Primary care patients' personal illness models for ... - Oxford Journals

Family Practice © Oxford University Press 2001

Vol. 18, No. 3 Printed in Great Britain

Primary care patients’ personal illness models for depression: a preliminary investigation Charlotte Brown, Jacqueline Dunbar-Jacoba, Deena R Palenchar, Kelly J Kelleher, Richard D Bruehlmanb, Susan Sereikaa and Michael E Thase Brown C, Dunbar-Jacob J, Palenchar DR, Kelleher KJ, Bruehlman RD, Sereika S and Thase ME. Primary care patients’ personal illness models for depression: a preliminary investigation. Family Practice 2001; 18: 314–320. Background. Despite the fact that more than half of depressed persons are treated for this disorder by primary care physicians, depression is often under-recognized or treated inadequately. There is continued emphasis on effective treatment of depression in primary care patients, but little attention has been paid to the role of the depressed person’s illness cognitions in coping with this disorder. Given the often recurring and chronic nature of depression, the individual’s self-management strategies may be critical to effective treatment, recovery and remaining well. Objectives. The purpose of this pilot study was to determine whether primary care patients’ personal illness cognitions for depression are associated with depression coping strategies and treatment-related behaviour. Methods. Forty-one primary care patients with depressive symptoms or disorder completed interviews and questionnaires assessing illness cognitions for depression, depression coping strategies and other treatment-related behaviour. Descriptive statistics are used to present patients’ illness cognitions for depression. t-tests and correlational analyses were completed to assess the relationship between illness cognitions, depression coping strategies and treatmentrelated behaviour. Results. Preliminary data describing illness cognitions for depression are presented. Participants’ illness cognitions for depression were significantly associated with current and past treatmentseeking behaviour, medication adherence and coping strategies. Conclusions. Although preliminary, these findings indicate that patients’ understanding of depression and its consequences are associated with how they manage this illness. Future research is needed to examine the mediating and moderating effects of illness cognitions for depression on medication adherence and other self-management behaviours of depressed primary care patients. Knowledge about primary care patients’ personal illness models will aid in the development of adherence interventions, self-management training and support services appropriate to patients’ needs in the primary care setting. Keywords. Depression, illness cognitions, primary care patients.

Introduction

Received 22 May 2000; Revised 3 October 2000; Accepted 3 January 2001. Western Psychiatric Institute and Clinic, bDepartment of Family Medicine and Clinical Epidemiology, University of Pittsburgh School of Medicine and aDepartment of Health and Community Systems, University of Pittsburgh School of Nursing, Pittsburgh, PA, USA. Correspondence to Charlotte Brown, Western Psychiatric Institute and Clinic, University of Pittsburgh School of Medicine, 3811 O’Hara Street, Pittsburgh, PA 15213, USA.

Despite the availability of effective treatments for depression, caring for depressed primary care patients and reducing the personal and societal costs of depression remain a public health priority. Overall, treatment costs for depression (in excess of $12 billion in 1990),1 its high prevalence and greater opportunity for detection of depression in the primary care service 314

Personal illness models for depression

sector,2 and associated functional impairment3 have led to the continued focus on delivery of effective depression treatment for primary medical care patients. Like other chronic diseases, management of depression requires considerable effort at self-management by persons with depression. Self-management refers to (i) engaging in activities that promote health and prevent adverse outcomes; (ii) interacting with health care providers and adhering to recommended treatment regimens; (iii) monitoring physical and emotional status and making appropriate illness management decisions on the basis of this self-monitoring; and (iv) managing the effects of illness on the ability to function in important roles and on emotions, self-esteem and relationships with others.4 Depressed primary care patients who take their medication as prescribed, participate in depression education activities and attend scheduled sessions with behavioural specialists have fewer symptoms and improved functional status compared with those who do not. Such findings have led to recommendations for patient activation interventions as a central component of depression treatment.5 Unfortunately, little attention has been given to why some patients readily engage in depression self-management while others do not. One possibility is that some individuals with depression do not believe that their symptoms and functional impairment are related to depression or that they can be improved with treatment. Depression frequently accompanies chronic physical illness,2 and this fact may contribute to the patient’s misinterpretation of depressive symptoms. Further, even when recognized, patients may not understand the seriousness of the disorder, its clinical course, impact on functioning and amenability to treatment. Determining how individuals understand and define depression and its consequences as a health problem is likely to be critical in understanding how they manage this illness. In fact, Von Korff and colleagues4 maintain that medical care for chronic conditions is rarely effective in the absence of adequate self-management. Understanding patients’ illness cognitions or models may be important for studying help-seeking, coping and treatment adherence for chronic disorders, especially in depression. Further, knowledge about primary care patients’ model of their illness can help us to develop adherence interventions, self-management training and support services appropriate to depressed patients’ needs in the primary care setting. Leventhal’s6 self-regulatory model of illness cognitions has been employed successfully to understand better patient behaviour in chronic medical disorders. According to this model, illness cognitions have five distinct components: • identity—the label the person uses to describe the illness and the symptoms the patient views as being part of the disorder;

315

• cause—personal ideas about the cause of the illness; • timeline—how long the patient believes the illness will last, i.e. is it acute, chronic or cyclic?; • consequences—expected effects and outcome of the illness; • perceived controllability—the responsiveness of the symptom or illness to self- and/or professional intervention. Research on the personal illness models of patients with chronic medical illnesses such as hypertension,7 diabetes,8,9 heart disease,10 arthritis11 and chronic fatigue syndrome12 have yielded promising findings, suggesting that how patients understand their illness has a significant impact on illness coping strategies, medication adherence and treatment referral adherence. However, the influence of personal illness models on such outcomes has not been examined in depressive disorders. One of the challenges of examining illness cognitions for depression is that depression itself is characterized by changes in cognitive, somatic and affective symptoms. Of particular significance is the fact that cognitive distortions are evident in depression, and this may be a confounding factor. Thus, it may be the severity of the depressive symptoms that is the primary influence on how patients conceptualize depression and how they manage it. We believe that personal illness models may show promise for better understanding depressed primary care patients’ treatment-related behaviours and present preliminary data on this. We describe patients’ conceptualization of their depressive symptoms according to Leventhal’s model, and the association between illness cognitions for depression, self-reported depression coping strategies and treatment-related behaviours. It is hypothesized that illness cognitions will be significantly associated with self-reported coping strategies independent of depressive severity. We discuss the utility of assessing patients’ understanding of depression in order to determine how personal illness models impact treatment-seeking behaviour, treatment adherence, and how knowledge of patients’ models can aid primary care physicians in collaborative management of chronic illnesses such as depression.

Method Procedure and participants Participants for this study were recruited from two sources, patients presenting for appointments in two general internal medicine clinics and patients participating in an on-going observational study of the course of minor depression in primary care patients. Primary care patients presenting for appointments in two urban general internal medicine clinics were approached by study research associates and asked (by clinic receptionists) to

316

Family Practice—an international journal

complete the Primary Care Evaluation of Mental Disorders Patient Questionnaire (PRIME-MD PQ13). Patients were also asked for permission to be approached by study research associates. Those who reported being ‘bothered a lot’ by a symptom of depression (loss of interest or depressed mood) in the past month were administered the mood module of the PRIME-MD Clinician Evaluation Guide (CEG).13 Of the 271 patients approached, 98% (n = 266) completed the PQ, and 18% (n = 47) endorsed loss of interest or depressed mood in the past month and were administered the PRIME-MD CEG. Patients were deemed ineligible if they had a terminal illness, neurological disorder, cognitive impairment, substance use disorder, bipolar illness or psychotic disorder. Of the 29 eligible individuals, 21 (72%) completed an interview on their symptom management strategies. Twenty participants were recruited from an on-going study of minor depression and completed the PRIME-MD PQ and CEG and an interview on symptom management. These participants were identified originally using an identical recruitment strategy. All participants (n = 41) completed self-report assessments of depressive symptoms, mental and physical health-related functioning, depression coping strategies, medication adherence, questions pertaining to personal demographics and past mental health visits, and were administered a modified version of the Illness Perception Questionnaire (IPQ).14 Measures Demographic characteristics. Demographic variables included age, gender, self-reported race (African American, White or other), employment status (employed full or part time versus unemployed), marital status (married versus unmarried) and education (>high school education versus ,high school education). Clinical characteristics. All patients who endorsed a symptom of depression (i.e. being bothered a lot by loss of interest or depressed mood in the past month) were administered the mood module of the PRIME-MD CEG, a semi-structured interview schedule designed to assess mood, anxiety, somatoform, eating and alcohol use disorders in the primary care setting. The mood module provides DSM-IV diagnoses for major and minor depression and dysthymia. Severity of depressive symptoms was assessed with the 21-item Beck Depression Inventory (BDI).15 Self-reported functional disability was assessed with the Medical Outcomes Study 36-item Short-Form Health Survey. Summary scores for physical and mental health-related functioning were computed using procedures developed by Ware et al.16 Depression coping strategies. Behaviours used to cope with depressive symptoms were assessed with an adapted version of the 28-item brief COPE. This scale is

a brief version of the COPE inventory which has been used extensively in health-related research. The scale has 14 conceptually distinct subscales (e.g. active coping, planning and acceptance) with two items per scale.17 We adapted instructions to the brief COPE by asking respondents to indicate how they coped with depressive symptoms in the past month. Additionally, several items were modified to refer to symptoms rather than situations (e.g. “I refuse to believe that it has happened” was changed to “I refuse to believe that I have these symptoms;” “I’ve been giving up trying to deal with it” was changed to “I’ve been giving up trying to deal with these symptoms”. One item adapted from Cameron et al.18 assessed perceived effectiveness of coping efforts, i.e. “On the whole, how effective have these coping strategies been in dealing with your depressive symptoms?”, rated on a 5-point Likert scale, with responses ranging from ‘strongly agree’ to ‘strongly disagree’. Current and past treatment for depression. Participants were asked if they had ever been treated for depression, and whether they were currently taking antidepressant medication. Medication adherence. Medication-taking behaviour was assessed with the 4-item Morisky Medication Adherence Scale.19 Items assess the degree to which patients self-report non-adherence with prescribed medication due to forgetting, carelessness, stopping the drug when feeling better or stopping the drug when feeling worse. Personal illness models for depression. A modified version of the IPQ14 was administered to all participants. This instrument is based on Leventhal et al.’s20 selfregulatory model of illness behaviour, and was designed to assess illness cognition in the context of chronic illness. The authors note that the scale can be adapted for use in a variety of illnesses. The five scales assess illness identity, cause, timeline, consequences and perceived controllability dimensions that underlie a patient’s representations of illness. Scale items are scored on a 4-point (all of the time to never) or a 5-point Likert scale (strongly agree to strongly disagree). Several components of the original scale were adapted for the assessment of illness cognitions in a depressed sample. Eleven items were added to the illness identity scale which assesses the frequency of symptoms endorsed as part of the illness. These items were included because they are commonly reported by depressed patients and are included in DSM-IV criteria for major depressive disorder or dysthymia.21 They included symptoms such as: little interest or pleasure in doing things; feeling sad or depressed; thoughts of death or suicide; trouble concentrating or making decisions; etc. The cause subscale assessed respondents’ perceived causes of depressive symptoms. Six items were added to the original scale to

317


assess psychosocial factors, interpersonal loss and medical factors as perceived causes of depressive symptoms. Four items assessed perceived duration of symptoms. One item was added to assess expected recurrence of depressive symptoms, i.e. “my symptoms will come and go”. Perceived consequences of depressive symptoms was assessed with seven items. Perceived controllability of depressive symptoms was assessed with seven items. One item was added to assess religious faith as a means of managing depressive symptoms. Statistical analyses Complete data (n = 41) were available for the IPQ, and questions assessing coping effectiveness, past mental health treatment and current antidepressant treatment. One participant did not complete the MOS SF-36 and the BDI, while two individuals did not complete the COPE. Descriptive statistics (i.e. means and proportions) were used to describe the study group’s demographic and clinical characteristics. Mean BDI, MOSPCS, MOSMCS and IPQ subscale (identity, timeline, consequences and controllability) scores approximated the normal distribution. Frequency distributions were used to examine commonly held illness cognitions for depressive symptoms and perceived effectiveness of coping strategies. The association between illness cognitions for depression (identity, causes, timeline, consequences and perceived controllability), depression coping strategies and medication adherence were assessed using partial correlation coefficients (controlling for depressive severity). The relationship between current and past mental health treatment and illness cognitions was assessed with t-tests.

Results Demographic and clinical characteristics of the sample As shown in Table 1, this sample of 41 primary care patients had a mean age of 43 years, was predominantly female, unemployed, unmarried and had at least a high school education. The proportions of African American and White participants were comparable. The sample was comprised of patients diagnosed with major depression, minor depression, dysthymia and those with depressive symptoms but no depressive disorder. Overall, patients reported mild to moderate symptoms of depression, and moderate levels of impairment in physical and psychosocial functioning. Approximately one-third were currently using antidepressant medication and about half had received prior treatment for depression. Description of illness cognitions for depressive symptoms Identity. All patients reported experiencing anhedonia or depressed mood and at least one other DSM-IV

TABLE 1

Sample characteristics n

Demographics Mean (SD) age % Female % White % Employed full/part time % Married % >High school

43 (15.7) 66 46 44 20 88

41 41 41 41 41 41

Clinical characteristics Mean (SD) BDI Mean (SD) physical component score Mean (SD) mental component score

17.5 (8.3) 39.5 (10.2) 34.7 (11.0)

40 40 40

% Major depression % Dysthymia % Minor depression % Depressive symptoms only

44 5 15 37

41 41 41 41

% Currently taking antidepressants % Previous mental health treatment

29 51

41 41

depression symptom at least occasionally in the past month, while 56% (n = 23) reported having these symptoms frequently in the past month. Other types of co-occurring symptoms were quite common. Thus, 54% (n = 22) of patients also reported frequent gastrointestinal symptoms (nausea, gas or constipation), 66% (n = 27) had frequent pain (general aches and pains, headaches, stomach pain or chest pain), 51% (n = 21) reported frequent irritability or anxiety and 27% (n = 11) had frequent symptoms of autonomic arousal (i.e. dizziness, fainting or racing heart) in the preceding month. Cause. Rankings of the five most commonly reported causes of depressive symptoms are listed in Table 2. Stressors of varied types and heredity were the most commonly endorsed causes of depressive symptoms, while factors related to medical illness were endorsed by

TABLE 2

Rankings of primary care patients’ perceived causes of depressive symptoms

Rank

Perceived cause

n

%

1

Stress

28

68

2

Heredity—runs in the family

17

41

3

Not taking care of physical health

16

39

3

Marriage or relationship problems

16

39

4

Interpersonal difficulties

14

34

5

Medical illness

13

32

5

Reaction to medical illness

13

32

318


nearly one-third of patients. There was also considerable overlap in the number of causes endorsed, with most respondents (63%, n = 26) endorsing at least two causes of depressive symptoms. Timeline. Four questions assessed patient expectations for depressive symptom duration. Patients were asked if they thought symptoms would ‘last a short time’ (acute), ‘last a long time or permanent’ (chronic) or ‘come and go’ (intermittent). The majority of patients (63%, n = 26) characterized depressive symptoms as fluctuating or intermittent. As shown in Table 3, this aspect of depressive symptoms was prominent even when symptoms were also described as either acute or chronic in nature. Further, almost half of the sample (49%, n = 20) described depressive symptoms as chronic, while only 24% (n = 10) of patients described symptoms as acute. Consequences. Most (68%, n = 28) participants viewed depression as having significant negative consequences, and only 7% (n = 3) viewed depression as having only minimal consequences on their life. Nevertheless, only 39% (n = 16) endorsed depression as a serious condition. Many respondents viewed depressive symptoms as having significant financial consequences (39%, n = 16), and a significant impact on how they viewed themselves (68%, n = 28) and on how others viewed them (41%, n = 17). Interestingly, despite the perceived negative impact of these symptoms, a substantial minority reported that depressive symptoms had become easier to live with (39%, n = 16). Perceived controllability. Most patients (63%, n = 26) thought depressive symptoms could be controlled and that such symptoms would improve in time (59%, n = 24), while 15% (n = 6) of patients viewed depressive symptoms as uncontrollable. However, only 39% (n = 16) of participants thought treatment would be effective in improving symptoms. In contrast, 63% (n = 26) thought that religious coping strategies (e.g. faith in God) would be effective in improving symptoms.

TABLE 3

Expected duration of depressive symptoms

Acute and intermittent Chronic and intermittent

n

%

6

14

13

32

Intermittent only

7

17

Acute only

4

10

Chronic only

7

17

4

10

None of the a

abovea

Four patients did not endorse any of the above descriptions of their depressive symptoms.

Association of illness cognitions with depression coping strategies As expected, patient perception of depressive symptoms was also related to strategies used to manage them. Low to moderate partial correlations (r = 0.32 to r = 0.58) indicated that several coping strategies remained significantly associated with patients’ illness cognitions independent of depressive severity. These coping behaviours included active coping, planning, religious coping, self-blame, self-distraction, emotional venting and positive reframing. Partial correlations indicated that perceived negative consequences for depression were associated with more active coping (r = 0.36, P = 0.03), religious coping (r = 0.39, P = 0.02) and self-blame (r = 0.43, P = 0.008), while less perceived controllability was associated with more religious coping (r = 0.38, P = 0.02). Perception of depressive symptoms as chronic was associated with less planning (r = –0.40, P = 0.02), and endorsement of more depressive symptoms was associated with more self-blame (r = 0.58, P = 0.0001), more self-distraction (r = 0.35, P = 0.03) and more emotional venting (r = 0.38, P = 0.02). Perceived effectiveness of coping strategies Few respondents (17%, n = 7) described their coping behaviour as ‘quite a bit’ or ‘extremely’ effective. Approximately half (51%, n = 21) of participants described their coping behaviour as moderately effective in managing depressive symptoms, while an additional 27% (n = 11) described coping efforts as only ‘a little’ effective, and only 5% (n = 2) viewed coping as ‘not at all’ effective. Association of illness cognitions with mental health treatment and current medication adherence Patients’ illness cognitions (i.e. timeline, consequences and cause) were also associated with other illness management behaviours such as prior mental health treatment, current antidepressant treatment and medication adherence. Patients who had received prior mental health treatment (n = 20) perceived depressive symptoms as being more chronic in nature (t = 2.39, df = 38, P = 0.02) and as having a greater negative impact on functioning (t = 2.42, df = 38, P = 0.02) than patients with no prior mental health treatment. Compared with those not currently taking antidepressants, patients currently taking antidepressants were also more likely to view depressive symptoms as chronic (t = –3.5, df = 39, P = 0.001). Of the participants currently prescribed antidepressants (n = 12), five (42%) endorsed no problems with adherence. Among those (58%) who endorsed at least one problem with adherence (i.e. forgetting to take medicine, careless about taking medicine, stopping medicine when feeling better or stopping medicine if feeling worse after taking it), partial correlations, controlling for severity of depression, indicated that poorer adherence was associated with endorsement of


interpersonal difficulties as a cause of depressive symptoms (r = 0.59, P = 0.05).

Discussion This preliminary investigation describes primary care patients’ understanding of their depressive symptoms (i.e. identity, perceived cause, consequences, time course and controllability) and examines its relationship to illness behaviour. While personal illness models have been examined in chronic disorders such as diabetes, heart disease, arthritis, etc., patients’ illness cognitions and their relationship to illness behaviour have not been studied in depression. Consistent with prior research in chronic physical disorders, we found illness cognitions for depression associated with current and past treatment-seeking behaviour, medication adherence and coping strategies. Before discussing our findings in more detail, the limitations of these preliminary findings should be noted. We describe preliminary findings of a small (n = 41) nonrepresentative sample of primary care patients. This pilot study was designed to examine the utility of assessing various aspects of patients personal illness models for depression to better understand the nature of patients’ self-management strategies. Our findings should therefore be considered hypothesis generating rather than conclusive. This sample was also largely untreated for current depression. Only 29% (n = 12) of patients studied were currently using an antidepressant, and 51% (n = 21) had received prior treatment for depression. Participation in psychotherapy was not assessed. Thus, results can only be generalized to other samples with similar treatment profiles. Participant’s models of depressive symptoms indicate that these symptoms are described as frequently cooccurring with physical symptoms such as pain, gastrointestinal distress and to a lesser extent with autonomic arousal symptoms. Depressive symptoms are viewed most commonly as caused by stress or heredity and are described as fluctuating over time and frequently following a chronic course. Although most patients acknowledged depression’s significant negative impact on their lives, many thought symptoms could be controlled and would improve over time. However, most patients described their own coping efforts as only ‘moderately’ or ‘a little’ effective, and ,40% viewed depression treatment as effective in improving symptoms. We also examined the relationship between patient’s perceptions of depressive symptoms and illness behaviours such as prior and current mental health treatment, medication adherence and illness coping strategies. Consistent with previous findings, we found illness cognitions for depression significantly correlated with coping strategies,12 even when the level of depression was controlled. Thus, the degree to which patients used

319

active coping strategies, planning and religious coping was associated with their perception of the controllability, chronicity and negative consequences of depressive symptoms, while the number of depressive symptoms experienced was associated with more coping strategies such as self-distraction, emotional venting and selfblame. In understanding the relationship between patients’ perception of their symptoms and behaviour, it is noteworthy that only 29% of the sample currently were treated pharmacologically, although 51% had had prior treatment for depression. This is not surprising given that study participants generally viewed depression as often due to stress, and not as a serious condition or as amenable to treatment. Patient’s views about the fluctuating time course of depressive symptoms, and their controllability, may have led some to believe that their self-management strategies were the most appropriate (if not always the most effective) means of handling depression. In fact, many reported that depressive symptoms had become easier to cope with over time. Further evidence of the importance of patients’ beliefs is found among those treated with antidepressants. Poorer medication adherence was associated with the belief that depressive symptoms were caused by relationship problems. Although these patients sought treatment, their beliefs about the causes of depression were associated with how they managed the treatment regimen. Hunt et al.22 described patients’ understanding of the cause of their illness as complex and long lasting. They found that while many patients often modified their ‘illness explanations’ over a 4-month period after a physician consultation, most reverted back to their original belief, often with biomedical embellishments. They concluded that this reflected patients’ efforts to link illness and treatment in a meaningful way within the context in which it is experienced. Research has shown that collaborative treatment approaches are effective in improving adherence and treating depressed primary care patients.23 Collaborative management begins when both physician and patient have a shared understanding of the presenting problem.4 Our preliminary findings suggest that patients’ views about what causes depression, the disorder’s amenability to treatment, perceived consequences of depression and perceived duration of symptoms are all associated with various aspects of self-management behaviour. Knowledge about a patient’s personal illness model can provide initial guidance regarding those aspects of depression that the physician should inquire about. This would provide the opportunity to clarify misconceptions about the nature of depression (e.g. cause, treatability and duration), and identify possible impediments to treatment adherence. Knowledge about a patient’s usual strategies for coping with depressive symptoms can provide valuable information that can help guide physicians in tailoring treatment regimens to the patient’s needs.

320


Eliciting this type of information can facilitate collaborative management of depression by assisting patients and physicians in defining a problem that both agree is important, establishing a mutually agreed upon behavioural goal and developing an action plan for achieving the goal. These steps have been found beneficial in managing depression as well as other chronic illnesses.5,24–26 While the small sample size does not permit the use of multivariate statistics, these preliminary findings are promising and indicate the need for future research to examine the mediating and moderating effects of cognition, affect and coping strategies on the illness management behaviours of depressed primary care patients. Elucidation of these relationships will enable us to identify factors associated with effective and ineffective depression management strategies. Depression intervention strategies can then be targeted toward improving ineffective self-management strategies. Knowledge of those aspects of patients’ illness beliefs which affect medication adherence and other self-management behaviours will also aid primary care physicians as they work toward developing more effective collaborative management strategies with depressed patients.

5

6

7

8

9

10

11

12

13

14

15

Acknowledgements 16

The authors would like to thank Denise Paris for assistance with participant interviews. This work was supported by grant K20-MH 01328 (to CB) from the National Institute of Mental Health.

17

References

19

1

2

3

4

Greenberg PE, Stiglin LE, Finkelstein SN, Berndt ER. The economic burden of depression. J Clin Psychiatry 1993; 54: 405–418. AHCPR Depression Guideline Panel. Depression in Primary Care: Vol. 1. Detection and Diagnosis: Clinical Practice Guideline. Washington, DC: US Government Printing Office, 1993. Wells KB, Stewart A, Hays RD et al. The functioning and well-being of depressed patients: results from the medical outcomes study. J Am Med Assoc 1989; 262: 914–919. Katon W, VonKorff M, Lin E et al. Population-based care of depression: effective disease management strategies to decrease prevalence. Gen Hosp Psychiatry 1997; 19: 169–178.

18

20

21

22

23

VonKorff M, Gruman J, Schaefer J, Curry SJ, Wagner EH. Collaborative management of chronic illness. Ann Intern Med 1997; 127: 1097–1102. Leventhal H, Diefenbach M, Leventhal EA. Illness cognition: using common sense to understand treatment adherence and affect cognition interactions. Cogn Ther Res 1992; 16: 143–163. Meyer D, Leventhal H, Guttmann M. Commonsense models of illness: the example of hypertension. Health Psychol 1985; 4: 115–135. Hampson SE, Glasgow RE, Foster LS. Personal models of diabetes among older adults: relationship to self-management and other variables. Diabet Educ 1995; 21: 300–307. Hampson SE, Glasgow RE, Toobert DJ. Personal model of diabetes and their relations to self-care activities. Health Psychol 1990; 9: 632–646. Petrie KJ, Weinman J, Sharpe N, Buckley J. Role of patient’s view of their illness in predicting return to work and functioning after myocardial infarction: longitudinal study. Br Med J 1996; 312: 1191–1194. Hampson SE, Glasgow RE, Zeiss AM. Personal models of osteoarthritis and their relation to self-management activities and quality of life. J Behav Med 1994; 17: 143–158. Moss-Morris R, Petrie KJ, Weinman J. Functioning in chronic fatigue syndrome: do illness perceptions play a regulatory role? Br J Health Psychol 1996; 1: 15–25. Spitzer R, Williams J, Kroenke K et al. Utility of a new procedure for diagnosing mental disorders in primary care: the PRIME-MD 1000 study. J Am Med Assoc 1994; 272: 1749–1756. Weinman J, Petrie KJ, Moss-Morris R, Horne R. The illness perception questionnaire: a new method for assessing the cognitive representation of illness. Psychol Health 1996; 11: 431–445. Beck A, Ward C, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961; 4: 561–571. Ware JE Jr, Kosinski M, Keller SD. SF-36 Physical and Mental Health Summary Scales: A User’s Manual. Boston, MA: The Health Institute, New England Medical Center, 1994. Carver CS: You want to measure coping but your protocol’s too long: consider the brief COPE. Int J Behav Med 1997; 4: 92–100. Cameron L, Leventhal EA, Leventhal H. Symptom representations and affect as determinants of care seeking in a communitydwelling, adult sample population. Health Psychol 1993; 12: 171–179. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care 1986; 24: 67–74. Leventhal H, Meyer D, Nerenz D. The common sense representation of illness danger. In Rachman S (ed.), Medical Psychology, Vol 2. New York: Pergamon Press, 1980: 7–30. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Association, 1994. Hunt LM, Jordan B, Irwin S. Views of what’s wrong: diagnosis and patients’ concepts of illness. Soc Sci Med 1989; 28: 945–956. Katon W, VonKorff M, Lin E et al. Collaborative management to achieve treatment guidelines. J Am Med Assoc 1995; 273: 1026–1031.