer. The effects of experienceof breast cancer in the family and illness perceptions. University of Edinburgh, Unpublished DoctoralDissertation.

Rees, G., Fry, A. and Cull, A. (2001). A family history of breast cancer. Women’s experiences from atheoretical perspective. Social Science and Medicine, 52, 1433–1440.

Richards, M. (1999). Genetic counselling for those with a family history of breast or ovarian cancer. ActaOncologica, 38, 559–565.

Rimer, B.K., Schildkraut, J.M., Lerman, C., Lin, T.H. and Audrain, J. (1996). Participation in a women’sbreast cancer risk counseling trial: Who participates? Who declines? Cancer, 77, 2348–2355.

Rosenstock, I.M. (1966). Why people use health services. Milband Memorial Fund Quarterly, 44, 94–127.Scharloo, M. and Kaptein, A. (1997). Measurement of illness perceptions in patients with chronic somatic ill-

nesses: a review. In: Petrie, K.J. and Weinman, J.A. (Eds.), Perceptions of Health and Illness, pp. 103–154,Harwood Academic Publishers, Singapore.

Scharloo, M., Kaptein, A.A., Weinman, J., Hazes, J.M., Willems, L.N.A., Bergman, W. and Rooijmans,H.G.M. (1998). Illness perceptions, coping and functioning in patients with rheumatoid arthritis, chronicobstructive pulmonary disease and psoriasis. Journal of Psychosomatic Research, 44, 573–585.

Scharloo, M., Kaptein, A.A., Weinman, J., Bergman, W., Vermeer, B.J. and Rooijmans, H.G.M. (2000).Patients’ illness perceptions and coping as predictors of functional status in psoriasis: a 1-year follow-up. British Journal of Dermatology, 142, 899–907.

764 G. REES et al.

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14

Schiaffino, K.M., Shawaryn, M.A. and Blum, D. (1998). Examining the impact of illness representations onpsychological adjustment to chronic illness. Health Psychology, 17, 262–268.

Scottish Executive (2001). Cancer Genetics Services in Scotland: Guidance to Support the Implementationof Genetics Services for Breast, Ovarian and Colorectal Cancer Predisposition. Scottish Executive,Edinburgh.

Scottish Intercollegiate Guidelines Network [SIGN] (1998). Breast Cancer in Women. A National ClinicalGuideline. SIGN Publication No. 29. Edinburgh, NHS Scotland.

Struewing, J.P., Hartge, P., Wacholder, S., Baker, S.M., Berlin, M., McAdams, M. et al. (1997). The risk ofcancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. The NewEngland Journal of Medicine, 336, 1401–1408.

Taylor, S.E., Lichtman, R.R. and Wood, J.V. (1984). Attributions, beliefs about control and adjustment tocancer. Journal of Personality and Social Psychology, 46, 489–502.

Thewes, B., Meiser, B. and Hickie, I.B. (2001). Psychometric properties of the Impact of Event scale amongstwomen at increased risk of hereditary breast cancer. Psycho-oncology, 10, 459–468.

Tilanus-Linthorst, M., Bartels, C., Obdeijn, A. and Oudkerk, M. (2000). Earlier detection of breast cancer bysurveillance of women at familial risk. European Journal of Cancer, 36, 514–519.

Vogel, V. (2000). Breast cancer prevention: a review of current evidence. CA–A Cancer Journal for Clinicians,50, 156–170.

Watson, M., Lloyd, S., Davidson, J., Meyer, L., Eeles, R., Ebbs, S. et al. (1999). The impact of genetic coun-selling on risk perception and mental health in women with a family history of breast cancer. BritishJournal of Cancer, 79, 868–874.

Weinman, J., Petrie, K.J., Moss-Morris, R. and Horne, R. (1996). The illness perception questionnaire: a newmethod for assessing the cognitive representation of illness. Psychology & Health, 11, 431–445.

Wellisch, D.K., Gritz, E., Schain, W., Wang, H.J. and Siau, J. (1991). Psychological functioning of daughtersof breast cancer patients: part I daughters and comparison subjects. Psychosomatics, 32, 324–336.

Wellisch, D.K., Gritz, E.R., Schain, W., Wang, H.J. and Siau, J. (1992). Psychological functioning of daugh-ters of breast cancer patients: part II characterizing the distressed daughter of the breast cancer patient.Psychosomatics, 33, 171–179.

Wellisch, D.K., Schains, W., Gritz, E.R. and Wang, H.J. (1996). Psychological functioning of daughters ofbreast cancer patients: part III expereinces and perceptions of daughters related to mother’s breastcancer. Psycho-oncology, 5, 271–281.

Wooster, R., Bignell, G., Lancaster, J., Swift, S., Seal, S., Mangion, J. et al. (1995). Identification of the breastcancer susceptibility gene BRCA2. Nature, 378, 789–792.

Zakowski, S., Valdimarsdottir, H., Bovbjerg, D., Borgen, P., Holland, J. and Kash, K. (1997). Predictors ofintrusive thoughts and avoidance in women with family histories of breast cancer. Journal of BehavioralMedicine, 19, 362–369.

ILLNESS PERCEPTIONS AND BREAST CANCER 765

Dow

nloa

ded

by [

Ban

gor

Uni

vers

ity]

at 0

7:59

20

Dec

embe

r 20

14