ORIGINAL ARTICLE
Psychiatrists’ and Primary Care Physicians’ Beliefs About Overtreatment of Depression and Anxiety Ryan E. Lawrence, MD, MDiv,* Kenneth A. Rasinski, PhD,† John D. Yoon, MD,‡ and Farr A. Curlin, MD§ Abstract: Critics say that physicians overdiagnose and overtreat depression and anxiety. We surveyed 1504 primary care physicians (PCPs) and 512 psychiatrists, measuring beliefs about overtreatment of depression and anxiety and predictions of whether persons would benefit from taking medication, investing in relationships, and investing in spiritual life. A total of 63% of PCPs and 64% of psychiatrists responded. Most agreed that physicians too often treat normal sadness as a medical illness (67% of PCPs and 62% of psychiatrists) and too often treat normal worry and stress as a medical illness (59% of PCPs, 55% of psychiatrists). Physicians who agreed were less likely to believe that depressed or anxious people would benefit “a lot” from taking an antidepressant (36% vs. 58% of PCPs) or antianxiety medication (25% vs. 42% of PCPs, 42% vs. 57% of psychiatrists). Most PCPs and psychiatrists believe that physicians too often treat normal sadness and worry as a medical illness. Key Words: Depression, Anxiety, Antidepressant, Anxiolytic, Religion
Within this debate, some have argued from explicitly religious frameworks to critique what they perceive as the medicalization of normal emotional problems and life stress (DeMoss, 2006). Jay Adams, in his controversial yet popular (over 30 printings) book Competent to Counsel, argued that clergy, not psychiatrists, are most qualified to help people address emotional problems and life stress (Adams, 1970). Continuing in this tradition, David Powlison, director of the Christian Counseling & Educational Foundation, wrote: Biopsychiatry will cure a few things… But in the long run, unwanted and unforeseen side effects will combine with vast disillusionment. The gains will never live up to the promises. And the lives of countless people, whose normal life problems are now being medicated, will not be qualitatively changed and redirected (Powlison, 2010, p. 266).
(J Nerv Ment Dis 2015;203: 120–125)
A
fter fluoxetine—the first selective serotonin reuptake inhibitor (SSRI)—was introduced in the United States in 1988, the number of physician visits for depression increased dramatically (10.99 million in 1985, 20.43 million in 1993–1994). Visits involving an antidepressant prescription similarly increased, and this growth can be entirely accounted for by the use of SSRIs (Pincus et al., 1988). Subsequently, a debate emerged over whether depression is overdiagnosed and overmedicated. Some argue that current diagnostic criteria medicalize normal sadness (Horwitz and Wakefield, 2007; Parker, 2007), whereas others suggest that increased diagnosis and treatment are beneficial because they reduce suicides, reduce stigma, and increase productivity (Hickie, 2007). A similar debate surrounds the diagnosis and treatment of anxiety—the most common mental health disorder internationally (The WHO World Mental Health Survey Consortium, 2004). Some argue that a degree of anxiety is a normal response to life stressors (Abrams, 2012) and that “The hasty, overmedicated handling of anxiety is a direct result of DSM-III and the large number of new disorders it created” (Lane, 2008, p. 139). At the same time, however, primary care data from the 1990s showed that half of patients with significant anxiety were undiagnosed and untreated and that patients with untreated anxiety exhibit worse physical functioning, social functioning, emotional problems, general mental health, energy, pain, and general health perceptions and had more limitations in daily activity caused by physical health (Fifer et al., 1994).
*Department of Psychiatry, Columbia University Medical Center and the New York State Psychiatric Institute, New York, NY; †Program on Medicine and Religion and ‡Department of Medicine and the MacLean Center for Clinical Medical Ethics, University of Chicago, Illinois; and §Trent Center for Bioethics, Humanities, and History of Medicine, and The Divinity School, Duke University, Durham, NC. Send reprint requests to Ryan E. Lawrence, MD, MDiv, Department of Psychiatry, Columbia University Medical Center, New York, Presbyterian Hospital, 9 Garden North, 177 Fort Washington Avenue, New York, NY 10032. E-mail: [email protected]. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0022-3018/15/20302–0120 DOI: 10.1097/NMD.0000000000000247
120
www.jonmd.com
The current study attempts to describe relevant attitudes regarding the medical diagnosis and treatment of depression and anxiety among US psychiatrists and primary care physicians and the extent to which such attitudes are associated with physician specialty and physician religiosity. We tested several hypotheses. First, we expected physicians to be divided about the medicalization of depression and anxiety, with large proportions believing that normal sadness and worry are too often treated as medical conditions. Second, we predicted that large proportions would also believe that undertreatment of depression and anxiety are greater problems than overtreatment, with psychiatrists being more likely than primary care physicians to believe this. Third, we tested whether physicians who believe that normal sadness and worry are too often treated as a medical problem will be less likely to endorse a benefit of using antidepressants or antianxiety medication and more likely to endorse a benefit of patients paying more attention to their relationships and spiritual lives. Fourth, we predicted that physicians who are more religious will be more likely to believe that normal sadness and worry are overly medicalized and less likely to believe that patients will significantly benefit from using medication.
METHODS Between September 2009 and June 2010, we mailed a confidential self-administered questionnaire to a stratified random sample consisting of 1504 US primary care physicians and 512 US psychiatrists 65 years old or younger. The sample was generated from the American Medical Association Physician Masterfile, a database intended to include all practicing US physicians. To increase minority religious group representation in the primary care sample, we used validated surname lists (Lauderdale, 2006; Lauderdale and Kestenbaum, 2000; Sheskin, 1998) to create four strata and oversampled from these strata. We sampled a) 121 primary care physicians with typical Asian surnames (Chinese, Japanese, Filipino, Korean, Asian Indian, Vietnamese), b) 171 primary care physicians with typical Arabic surnames, c) 86 primary care physicians with typical Jewish surnames, and d) 1126 additional primary care physicians (from all those whose surnames were not on one of these ethnic lists). The psychiatrist sample was not sufficiently large to warrant oversampling by ethnic surname. Physicians received up to three separate mailings of the questionnaire. The first
The Journal of Nervous and Mental Disease • Volume 203, Number 2, February 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
The Journal of Nervous and Mental Disease • Volume 203, Number 2, February 2015
mailing included a $20 bill and the third offered an additional $30 for participation. Data were double keyed, cross-compared, and corrected against the original questionnaires. The study was approved by the University of Chicago institutional review board. The primary outcome measures were physicians' beliefs on two statements: “Physicians too often treat normal sadness (or normal worry and stress) as if it were a medical illness” and “Undertreatment of depression (or anxiety disorders) is a more pressing problem than overtreatment of normal sadness.” The survey also asked respondents how much depressed or anxious patients would benefit from paying more attention to their relationships, paying more attention to their spiritual lives, and taking antidepressant or antianxiety medication. The exact wordings and response categories are presented in Table 2. To test whether physicians' views were associated with their religious characteristics, we asked physicians to self-report their religious affiliation and attendance at religious services. Affiliation was classified as no affiliation, Hindu, Jewish, Muslim, Catholic (includes 27 Eastern Orthodox Christians), Evangelical Protestant, non-Evangelical Protestant, and other religion. Attendance was classified as never, once a month or less, and twice a month or more. Other covariates included physician sex, race/ethnicity, and geographic region.
Statistical Analysis Stratum weights for the primary care sample were calculated to account for oversampling from the ethnic surname strata (the design weight). We also created a poststratification adjustment weight to correct for a slightly higher response rate among US medical school graduates (65% response) versus international medical school graduates (56% response; p = 0.002) and among physicians whose roles are primarily teaching or “other” (75% response, 103/138) versus office based, hospital based, research, administrative, or unclassified (62% response, 793/1288; p = 0.004). Weights were the inverse probability of a person with the relevant characteristic being in the final data set. The final weight for each case/respondent was the product of the design weight and the poststratification adjustment weight. This enabled us to adjust for sample stratification and variable response rates to generate estimates for the population of US primary care physicians. Weights were not calculated for the psychiatrist sample because no disproportionate sampling by name strata was performed and because response rates for background variables from the Masterfile did not differ significantly. We first generated estimates of the prevalence of each belief for all US primary care physicians and psychiatrists. We then dichotomized responses to each question, trying to divide responses as close to the middle as possible. This resulted in dividing physicians into those who “agree somewhat or strongly” (versus disagree strongly or somewhat) and those who believe that patients would benefit “a lot” (versus not at all, a little, and somewhat). We then used multivariable logistic regression to look for associations with physician specialty (primary care versus psychiatry), religious affiliation, and attendance at religious services. We measured whether physicians' beliefs about overtreatment and undertreatment were associated with predictions of whether patients would benefit “a lot” from each intervention. All multivariable logistic regression models adjusted for physician sex, region, and race/ethnicity. Bonferroni corrected alpha was used to adjust for multiple comparisons.
RESULTS Response The response rate was 63% (896/1427) for primary care physicians and 64% (312/487) for psychiatrists, after excluding 77 primary care physicians and 25 psychiatrists who had invalid addresses or were no longer practicing. The response rate for primary care physicians varied by stratum: it was 53% (85/162) among those with Arabic surnames, 56% (63/112) among those with Asian surnames, 70%
Overtreating Depression and Anxiety
(59/84) among those with Jewish surnames, and 64% (689/1069) among the remaining physicians. Response rates did not differ by age, sex, region, or board certification. Respondent characteristics are summarized in Table 1.
Beliefs About Depression Two thirds of primary care physicians (67%) and psychiatrists (62%) agreed somewhat or strongly that physicians too often treat normal sadness as if it were a medical illness (difference not significant). Psychiatrists (87%) were more likely than primary care physicians (77%) to believe that undertreatment of depression is a more pressing TABLE 1. Respondent Demographic Characteristics Primary Care Physicians Psychiatrists Sex Female Male Race White, non-Hispanic Black, non-Hispanic Asian Hispanic/Latino Other Age 25–36 yrs 37–44 yrs 45–53 yrs 54–65 yrs Region South Northeast Midwest West Immigration history US born Immigrant Education US medical school International medical school Board certified (yes) Religious affiliation No religion Hindu Jewish Muslim Catholic Evangelical Protestant Non-Evangelical Protestant Other religion Attendance at religious services Never Once a month or less Twice a month or more
N
%
N
%
324 572
36 64
133 179
43 57
625 53 142 41 22
71 6 16 5 2
198 23 64 17 8
64 7 21 5 3
226 224 225 221
25 25 25 25
80 42 92 98
26 13 29 31
295 198 216 187
33 22 24 21
89 102 61 60
29 33 30 19
637 249
72 28
214 96
69 31
678 218 639
76 24 71
220 92 187
71 29 60
96 42 97 60 212 95 227 39
11 5 11 7 24 11 26 4
48 24 41 8 68 20 71 27
16 8 13 3 22 7 23 9
118 413 338
14 48 39
53 162 94
17 52 30
Note. For primary care physicians, mean (SD) age is 44.7 (10.2) years (range, 25–65 years). For psychiatrists, mean (SD) age is 46.5 (10.9) years (range, 26–65 years). Percentages are not adjusted for survey design. They reflect the percentages in our sample.
© 2015 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.jonmd.com
121
Lawrence et al.
The Journal of Nervous and Mental Disease • Volume 203, Number 2, February 2015
TABLE 2. Overall Beliefs About Treating Depression, by Specialty Primary Care Physicians To what degree do you agree or disagree with the following statements? Physicians too often treat normal sadness as if it were a medical illness. Disagree strongly Disagree somewhat Agree somewhat Agree strongly Undertreatment of depression is a more pressing problem than overtreatment of normal sadness. Disagree strongly Disagree somewhat Agree somewhat Agree strongly In general, how much would people with depression benefit from each of the following? Paying more attention to their relationships Not at all A little Somewhat A lot Paying more attention to their spiritual life Not at all A little Somewhat A lot Taking antidepressant medications Not at all A little Somewhat A lot
Psychiatrists
OR (95% CI)
N
%a
N
%a
49 240 450 144
6 27 51 16
37 80 131 62
12 26 42 20
26 173 430 255
3 20 49 28
6 34 127 142
2 11 41 46
7 91 338 452
Psychiatrists’ and Primary Care Physicians’ Beliefs About Overtreatment of Depression and Anxiety Ryan E. Lawrence, MD, MDiv,* Kenneth A. Rasinski, PhD,† John D. Yoon, MD,‡ and Farr A. Curlin, MD§ Abstract: Critics say that physicians overdiagnose and overtreat depression and anxiety. We surveyed 1504 primary care physicians (PCPs) and 512 psychiatrists, measuring beliefs about overtreatment of depression and anxiety and predictions of whether persons would benefit from taking medication, investing in relationships, and investing in spiritual life. A total of 63% of PCPs and 64% of psychiatrists responded. Most agreed that physicians too often treat normal sadness as a medical illness (67% of PCPs and 62% of psychiatrists) and too often treat normal worry and stress as a medical illness (59% of PCPs, 55% of psychiatrists). Physicians who agreed were less likely to believe that depressed or anxious people would benefit “a lot” from taking an antidepressant (36% vs. 58% of PCPs) or antianxiety medication (25% vs. 42% of PCPs, 42% vs. 57% of psychiatrists). Most PCPs and psychiatrists believe that physicians too often treat normal sadness and worry as a medical illness. Key Words: Depression, Anxiety, Antidepressant, Anxiolytic, Religion
Within this debate, some have argued from explicitly religious frameworks to critique what they perceive as the medicalization of normal emotional problems and life stress (DeMoss, 2006). Jay Adams, in his controversial yet popular (over 30 printings) book Competent to Counsel, argued that clergy, not psychiatrists, are most qualified to help people address emotional problems and life stress (Adams, 1970). Continuing in this tradition, David Powlison, director of the Christian Counseling & Educational Foundation, wrote: Biopsychiatry will cure a few things… But in the long run, unwanted and unforeseen side effects will combine with vast disillusionment. The gains will never live up to the promises. And the lives of countless people, whose normal life problems are now being medicated, will not be qualitatively changed and redirected (Powlison, 2010, p. 266).
(J Nerv Ment Dis 2015;203: 120–125)
A
fter fluoxetine—the first selective serotonin reuptake inhibitor (SSRI)—was introduced in the United States in 1988, the number of physician visits for depression increased dramatically (10.99 million in 1985, 20.43 million in 1993–1994). Visits involving an antidepressant prescription similarly increased, and this growth can be entirely accounted for by the use of SSRIs (Pincus et al., 1988). Subsequently, a debate emerged over whether depression is overdiagnosed and overmedicated. Some argue that current diagnostic criteria medicalize normal sadness (Horwitz and Wakefield, 2007; Parker, 2007), whereas others suggest that increased diagnosis and treatment are beneficial because they reduce suicides, reduce stigma, and increase productivity (Hickie, 2007). A similar debate surrounds the diagnosis and treatment of anxiety—the most common mental health disorder internationally (The WHO World Mental Health Survey Consortium, 2004). Some argue that a degree of anxiety is a normal response to life stressors (Abrams, 2012) and that “The hasty, overmedicated handling of anxiety is a direct result of DSM-III and the large number of new disorders it created” (Lane, 2008, p. 139). At the same time, however, primary care data from the 1990s showed that half of patients with significant anxiety were undiagnosed and untreated and that patients with untreated anxiety exhibit worse physical functioning, social functioning, emotional problems, general mental health, energy, pain, and general health perceptions and had more limitations in daily activity caused by physical health (Fifer et al., 1994).
*Department of Psychiatry, Columbia University Medical Center and the New York State Psychiatric Institute, New York, NY; †Program on Medicine and Religion and ‡Department of Medicine and the MacLean Center for Clinical Medical Ethics, University of Chicago, Illinois; and §Trent Center for Bioethics, Humanities, and History of Medicine, and The Divinity School, Duke University, Durham, NC. Send reprint requests to Ryan E. Lawrence, MD, MDiv, Department of Psychiatry, Columbia University Medical Center, New York, Presbyterian Hospital, 9 Garden North, 177 Fort Washington Avenue, New York, NY 10032. E-mail: [email protected]. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0022-3018/15/20302–0120 DOI: 10.1097/NMD.0000000000000247
120
www.jonmd.com
The current study attempts to describe relevant attitudes regarding the medical diagnosis and treatment of depression and anxiety among US psychiatrists and primary care physicians and the extent to which such attitudes are associated with physician specialty and physician religiosity. We tested several hypotheses. First, we expected physicians to be divided about the medicalization of depression and anxiety, with large proportions believing that normal sadness and worry are too often treated as medical conditions. Second, we predicted that large proportions would also believe that undertreatment of depression and anxiety are greater problems than overtreatment, with psychiatrists being more likely than primary care physicians to believe this. Third, we tested whether physicians who believe that normal sadness and worry are too often treated as a medical problem will be less likely to endorse a benefit of using antidepressants or antianxiety medication and more likely to endorse a benefit of patients paying more attention to their relationships and spiritual lives. Fourth, we predicted that physicians who are more religious will be more likely to believe that normal sadness and worry are overly medicalized and less likely to believe that patients will significantly benefit from using medication.
METHODS Between September 2009 and June 2010, we mailed a confidential self-administered questionnaire to a stratified random sample consisting of 1504 US primary care physicians and 512 US psychiatrists 65 years old or younger. The sample was generated from the American Medical Association Physician Masterfile, a database intended to include all practicing US physicians. To increase minority religious group representation in the primary care sample, we used validated surname lists (Lauderdale, 2006; Lauderdale and Kestenbaum, 2000; Sheskin, 1998) to create four strata and oversampled from these strata. We sampled a) 121 primary care physicians with typical Asian surnames (Chinese, Japanese, Filipino, Korean, Asian Indian, Vietnamese), b) 171 primary care physicians with typical Arabic surnames, c) 86 primary care physicians with typical Jewish surnames, and d) 1126 additional primary care physicians (from all those whose surnames were not on one of these ethnic lists). The psychiatrist sample was not sufficiently large to warrant oversampling by ethnic surname. Physicians received up to three separate mailings of the questionnaire. The first
The Journal of Nervous and Mental Disease • Volume 203, Number 2, February 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
The Journal of Nervous and Mental Disease • Volume 203, Number 2, February 2015
mailing included a $20 bill and the third offered an additional $30 for participation. Data were double keyed, cross-compared, and corrected against the original questionnaires. The study was approved by the University of Chicago institutional review board. The primary outcome measures were physicians' beliefs on two statements: “Physicians too often treat normal sadness (or normal worry and stress) as if it were a medical illness” and “Undertreatment of depression (or anxiety disorders) is a more pressing problem than overtreatment of normal sadness.” The survey also asked respondents how much depressed or anxious patients would benefit from paying more attention to their relationships, paying more attention to their spiritual lives, and taking antidepressant or antianxiety medication. The exact wordings and response categories are presented in Table 2. To test whether physicians' views were associated with their religious characteristics, we asked physicians to self-report their religious affiliation and attendance at religious services. Affiliation was classified as no affiliation, Hindu, Jewish, Muslim, Catholic (includes 27 Eastern Orthodox Christians), Evangelical Protestant, non-Evangelical Protestant, and other religion. Attendance was classified as never, once a month or less, and twice a month or more. Other covariates included physician sex, race/ethnicity, and geographic region.
Statistical Analysis Stratum weights for the primary care sample were calculated to account for oversampling from the ethnic surname strata (the design weight). We also created a poststratification adjustment weight to correct for a slightly higher response rate among US medical school graduates (65% response) versus international medical school graduates (56% response; p = 0.002) and among physicians whose roles are primarily teaching or “other” (75% response, 103/138) versus office based, hospital based, research, administrative, or unclassified (62% response, 793/1288; p = 0.004). Weights were the inverse probability of a person with the relevant characteristic being in the final data set. The final weight for each case/respondent was the product of the design weight and the poststratification adjustment weight. This enabled us to adjust for sample stratification and variable response rates to generate estimates for the population of US primary care physicians. Weights were not calculated for the psychiatrist sample because no disproportionate sampling by name strata was performed and because response rates for background variables from the Masterfile did not differ significantly. We first generated estimates of the prevalence of each belief for all US primary care physicians and psychiatrists. We then dichotomized responses to each question, trying to divide responses as close to the middle as possible. This resulted in dividing physicians into those who “agree somewhat or strongly” (versus disagree strongly or somewhat) and those who believe that patients would benefit “a lot” (versus not at all, a little, and somewhat). We then used multivariable logistic regression to look for associations with physician specialty (primary care versus psychiatry), religious affiliation, and attendance at religious services. We measured whether physicians' beliefs about overtreatment and undertreatment were associated with predictions of whether patients would benefit “a lot” from each intervention. All multivariable logistic regression models adjusted for physician sex, region, and race/ethnicity. Bonferroni corrected alpha was used to adjust for multiple comparisons.
RESULTS Response The response rate was 63% (896/1427) for primary care physicians and 64% (312/487) for psychiatrists, after excluding 77 primary care physicians and 25 psychiatrists who had invalid addresses or were no longer practicing. The response rate for primary care physicians varied by stratum: it was 53% (85/162) among those with Arabic surnames, 56% (63/112) among those with Asian surnames, 70%
Overtreating Depression and Anxiety
(59/84) among those with Jewish surnames, and 64% (689/1069) among the remaining physicians. Response rates did not differ by age, sex, region, or board certification. Respondent characteristics are summarized in Table 1.
Beliefs About Depression Two thirds of primary care physicians (67%) and psychiatrists (62%) agreed somewhat or strongly that physicians too often treat normal sadness as if it were a medical illness (difference not significant). Psychiatrists (87%) were more likely than primary care physicians (77%) to believe that undertreatment of depression is a more pressing TABLE 1. Respondent Demographic Characteristics Primary Care Physicians Psychiatrists Sex Female Male Race White, non-Hispanic Black, non-Hispanic Asian Hispanic/Latino Other Age 25–36 yrs 37–44 yrs 45–53 yrs 54–65 yrs Region South Northeast Midwest West Immigration history US born Immigrant Education US medical school International medical school Board certified (yes) Religious affiliation No religion Hindu Jewish Muslim Catholic Evangelical Protestant Non-Evangelical Protestant Other religion Attendance at religious services Never Once a month or less Twice a month or more
N
%
N
%
324 572
36 64
133 179
43 57
625 53 142 41 22
71 6 16 5 2
198 23 64 17 8
64 7 21 5 3
226 224 225 221
25 25 25 25
80 42 92 98
26 13 29 31
295 198 216 187
33 22 24 21
89 102 61 60
29 33 30 19
637 249
72 28
214 96
69 31
678 218 639
76 24 71
220 92 187
71 29 60
96 42 97 60 212 95 227 39
11 5 11 7 24 11 26 4
48 24 41 8 68 20 71 27
16 8 13 3 22 7 23 9
118 413 338
14 48 39
53 162 94
17 52 30
Note. For primary care physicians, mean (SD) age is 44.7 (10.2) years (range, 25–65 years). For psychiatrists, mean (SD) age is 46.5 (10.9) years (range, 26–65 years). Percentages are not adjusted for survey design. They reflect the percentages in our sample.
© 2015 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.jonmd.com
121
Lawrence et al.
The Journal of Nervous and Mental Disease • Volume 203, Number 2, February 2015
TABLE 2. Overall Beliefs About Treating Depression, by Specialty Primary Care Physicians To what degree do you agree or disagree with the following statements? Physicians too often treat normal sadness as if it were a medical illness. Disagree strongly Disagree somewhat Agree somewhat Agree strongly Undertreatment of depression is a more pressing problem than overtreatment of normal sadness. Disagree strongly Disagree somewhat Agree somewhat Agree strongly In general, how much would people with depression benefit from each of the following? Paying more attention to their relationships Not at all A little Somewhat A lot Paying more attention to their spiritual life Not at all A little Somewhat A lot Taking antidepressant medications Not at all A little Somewhat A lot
Psychiatrists
OR (95% CI)
N
%a
N
%a
49 240 450 144
6 27 51 16
37 80 131 62
12 26 42 20
26 173 430 255
3 20 49 28
6 34 127 142
2 11 41 46
7 91 338 452
