Primary Care Patients’ Opinions Regarding the Importance of Various Aspects of Care for Depression Lisa A. Cooper, M.D., M.P.H., Charlotte Brown, Ph.D., Hong Thi Vu, M.H.S., Deena R. Palenchar, M.S., Junius J. Gonzales, M.D., Daniel E. Ford, M.D., M.P.H., and Neil R. Powe, M.D., M.P.H., M.B.A. Abstract: The objectives of this study were to 1) ascertain the importance of various aspects of depression care from the patient’s perspective and 2) select items and scales for inclusion in a new instrument to measure primary care patients’ attitudes toward and ratings of depression care. We used a crosssectional survey at a university-based urban primary care clinic; the subjects were adult patients being recruited for a study of minor depression. To help prioritize attitudinal domains, including 126 items identified previously in focus groups, we asked patients to rate the importance of each aspect of depression care on a five-point scale. Items were ranked according to mean scores and the percentage of patients ranking the items as extremely important. The items were selected for inclusion in an instrument to measure patients’ attitudes toward depression care based on their importance ratings. We performed reliability and validity testing of scales comprising the 30 most important items and a shortened version that includes 16 items. The sample included 76 patients (mean age Presented in part at the 15th Annual Meeting of the Association for Health Services Research, Washington, D.C., June 21–23, 1998. This work was partially supported by research grants from The Commonwealth Fund (Picker/Commonwealth Scholars Program—L.A.C.), the Robert Wood Johnson Foundation (Minority Medical Faculty Development Program—LCP), and the National Institute of Mental Health (U01-MH454443). Department of Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland (L.A.C., H.T.V., D.E.F., N.R.P.), Department of Health Policy and Management, The Johns Hopkins University School of Hygiene and Public Health, Baltimore, Maryland (L.C-P., D.E.F., N.R.P.), Western Psychiatric Institute and Clinic and the Department of Psychiatry, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania (C.B., D.R.P.), Department of Psychiatry, Georgetown University Medical Center, Washington, DC (J.J.G.). Address reprint requests to: Lisa A. Cooper, M.D., M.P.H., Welch Center for Prevention, Epidemiology, and Clinical Research, 2024 East Monument Street, Suite 2-500, Baltimore, MD 21205-2223.
General Hospital Psychiatry 22, 163–173, 2000 © 2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
34.8 years; mean CES-D score, 22.2; 72% women; 36% African-American; 32% college graduates). Forty-six percent had visited a mental health professional in the past. The top 30 items for the overall sample came from the following domains: 1) health care providers’ interpersonal skills, 2) primary care provider recognition of depression, 3) treatment effectiveness, 4) treatment problems, 5) patient understanding about treatment, 6) intrinsic spirituality, and 7) financial access to services. Scales comprising items from these domains show adequate internal consistency (Cronbach’s alpha ⬎0.70) as well as convergent and discriminant validity. We have designed a brief patient-centered instrument for measuring attitudes toward depression care that has evidence for internal item consistency reliability and discriminant validity. © 2000 Elsevier Science Inc.
Background The majority of individuals with depression in the United States who seek care receive all or part of their mental health care in primary care settings [1– 4]. Reducing the burden of suffering from depression in the population will therefore require enhancing management in primary care settings. Treatment strategies for depression currently include various forms of psychotherapy (supportive listening, cognitive therapy, interpersonal therapy) and pharmacotherapy ranging from traditional tricyclic antidepressants to newer agents such as selective serotonin reuptake inhibitors and heterocyclic antidepressants. Despite the availability of these similarly efficacious treatments for depression, a large proportion of patients receive inadequate treatment [5,6]. Barriers to adequate treatment of depression in
163 ISSN 0163-8343/00/$–see front matter PII S0163-8343(00)00073-6
L.A. Cooper et al.
primary care settings include patient, physician, and health system factors. Many studies and interventions have targeted physician barriers, such as lack of skills to appropriately recognize and manage depression, and health system barriers, such as lack of adequate resources for regular follow-up visits, education, and reminders for patients. However, few studies have systematically incorporated patient perspectives to understand barriers to effective treatment of depression and to explain differences between patients in the type of depression treatment received. In conditions like depression, where several treatment options exist, patient attitudes, preferences, and ratings of care could play an important role in understanding and improving adherence to treatment. Patient attitudes regarding the appropriateness of requesting help from primary care physicians for psychosocial problems have been cited as potential barriers to appropriate recognition and treatment of depressive symptoms [7]. The most common reasons for not seeking treatment cited by depressed individuals participating in National Depression Day 1993 included lack of knowledge regarding where to go for treatment, perceptions that treatment is either too expensive or that insurance will not cover it, lack of awareness of ones’ symptoms as requiring medical care, and social stigma [8]. In focus groups, depressed patients identified concerns about patient–provider relationships, technical aspects of care, and specific attributes of medications and counseling. They discussed the impact of their spirituality, other coping strategies, life experiences, social support systems, and perceptions of stigma on their help-seeking behavior and adherence to treatment [9]. Treatment drop-out rates among primary care patients with depression are high for both medications and counseling. Studies have shown that the presence of unpleasant side effects, early recovery from depressive symptoms, and worsening of depressive symptoms are a few of the reasons patients stop antidepressant treatment [10,11]. However, not much is known about other reasons patients stop medication or fail to follow up for counseling visits. Knowledge of patients’ attitudes toward antidepressant medications and counseling would help practicing physicians, researchers, health educators, and managers of health care plans target their efforts to improve quality of care for depression in primary care. The primary objectives of this study were to 1) ascertain the importance of various aspects of de-
164
pression treatment (medical care and self-care) from the perspective of patients and 2) select domains and items for inclusion in an instrument to measure primary care patients’ attitudes toward depression care. Such an instrument might be used by researchers to explain differences between patients in preferences for and receipt of depression care. Additionally, the instrument might be used by clinicians to facilitate joint decision-making regarding depression treatment with patients.
Methods Study Design and Patient Selection The study design was a cross-sectional survey of attenders of an urban university-based primary care clinic who had been recruited for a study of minor depression. To be eligible for the study, patients had to be English-speaking adults between the ages of 18 and 64, African American or White, and able to give informed consent. As part of the study, patients were asked to complete the 20-item Center for Epidemiologic Studies Depression Scale (CES-D). The CES-D is a self-report instrument with established reliability and validity [12]. A score of 16 or higher on the CES-D is generally used to identify patients with clinically significant depressive symptoms [13–15]. However, this study included individuals with lower scores, many of whom had minor depression, dysthymia, major depressive disorder in partial remission, and mixed anxiety/depressive conditions. Previous work has suggested that individuals with milder symptoms may still benefit from close follow-up and treatment. Most of these individuals have had prior experiences with depression treatment, have significant functional disability, and are potential candidates for early intervention [16,17]. The CES-D was not used in this study as a diagnostic instrument, but rather as a measure of depression symptom level. There is controversy regarding whose perspective should be obtained when measuring patients’ attitudes and preferences. We chose to study this group of patients because 1) we believe patients with no symptoms are unlikely to have experiences upon which to base ratings of the importance of various aspects of depression treatment and 2) we believe limiting our study to only individuals meeting diagnostic criteria for major depressive disorder would exclude the majority of primary care patients with depressive symptoms. These patients’
Patient Attitudes about Depression Care
depressive symptoms are not static. While some symptoms may resolve spontaneously, others progress to a full disorder, are associated with functional impairment, and respond to treatment with antidepressant medications or counseling.
Domain and Item Selection for Patient Attitudes towards Depression Care Instrument Focus Groups. In a previous study, we conducted focus groups to identify specific domains of patients’ help-seeking behavior and attitudes towards depression treatment [9]. Three focus groups were held with African-American patients, White patients, and a biracial group of health professionals. Discussions were audiotaped, transcribed, and reviewed by investigators to identify and group distinct thoughts and concepts into related categories. Patients identified a wide range of factors that they felt played a role in their help-seeking behavior and adherence to depression treatment. The comments were grouped into 16 categories or themes related to depression care using content analysis. Patients made more comments than professionals regarding the impact of spirituality, social support systems, coping strategies, life experiences, patient-physician relationships, and specific attributes of medications and counseling. African Americans raised more concerns than whites regarding stigma, and unlike whites, they tended to view depression and its treatment in a religious or spiritual framework [9]. Importance Survey. For this study, we designed a self-administered survey that we asked patients to complete while at the clinic or return by mail. In the introduction to the survey, patients were given a brief description of the symptoms and criteria for major depression. They were told that depression is treatable and given a brief description of the major types of treatment. They were also told that there were no right or wrong answers to the items, and that we sought their opinion regarding the importance of each item for good quality of care for depression. We told patients, “We would like to know what you think about getting treatment for depression and what would be most important to you if you had to make this decision.” For example, an item that measured the importance of trust between patients and health professionals (included in the domain of health care providers’ interpersonal skills) asked, “How important would it be
that you could trust the doctor or therapist to act in your best interests?” The importance survey included 126 items that were developed using comments from focus groups of health professionals and depressed patients. The items were grouped according to categories or themes from focus groups. Categories generally included 5 to 13 items constructed using the actual words and descriptions given by patients in focus groups [9]. We asked patients to rate each item on a five-point scale according to its importance for high quality care for depression: 1, not at all important; 2, not very important; 3, somewhat important; 4, very important; and 5, extremely important. All items were related to care for or experience with depression. The availability of coping strategies might impact on patients’ choices for formal treatment. Because of the salience of spiritual beliefs and practices as a method of self-care or coping with depression in focus group discussions, we included a separate domain for spirituality and used the same importance scale used for other aspects of care. For coping strategies other than spirituality, we asked patients to rate how much of the time in the past 6 months they had tried each method of coping when feeling “down” on a five-point scale: 1, none of the time; 2, some of the time; 3, about half of the time; 4, most of the time; 5, all of the time. This category, called “other coping strategies,” included 37 items. The importance survey also included questions regarding lifetime experience with depressed mood and loss of interest or pleasure (yes/no), past visits to mental health professionals (yes/no), recency of mental health visits (within the past month, 1– 6 months ago, 7–12 months ago, ⬎12 months ago), current impairment, such as difficulty doing work, taking care of things at home, or getting along with other people, attributed to depressive symptoms (yes/no/sometimes), and demographic characteristics including age, gender, race, education, marital and employment status.
Analyses Importance Rankings. We ranked the importance of the 126 items by mean scores and by the percentage of patients rating the items as extremely important. The top 30 items by either method were identified. Spearman’s rank-order correlation coefficient tests were then used to compare rankings of these 30 items sorted by mean score and percent extremely important.
165
L.A. Cooper et al.
Scale Development and Preliminary Psychometric Testing. To develop an instrument to measure patient attitudes toward and ratings of care for depression, we began with the domains identified in focus groups and selected items for inclusion based upon their importance ratings by patients. Because we wanted to minimize patient burden with a brief questionnaire and focus on their most important concerns, we first included the top 30 most important items (either by mean score or percent extremely important). To examine the construct validity of the scales, we tested several hypotheses. First, we examined the convergent and discriminant construct validity of the item-scales by determining if each item within a scale correlated with the remaining items in the same scale (Pearson’s correlation coefficient ⬎0.40) and if each item correlated more highly with other items in its putative scale than with other scales. Then we examined whether each scale’s internal consistency reliability coefficient (the degree of convergence among all items in the scale) or Cronbach’s alpha was ⬎0.70 and whether it exceeded that scales’ correlations with other scale scores. To do this, we used the Multitrait Analysis Program (MAP) [18,19]. Using the MAP, internal item-scale consistency is defined by having at least 90% of item-scale correlations that are greater than or equal to 0.40 (corrected). Scaling successes for item discriminant validity are defined by having at least 80% of itemscale correlations in the total data set and within each individual scale be at least 2 standard errors greater than the correlations of the item to other scales. In order to meet the standard of scale reliability, all scales must have Cronbach’s alpha coefficients of 0.70. To create an even shorter instrument, we repeated these analyses using a subset of the 30 most important items. The two items with the highest importance ratings (either by mean score or percent extremely important) from six of the seven scales were selected for inclusion in the short version. The abbreviated treatment effectiveness scale, however, included the four items with the highest importance ratings (two related to antidepressant medication and two related to individual counseling). The long and short versions of the questionnaire include 30 and 16 items, respectively, that were converted from importance ratings to attitudinal statements and reviewed by our research team, which includes three general internists/health services researchers, a psychiatrist, and a clinical psy-
166
chologist. Items were deleted by consensus of the investigators if they were endorsed infrequently or rated as not important by patients, felt to be redundant or unclear by the investigators, and if they did not correlate well with any scale.
Results Characteristics of the Study Sample A total of 97 patients were eligible to complete the questionnaire. Six patients refused, including five Whites (3 women, 2 men) and one AfricanAmerican man. Fifteen patients failed to return the questionnaire by mail. Compared to respondents, nonrespondents were somewhat older, more likely to be college graduates, and had significantly lower CES-D scores. A sample of 76 patients completed the questionnaire (response rate 78%). The average age of patients was 34.8 years. The sample was 72% women, 36% African-American, 64% White, 29% married, 32% college graduates, and 49% were employed full-time. The mean score on the Center for Epidemiologic Studies Depression Scale was 22.2. Fiftyfour percent of the patients reported definite impairment in functioning related to their symptoms, and another 24% reported that they felt impaired “sometimes.” Over half of the patients reported having depression in the past, and 46% of the sample had visited a mental health professional sometime in the past (Table 1). Table 2 shows the 16 categories of depression care elicited from patient and health professional focus groups. Thirteen of these categories were included in the importance survey designed for this study. Three of the original focus group categories were not included on the importance survey either because there were no patient comments or because the comments were of a general nature.
Importance Ratings We identified the top 30 items ranked by mean scores and by the percentage of patients rating the item as extremely important. Spearman’s rankorder correlation coefficient test showed that the rankings by these two methods were highly correlated (r s⫽.88, P⬍.0001). The most important items were: being able to trust the health professional to act in one’s best interests, specific attributes of medication and counseling (effectiveness, ability to restore function, side effects, negative aspects), patient understanding of depression (having
Patient Attitudes about Depression Care
Table 1. Characteristics of study participants (n⫽76) n Demographics Age 18–34 35⫹ Race African American White Gender Female Male Marital Status Married Never Married Separated/divorced/widowed Education High school or less Some college College graduate Employment Status Unemployed Employed part-time Employed full-time Depression Status CES-D Score 11–15 16–21 22⫹ Ever depressed in the past Yes No Currently impaired by depressive symptoms Yes Sometimes No Ever visited mental health professional Yes No Most recent visit to mental health professional (n⫽35) ⬍1 month ago 1–6 months ago 7–12 months ⬎12 months ago
Table 2. Categories of depression care elicited from patient and professional focus groups and used in importance survey %
42 34
55 45
27 49
36 64
55 21
72 28
22 41 13
29 54 17
24 21 29
39 28 32
27 12 37
35 16 49
16 27 33
21 36 43
41 32
56 44
39 17 16
54 24 22
35 41
46 54
4 3 0 28
11 9 0 80
information, knowing what to expect from treatment), health professionals’ interpersonal skills (knows all about patients, listens, understands patients’ problems, approaches patients as individuals, makes patients feel comfortable, supports and
Spirituality Social support Coping strategies Life experiences Psychological receptivity Stigma Provider recognition of depression Provider traininga Provider treatment biasa Patient–provider relationships Coordination and referral issues Physical health and depression Attributes of treatment Medications, counseling, patient education Access to mental health services General commentsa Advice for health professionals a
These categories were not included on the importance survey because there were either no patient comments or very general comments from patient focus groups.
encourages patients), and physician recognition and validation of patients’ depression (recognizes depression, believes patients’ symptoms are real). Two items related to access were among the top 30 items: being able to afford mental health treatment and having insurance coverage for mental health treatment. Four items related to spirituality were also among the top 30 items (Table 3). Items from the social support, other coping strategies, and stigma domains were not ranked among the top 30 items.
Importance Ratings by Demographic Variables There were no differences in importance ratings by age. Women gave higher ratings than men to whether the primary care provider recognized depression (P⫽.04) and whether the health care provider gave information (P⫽.02). There were differences in the importance ratings of several items by marital status, education, and employment status. However, none of these differences were consistent across several items within a domain. There were statistically significant differences in importance ratings of several items within the spirituality domain by race. Compared with White patients, African-American patients gave higher ratings to items related to spirituality. There was less racial disparity in the rankings of other aspects of depression care (effectiveness of medications and counsel-
167
L.A. Cooper et al.
Table 3. Most important aspects of depression care identified by 76 patients PARC-D item no.
Domain/ Scale
Aspect of care
Mean score
Rank
% EI
Rank
1 11 21 29 3 20 30 13 – 17 12 14 2 – 4 5 18 6 19 26 27 7 22 9 15 8 – 28 10 16 – 25 24 23
HP TE PE REC HP PE REC TE TP TP TE TE HP HP HP HP TP HP TP ACC ACC HP IS HP TE HP TE ACC HP TE TE IS IS IS
Trust HP to act in one’s best interests Antidepressant medications restore functioning Having information about treatment HP recognizes depression HP listens to you Knowing what to expect HP believes one’s symptoms Antidepressant medications are effective Antidepressant medications make you feel worse Antidepressant medications are addictive Counseling restores functioning Counseling helps with problems HP understands your problems HP gives information HP gives support HP approaches you as an individual Antidepressant medications make you feel drugged You feel comfortable with HP Antidepressant medications should fix, not just cover up Affording mental health treatment Having insurance for mental health One HP who knows all about you Having faith in God Whether HP judges you Able to talk about what bothers you HP gives you guidance Able to talk about life experiences The system gives you hassles You feel like a “guinea pig” Antidepressant medications take long to start working Antidepressant medications help you to sleep Asking God to forgive you Being able to thank God Prayer heals depression
4.61 4.53 4.53 4.50 4.50 4.50 4.48 4.47 4.45 4.43 4.43 4.42 4.41 4.30 4.30 4.29 4.28 4.28 4.27 4.24 4.07 4.07 4.05 4.04 4.03 4.01 3.97 3.91 3.91 3.87 3.83 3.81 3.73 3.66
1 2 2 4 4 4 7 8 9 10 10 12 13 14 14 16 17 17 19 20 21 21 23 24 25 26 27 28 28 30 NR NR NR NR
77 67 63 66 62 62 68 65 69 72 60 59 61 54 51 49 61 49 56 61 53 44 54 51 39 36 35 33 37 31 43 47 45 39
1 5 8 6 9 9 4 7 3 2 14 15 12 17 20 23 11 22 16 12 19 26 18 21 29 NR NR NR 30 NR 27 24 25 28
Abbreviations: %EI, percent of sample rating item as extremely important; HP, health care providers’ interpersonal skills; TE, treatment effectiveness; PE, patient education, information, and understanding; TP, treatment problems; IS, intrinsic spirituality; ACC, access; REC, primary care provider recognition of depression; NR, not ranked in top 30 items by this method; –, not included in PARC-D because of redundancies or poor correlations with other items/scales. Rank order from most highly rated by mean (⫽1) Spearman’s rho ⫽ 0.88, P⬍.0001, indicating rankings by mean score and %EI are highly correlated.
ing, health care providers’ technical and interpersonal skills, and access) (data not shown).
Importance Ratings by Depression Status Patients with CES-D scores ⱖ22 gave higher ratings to having faith in God (P⫽.04) and getting support from health professionals (P⫽.04). There were no differences in importance ratings by impairment status. Patients who had never visited a mental
168
health professional rated knowing what to expect from treatment as more important than patients who had visited a mental health professional (P⫽.04).
Scale Development and Preliminary Psychometric Testing The following scales were developed after tests of internal consistency and discriminant validity were
Patient Attitudes about Depression Care
performed using the 30 items rated as most important to high quality care of depression by either mean score or percent extremely important and a subset of 16 of these items: 1) health professionals’ interpersonal skills; 2) primary care physician recognition and validation of patients’ depression; 3) treatment effectiveness; 4) treatment problems; 5) patients’ need for education, information, and understanding regarding depression; 6) intrinsic spirituality; and 7) financial access to services. Conventional wisdom would suggest that patient education, information, and understanding are separate constructs. However, in this sample, ratings of the importance of these characteristics were highly correlated in factor analyses. Additionally, there was overlap among the health care provider skills, treatment, and patient education scales, indicating that patients might not easily differentiate among these domains. Because there was not complete overlap, we believe that primary care physicians and mental health specialists would probably find it helpful to have patients’ attitudes toward and ratings of these aspects of care identified separately. Therefore, we chose to keep these domains separate
from one another. Four of the top 30 items identified by either mean score or percent extremely important (34 items overall) were not included in the final instrument because they had poor correlations with other items or scales or were felt to be redundant or unclear: 1) medications will make me feel worse; 2) medications help me sleep; 3) I am able to talk about my life experiences with health professionals; and 4) my doctor gives me information. Table 4 shows descriptive statistics and reliability estimates for the scales or domains of the 30-item and 16-item versions of the Patient Attitudes toward and Ratings of Care for Depression (PARC-D) questionnaire. Results of internal consistency and a correlation matrix of the seven scales are shown in Table 5. Internal consistency reliability was generally acceptable, with Cronbach’s alpha ⬎0.70 for all scales in the 30-item and 16-item versions. Item convergent and discriminant validity were satisfactory. As expected, each multi-item scale’s internal consistency coefficient exceeds it correlations with other scales. Using the MAP, there were 64% scaling successes, 33% possible scaling failures (where
Table 4. Descriptive statistics and reliability estimates for domains of the PARC-D 30 item and 16 item versions
Domain/scale Health provider skills HP-10 HP-2 Treatment effectiveness TE-6 TE-4 Treatment problems TP-3 TP-2 Patient education PE-2 Intrinsic spirituality IS-4 IS-2 Access to care ACC-3 ACC-2 Recognition of depression REC-2
Mean
Standard deviation
Percentage lowest score
Percentage highest score
Reliability
50 10
82.10 89.15
17.76 19.78
1.5 2.9
7.4 57.4
0.91 0.87
6 4
30 20
82.84 87.31
18.47 18.80
1.5 1.5
14.7 42.6
0.90 0.92
3 2
15 10
84.31 85.66
23.55 25.21
1.5 4.4
42.6 60.3
0.87 0.89
2
10
88.24
17.66
1.5
54.4
0.90
4 2
20 10
69.97 71.88
31.98 33.13
7.4 7.4
32.4 44.1
0.94 0.91
3 2
15 10
76.47 78.13
27.62 32.57
2.9 11.8
26.5 50.0
0.85 0.94
2
10
89.34
17.84
0.0
61.8
0.81
Number of items
Number of levels
10 2
The mean and SD were calculated based on transformed scores (observed score-lowest possible score)/(highest possible score-lowest possible score) ⴱ 100. The estimates in this table were based on an analytic sample (n⫽68) with complete responses to all PARC-D 30 items.
169
L.A. Cooper et al.
Table 5. Reliability coefficients and inter-scale correlations patient attitudes toward and ratings of care for depression (PARC-D) questionnaire 30-item and 16-item versions Domain PARC-D 30
HP
TE
TP
PE
IS
HP (.91) (.87) .79 .72 .72 TE .77 (.90) (.92) .57 .75 TP .68 .62 (.87) (.89) .54 PE .71 .72 .56 (.90) (.90) IS .04 .03 .00 ⫺.10 ACC .51 .60 .29 .46 REC .40 .45 .13 .24 Total interscale correlation .89 .80 .86 .86 .68 .65 .71 .71 Scale-total correlation corrected for overlap Total 0.73 .73 0.77 .74 0.58 .51 .66 .62
ACC
Domain PARC-D 16
REC
⫺.10 .46 .37 ⫺.00 .53 .43 ⫺.10 .25 .16 ⫺.10 .35 .24 (.94) (.91) .16 .21 .14 (.85) (.94) .47 .20 .47 (.81) (.81)
.34
.28
.70
.73
.53
.59
.06
.01
.58
.55
.47
.49
HP TE TP PE IS ACC REC
Scale internal consistency reliability coefficients (Cronbach’s Alpha) are presented in parentheses along the diagonal. Italicized and shaded numbers correspond to the PARC-D 16. Other numbers correspond to the PARC-D 30. Abbreviations: HP, health care providers’ interpersonal skills; TE, treatment effectiveness; TP, treatment problems; PE, patient education, information, and understanding; IS, intrinsic spirituality; ACC, access; REC, primary care physician recognition of depression.
item-scale correlations were greater than the correlations of the item to other scales, but not by 2 standard errors), and only 3% definite scaling failures. Table 6 shows the 30 items comprising the final Patient Attitudes Toward and Ratings of Care for Depression (PARC-D) questionnaire, listed by domains. The response set is a five-point Likert scale with the following headings and scores: 1 (strongly disagree), 2 (disagree), 3 (neither agree nor disagree), 4 (agree), and 5 (strongly agree). Items that reflect negative attitudes have reverse scoring. The items included in the 16-item version are italicized and shaded.
Discussion We employed a comprehensive, patient-centered approach to develop an instrument to measure primary care patients’ attitudes toward and ratings of care for depression (PARC-D questionnaire). In this study, patients rated health professionals’ interpersonal skills, primary care physicians’ recognition of depressive symptoms, having health-related information and understanding of depression, specific attributes of antidepressant medication and counseling, intrinsic spirituality, and access as the most important aspects of good quality care for depres-
170
sion. Trust in health professionals was rated the most important aspect of depression care. Other researchers have cited the important influences of interpersonal trust and social trust on the quality of physician-patient interactions [20,21]. As in previous work, patient attitudes towards caring (humanness) and curing (competence) aspects of doctor behavior appear to be equally important [21,22]. These domains also demonstrate some overlap when measured from patients’ perspective. The high ratings of patient education and information items indicate, similar to previous work, that patients highly value the effective communication of health-related information [22,23]. The most important attributes of treatment (with medications and counseling) were their effectiveness, ability to restore patients to their usual level of functioning, ability to cure instead of just covering up the underlying problem, and the absence of perceived negative side effects, such as making patients feel drugged, taking too long to work, and causing patients to become addicted. Many dimensions of access have been cited as important to high quality patient care in previous work [24,25]. In this study, as in previous work, financial access appeared to be very important to patients [8]. This may be due to the fact that insur-
Patient Attitudes about Depression Care
Table 6. Questions and domains in the patient attitudes toward and ratings of care for depression (PARC-D 30) questionnairea Item 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.
Scale/domain
I trust my health care provider to act in my best interests. My health care provider understands my problems. My health care provider listens to me. My health care provider gives me support and encouragement. My health care provider approaches me as an individual. I am comfortable with my health care provider. I have one health care provider who knows all about me. My health care provider gives me guidance. My health care provider judges me. (R) I will feel like a guinea pig being experimented upon if I seek help for depression. (R) Antidepressant medications will restore me to my normal level of functioning. Counseling will restore me to my normal level of functioning. Most antidepressant medications are effective in treating depression. Counseling will help me just as much as antidepressant medication. I can talk about what bothers me in counseling. Antidepressant medications take too long to start working. (R) Antidepressant medications are addictive. (R) Antidepressant medications will make me feel drugged. (R) Antidepressant medications do not fix depression; they just cover it up. (R) I know what to expect regarding treatment of depression. I have enough information to deal with my depression. Faith in God will heal my depression. Prayer alone can heal depression. Thanking God helps depression to get better. Asking God for forgiveness will help heal my depression. I can afford mental health treatment for depression. I have health insurance that will cover enough of the costs of my depression care. The health care system is a hassle when I try to get help for depression. (R) My primary care provider recognizes when something is wrong with me. My primary care provider believes my symptoms are real.
HP HP HP HP HP HP HP HP HP HP TE TE TE TE TE TE TP TP TP PE PE IS IS IS IS ACC ACC ACC REC REC
a All items are rated on a five-point scale where 1 ⫽ strongly disagree, 2 ⫽ disagree, 3 ⫽ neither agree nor disagree, 4 ⫽ agree, and 5 ⫽ strongly agree. Items with (R) next to them have reverse scoring. PARC-D 16 items are italicized and shaded. Abbreviations: HP, health care providers’ interpersonal skills; IS, intrinsic spirituality; TE, treatment effectiveness; ACC, access; TP, treatment problems; REC, recognition of depression; PE, patient education, information, and understanding.
ance coverage for mental health services is frequently not as high as coverage for physical health problems. Spirituality was rated among the most important aspects of care for depression in our sample. Some authors suggest that religion and spirituality are most important to a patient’s coping when some ambiguity occurs with respect to the efficacy of the treatment [26]. In the case of depression, medications and counseling are each effective approximately 60 –70% of the time [27,28]. Patients’ preferences for treatment and their unique beliefs and coping strategies can therefore play an important role. Research indicates that patients want their physicians to address issues of faith and spirituality
in the course of their treatment, and that patients with strong spiritual and religious tendencies want physicians to pray with them [29 –31]. However, primary care physicians cite lack of time and inadequate training as the most important barriers to discussing spirituality with patients [32]. Studies are needed to determine, from the patient’s perspective, how spirituality can be incorporated into medical care and how patients’ spiritual beliefs and practices impact their interpretations of symptoms, perceived need for treatment, preferences, and adherence to treatment for depression and other serious medical conditions. With respect to other coping strategies, stigma, and social support, items from these domains were either
171
L.A. Cooper et al.
endorsed infrequently or not rated as being very important. The most frequently used coping strategies other than spirituality were taking care of oneself, waiting for things to get better, and listening to music. Surprisingly, stigma items (social embarrassment if family, friends, co-workers, or employers discovered one was being treated for depression) were not rated as important. Previous work has suggested that stigma is an important barrier to mental health treatment, particularly among the severely depressed [8,33]. However, recent work suggests that perceptions of mental illness stigma are not as high as previously believed [34]. Limitations of the study should be discussed. First, although we studied a randomly selected population, the study has a relatively small sample size from one geographic area. The results support previous findings from focus group participants in a different urban area, and the study also contributes to the literature because little work has been done previously to describe primary care patients’ priorities for treatment of depression. We have created short and long versions of the instrument that need to be validated in various settings and in patients with a range of treatment experiences and symptomatology. Settings might include primary care samples in other regions of the United States, specialty mental health settings, and the community. Second, this sample consisted of only Caucasian and African Americans; therefore, the instrument may not be applicable to other ethnic groups. Third, the CES-D is not a diagnostic instrument to measure major depressive disorder. However, these patients are representative of the majority of patients with depressive symptoms seen in primary care, many of whom are in partial remission from major depressive episode and have milder forms of depression (minor depression, mixed anxiety/depressive disorders). In our sample, over half of the patients reported having a prior episode of depression, and almost 80% reported that their symptoms impaired them. Fourth, this study used a crosssectional design and therefore could not provide information about the predictive validity of the patient attitudes and ratings of care questionnaire. Another limitation is the inclusion of only patient views in constructing the instrument. Depending upon the intent of its use, these items might need to be supplemented with theoretical constructs from the literature and with more input from health professionals. Finally, this study does not describe patient preferences; rather, it describes the factors pa-
172
tients think are most important and that could explain their preferences and choices of treatment. One of the primary dimensions of patientcentered care is a respect for patients’ values, preferences, and expressed needs [35]. To deliver patient-centered care, health care professionals must understand patients’ models of illness and incorporate this dimension into caring for patients. One of the benefits of patient-centered care is that patients are more informed about their illness and involved in decision-making regarding their treatment. Increasing patient involvement in care via negotiation and consensus seeking improves patient satisfaction and outcomes [36 –39]. By focusing on patient perspectives in the design of the PARC-D, we identified several important aspects of care that primary care physicians, health services researchers, health care managers, and health policy-makers should incorporate when caring for depressed patients, measuring quality of care, designing interventions to improve quality of care, and making policy decisions that affect care for individuals with depressive conditions. This study suggests that the patient-provider relationship is particularly important and should be considered in decisions affecting health policy for patients with depression. Policies that reduce continuity of care and limit the amount of time primary care patients spend with their physicians may compromise these relationships and lead to poorer depression outcomes. Future research should explore how patient attitudes and ratings of care differ across demographic groups, patients with different levels of depression and functioning, and patients with different experiences with treatment. Additionally, research using this instrument could determine how patient attitudes influence their preferences for treatment, receipt of guideline-concordant care, and depression outcomes.
References 1. Shapiro S, Skinner EA, Kessler LG, et al: Utilization of health and mental health services: Three epidemiologic catchment area sites. Arch Gen Psych 41:971– 978, 1984 2. Regier D, Goldberg I, Taube C: The defacto US Mental Health Services System. Arch Gen Psych 35:685– 693, 1978 3. Kessler L, Burns B, Shapiro S, Tischler G: Psychiatric diagnoses of medical service users: evidence from the Epidemiologic Catchment Area Program. Am J Public Health 77:18–24, 1987 4. Leaf P, Livingston M, Tischler G: Contact with health professionals for the treatment of psychiatric and emotional problems. Med Care 23:1322–1337, 1985
Patient Attitudes about Depression Care 5. Katon W, Von Korff M, Lin E, et al: Adequacy and duration of antidepressant treatment in primary care. Med Care 30:67–76, 1992 6. Wells KB, Katon W, Rogers B, Camp P: Use of minor tranquilizers and antidepressant medication by depressed outpatients: results from the Medical Outcomes Study. Am J Psychiatry 151:694–700, 1994 7. Good MJD, Good BJ, Cleary PD: Do patient attitudes influence physician recognition of psychosocial problems in primary care? J Fam Pract 25:53–59, 1987 8. Magruder KM: Community education and screening programs. In Jenkins R, Ustun TB (eds), Preventing Mental Illness. Mental Health Promotion in Primary Care. Chichester, John Wiley & Sons, 1998 9. Cooper-Patrick L, Powe NR, Jenckes MW, et al: Identification of patient attitudes and preferences regarding treatment of depression. J Gen Intern Med 12:431– 438, 1997 10. Lin EHB, Von Korff M, Katon W, et al: The role of the primary care physician in patients’ adherence to antidepressant therapy. Med Care 33:67–74, 1995 11. Thompson J, Rankin H, Aschcroft CW: The treatment of depression in general practice. Psychological Med 12:741–751, 1982 12. Radloff LS: The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychological Measurement 1:385–401, 1977 13. Roberts RE, Vernon SW: The Center for Epidemiologic Studies Depression scale: its use in a community sample. Am J Psychiatry 140:41–46, 1983 14. Schulberg HC, Saul M, McClelland M: Assessing depression in primary medical and psychiatric practices. Arch Gen Psychiatry 42:1164–1170, 1985 15. Katon W, Schulberg H: Epidemiology of depression in primary care. Gen Hosp Psychiatry 14:237–247, 1992 16. Wells KB, Stewart A, Hays RD, et al: The functioning and well-being of depressed patients. Results from the Medical Outcomes Study. JAMA 262:914–919, 1989 17. Broadhead WE, Blazer DG, George LK, Tse CK: Depression, disability days, and days lost from work in a prospective epidemiological survey. JAMA 264: 2524–2528, 1990 18. Hays RD, Hayashi T: Beyond internal consistency reliability: rationale and user’s guide for multitrait analysis program on the microcomputer. Behavior Research Methods, Instruments & Computers 22:167– 175, 1990 19. Ware JE Jr, Harris WJ, Gandek B, et al: MAP-R for Windows: Multitrait/Multi-Item Analysis Program—Revised User’s Guide. Boston, Health Assessment Lab, 1997 20. Mechanic D, Schlesinger M: The impact of managed care on patients’ trust in medical care and their physicians. JAMA 275:1693–1697, 1996 21. Thom DH, Campbell B: Patient-physician trust: an exploratory study. J Family Practice 44:169–176, 1997 22. Rubin HR, Jenckes M, Fink NE, et al: Patient’s view of
23.
24. 25. 26. 27.
28.
29. 30. 31. 32. 33. 34.
35.
36. 37. 38. 39.
dialysis care: development of a taxonomy and rating of importance of different aspects of care. Am J Kidney Dis 30:793–801, 1997 Laine C, Davidoff F, Lewis CE, et al: Important elements of outpatient care: a comparison of patients’ and physicians’ opinions. Ann Intern Med 125:640– 645, 1996 Andersen RM, McCutcheon A, Aday LA, et al: Exploring dimensions of access to medical care. Health Serv Res 18:49–74, 1983 Penchansky R, Thomas JW: The concept of access: definition and relationship to consumer satisfaction. Med Care 19:127–140, 1981 Jenkins RA, Pargament KI: Religion and spirituality as resources for coping with cancer. J Psychosocial Oncol 1:51–74, 1995 Agency for Health Care Policy Research. Depression in Primary Care: Volume 1. Detection and Diagnosis. Rockville, MD, U.S. Department of Health and Human Services, Public Health Service, 1993 Agency for Health Care Policy and Research. Depression in Primary Care: Volume 2. Treatment of Major Depression. U.S. Department of Health and Human Services, Public Health Service, Rockville, MD, 1998 King DE, Bushwick B: Beliefs and attitudes of hospital inpatients about faith healing and prayer. J Fam Pract 39:349–352, 1994 Oyama O, Koenig HG: Religious beliefs and practice in family medicine. Arch Fam Med 7:431–435, 1998 Daaleman TP, Nease DE Jr: Patient attitudes regarding physician inquiry into spiritual and religious issues. J Fam Pract 39:564–568, 1994 Ellis MR, Vinson DC, Ewigman B: Addressing spiritual concerns of patients: family physicians’ attitudes and practices. J Fam Pract 48:105–109, 1999 Sussman LK, Robins LN, Earl F: Treatment-seeking for depression by black and white Americans. Soc Sci Med 24:187–196, 1987 Cooper-Patrick L, Gallo JJ, Powe NR, et al: Mental health service utilization by African-Americans and whites: the Baltimore epidemiologic catchment area follow-up. Medical Care 37:1034–1045, 1999 Gerteis M, Edgman-Levitan S, Daley J, Delbanco TL: Through the Patient’s Eyes. Understanding and Promoting Patient-Centered Care. San Francisco, JosseyBass Publishers, 1993. Kaplan SH, Greenfield S, Gandek B, et al: Characteristics of physicians with participatory decisionmaking styles. Ann Intern Med 124:497–504, 1996 Barsky AJ, Kazis LE, Freiden RB, et al: Evaluating the interview in primary care medicine. Soc Sci & Med 14A:653–658, 1980 Greenfield S, Kaplan S, Ware JE Jr: Expanding patient involvement in care. Ann Intern Med 102:520–528, 1985 Kaplan SH, Greenfield S, Ware JE Jr: Assessing the effects of physician-patient interactions on the outcomes of chronic disease. Med Care 27:S110–S127, 1989
173