Mojtabai and Olfson's National Patterns in Antidepressant Treatment by Psychiatrists and General Med

Mojtabai and Olfson's National Patterns in Antidepressant Treatment by Psychiatrists and General Med

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: National Patterns in Antidepressant Treatment National Patterns in Antidepressant Treatment by Psychiatrists and General Medical Providers: Results From the National Comorbidity Survey Replication Ramin Mojtabai, M.D., Ph.D., M.P.H.; and Mark Olfson, M.D., M.P.H. Background: Primary care physicians, rather than psychiatrists, prescribe a majority of psychotropic medications in the United States. However, past research has shown significant differences in psychopharmacologic treatment practices of these 2 groups of physicians. The objective of this study was to compare patient characteristics and treatment patterns of adults in the United States treated with antidepressant medications by psychiatrists and other medical providers. Method: Data from the National Comorbidity Survey Replication (February 2001–April 2003) were used to compare characteristics of adults (aged ≥ 18 years) prescribed antidepressants by psychiatrists (N = 255) or other medical providers (N = 673). The treatment groups were also compared with respect to presenting problem, antidepressant type and dose, and continuity of treatment. Results: Approximately 1 in 10 adults (10.5%) were treated with an antidepressant in the past year, usually by a general medical provider (73.6%). Compared with those treated by psychiatrists, adults treated by general medical providers were significantly more likely to be at least 65 years of age and to reside in a nonurban area. By contrast, those treated by psychiatrists were significantly more likely to be male, to report significant distress, to present with serious mood or anxiety symptoms, and to meet DSM-IV criteria for mood and anxiety disorders. Individuals treated by psychiatrists typically received higher doses of medications, were less likely to stop the medication before 30 days, and were more likely to continue 90 days or longer. Conclusions: Most adults treated with antidepressants receive the medication from general medical providers. In comparison with adults treated by psychiatrists, those treated by general medical providers are less likely to meet the criteria for mood or anxiety disorders or to continue medication beyond the first month. Quality improvement initiatives in general medical settings should focus on better targeting and continuity of antidepressant medications. (J Clin Psychiatry 2008;69:1064–1074) J Clin Psychiatry 69:7, July 2008 Received June 24, 2007; accepted Nov. 7, 2007. From the Department of Psychiatry, Beth Israel Medical Center (Dr. Mojtabai), and the Department of Psychiatry, College of Physicians and Surgeons, Columbia University and New York State Psychiatric Institute (Dr. Olfson), New York, N.Y. This study was completed at the Department of Psychiatry, Beth Israel Medical Center, New York, N.Y. The National Comorbidity Survey Replication is supported by a grant from the National Institute of Mental Health (U01-MH60220), with supplemental support from grants from the National Institute on Drug Abuse, the Substance Abuse and Mental Health Services Administration, the Robert Wood Johnson Foundation (044708), and the John W. Alden Trust (principal investigator: Ronald C. Kessler, Ph.D.). Dr. Olfson’s work was supported in part by a grant from the Agency for Healthcare Research and Quality (U18-HS016097). Dr. Mojtabai has received research funding and consultant fees from Bristol-Myers Squibb. Dr. Olfson has received research funding from Eli Lilly, Bristol-Myers Squibb, and Janssen and has served as a paid consultant to Pfizer and McNeil. Corresponding author and reprints: Ramin Mojtabai, M.D., Ph.D., M.P.H., Department of Psychiatry, Beth Israel Medical Center, First Ave. at 16th St., New York, NY 10003 (e-mail: [email protected]). T here has been a dramatic increase in rates of outpatient treatment of depression and other common mental disorders in recent years.1,2 Several factors likely contributed to this trend including introduction of newer medications with fewer side effects along with expanding indications for these medications, growth of direct to consumer advertising by pharmaceutical manufacturers,3 public awareness campaigns,4 new social policies,5 and publicity concerning depression and antidepressant medications.6 This development also paralleled the growing role of the general medical sector in providing outpatient care for common mental disorders.2 Between the early 1990s and early 2000s, the percentage of individuals who sought mental health care from the general medical sector increased from 31.5% of all of those who sought mental health care to 49.6%, while the share of individuals seeking treatment from psychiatrists increased from 19.6% to 25.8%.2 This trend is likely partly responsible for the finding that the growth in the use of antidepressant medications was much greater than in the use of psychotherapy.1 Primary care physicians, rather than psychiatrists, have historically prescribed a majority of psychotropic medications in the United States.7,8 However, past research has found important differences in the quality and duration of psychopharmacologic management provided by these 2 groups of health professionals.9–14 Specifically, mental health care provided in general medical settings was PSYCHIATRIST.COM 1065 1064 Mojtabai and Olfson somewhat less likely than that provided in specialty mental health care settings to be guideline consistent and to extend for an adequate time period.9–11 Furthermore, the patients who sought care in general medical settings were different in important characteristics from those who sought care in specialized mental health care settings.15,16 However, past research on differences in treatment patterns across the 2 groups of providers mostly predates the recent expansion in outpatient treatment of common mental disorders. This study compared more recent patterns of antidepressant medication treatment by psychiatrists and general medical providers in a representative sample of the U.S. adult general population. More specifically, we compared individuals treated with antidepressants by general medical providers and psychiatrists with respect to sociodemographic characteristics, psychiatric diagnoses, presenting complaints, type of medications, dose ranges, continuity of treatment, adherence to medications, and perceptions of effectiveness of medication treatment. METHOD Sample Data were drawn from the National Comorbidity Survey Replication (NCS-R),2 a nationally representative cross-sectional survey of households in the 48 coterminous United States. The NCS-R was administered to 9282 individuals aged ≥ 18 years between February 2001 and April 2003 (response rate = 70.9%). The NCS-R interview included 2 parts administered in 1 interview session. Part 1 comprised the core diagnostic assessment module and the pharmacoepidemiology module administered to all participants. Part 2 included further diagnostic assessments as well as questions about current symptoms and was administered to 5692 part 1 participants who met lifetime criteria for any core disorder plus a probability subsample of other participants. Interviews were conducted in person. Informed consent was obtained before the interview. The human subjects committees of Harvard Medical School, Boston, Mass., and the University of Michigan, Ann Arbor, Mich., approved these recruitment and consent procedures. Secondary analyses of the data presented in this article were approved by the institutional review board of the Beth Israel Medical Center, New York, N.Y. Assessments Antidepressant medication use was assessed by presenting the participants with a list of specific psychotropic medications and asking them which medicines on the list they had taken in the past 12 months for problems with their “emotions, nerves, mental health, substance use, energy, concentration, sleep, or ability to cope with stress.” The list included 215 generic and proprietary names for 1065 1066 commonly used psychotropic medications. For this study, the analyses were limited to antidepressant medications. The participants were asked to check the medication bottle for the exact name of the medication. For the first 3 medication mentions, participants were asked who prescribed that medication—a psychiatrist, a general or family doctor, some other medical doctor, or some other health professional. For this report, participants were classified into 2 nonoverlapping categories: participants who took an antidepressant prescribed by a psychiatrist whether or not they also took antidepressants prescribed by a general medical provider and those who took an antidepressant prescribed by a general medical provider only (a general or family doctor, some other medical doctor, or some other health professional). Psychiatric diagnoses were ascertained using the Composite International Diagnostic Interview,17 a layadministered structured interview that provides psychiatric diagnoses based on the DSM-IV18 criteria. For this study, analyses were limited to lifetime diagnoses of common mood and anxiety disorders assessed in part 1 of the NCS-R and for which treatment with an antidepressant medication is indicated. Psychological distress was ascertained using the K6 screening instrument.19 The 6 items of the K6 probe how often the participant has felt nervous, restless, hopeless, worthless, extremely sad, or that “everything was an effort” during a 1-month period in the past 12 months when the participant was “the most depressed, anxious, or emotionally stressed.” Each K6 item is rated on a scale ranging from “none of the time” (0) to “all of the time” (4). Thus, K6 scores can range from 0 to 24. The K6 has been shown to have a high internal consistency reliability (Cronbach’s α = 0.89) and concurrent validity against the Structured Clinical Interview for DSM-IV.19 Consistent with past research, we used a cutoff point of < 13 vs. ≥ 13 to identify participants with serious psychological distress.19 Presenting problems were assessed for the first 3 medication mentions by giving the participants a list of common presenting problems and asking them to identify problems for which they had taken the medication. For this study, presenting problems were categorized into mood and anxiety symptoms, physical symptoms, cognitive symptoms, role functioning problems, and alcohol/ drug problems. Sociodemographic variables included age (18–24, 25–34, 35–44, 45–54 years), gender, race/ethnicity (nonHispanic white, Hispanic, non-Hispanic black, other), marital status (married or living as married, divorced/ separated/widowed, never married), education (0–11, 12, 13–15, ≥ 16 years), family income compared to U.S. federal poverty level for 2001 (low = < 1.5, lowaverage = 1.5 to < 3, high-average = 3 to < 6, and high ≥ 6), urbanicity (metropolitan counties with ≥ 1,000,000 PSYCHIATRIST.COM J Clin Psychiatry 69:7, July 2008 National Patterns in Antidepressant Treatment population, other urban metropolitan counties with < 1,000,000 population, and nonurban counties, which include all nonmetropolitan counties),2 and insurance status (any vs. no insurance, and for those with any insurance, whether preauthorization is required for specialist referral or not). Questions about income and preauthorization were included in part 2 of the NCS-R. Medication dose was ascertained by asking participants about their frequency of use and doses of each medication. Daily dose was computed by multiplying frequency of use by milligram amount in each dose of each medication. In addition, doses were compared with dose ranges recommended in the American Psychiatric Association’s Practice Guideline for the Treatment of Patients With Major Depressive Disorder.20 Medication continuity and adherence were assessed separately for each medication. Participants were asked how many days in the past 12 months they took the medication and if they were still taking the medication or had stopped. For those who were still taking the medication, participants were asked on how many days out of the past 30 days they took the medication. Consistent with past research,21 continuity of antidepressant use was operationalized in 2 ways: (1) stopping antidepressants before 30 days of use and (2) using antidepressants for at least 90 days. For the first 3 medication mentions, participants were also asked, “Think of a typical month when you took (the medication’s name) in the past 12 months. How many days out of 30 did you typically either forget to take it or take less of it than you were supposed to take?” The NCS-R did not obtain information on time since start of medication. Furthermore, for participants who reported taking more than 1 antidepressant medication, the order in which these medications had been prescribed was not ascertained. Perceived effectiveness of the antidepressant medication treatment was assessed by asking the participants, “Overall, how effective was (the medication name) in doing the things you expected it to do—very, somewhat, not very, or not at all effective?” The ratings were coded from 0 (not at all effective) to 3 (very effective). Data Analysis Sociodemographic characteristics, diagnoses, presenting problems, type of medications, and characteristics of medication treatment such as medication dose compared to the minimum recommended dose and continuity were compared across participants prescribed antidepressant medications by psychiatrists and general medical providers using bivariate logistic regression analyses. Dose ranges for individual medications were also computed and compared across psychiatrists and general medical providers using ordinal logistic regression models that accommodate nonnormally distributed data such as medication doses. J Clin Psychiatry 69:7, July 2008 The relationship of type of prescribing professional with characteristics of medication treatment, including continuity of treatment, adherence, and medication dose, compared with the minimum recommended antidepressant dose for all antidepressant medications together was further assessed using multivariate logistic regression analyses that adjusted for potentially confounding variables. Medication treatment characteristics found in bivariate analyses to be different across groups at a p < .05 level were included in these analyses. These analyses adjusted for sociodemographic and clinical characteristics that differed in the bivariate analysis at a p < .25 level.22 The NCS-R used a complex stratified sampling design. Survey weights and design elements were included in the analyses to adjust for their effects and to make samples representative of the U.S. population. All percentages reported are weighted by the NCS-R sampling weights. STATA 9.223 software was used for all analyses. A p < .05 level was adopted as the threshold for judging the significance of statistical tests. RESULTS Overall, 975 (10.5%) of 9282 NCS-R participants reported having taken an antidepressant medication in the past year. For 928 (96.6%) of these participants, information was available concerning the prescribing health care professional. Of these, 237 (24.7%) reported taking antidepressants prescribed by a psychiatrist only, 18 (1.7%) by a psychiatrist as well as a general medical provider, and 673 (73.6%) by a general medical provider only. For the present analyses, the group that reported taking antidepressants prescribed both by a psychiatrist and a general medical provider was combined with the group that reported taking antidepressants prescribed by a psychiatrist only. The analytic groups thus comprised 255 participants prescribed antidepressants by psychiatrists (or psychiatrists and general medical providers) and 673 participants prescribed antidepressants by general medical providers only. There were minor sociodemographic differences between these 2 groups (Table 1). As compared with participants treated by general medical providers, participants treated by psychiatrists were more likely to be male, to be from the “other” racial/ethnic group, and to have never married. In relation to participants treated with antidepressants by psychiatrists, those treated by general medical providers were more likely to be at least 65 years of age and to reside in a nonurban area (Table 1). Differences with regard to psychiatric diagnoses and psychological distress were more marked (Table 2). Compared with participants treated with antidepressants by general medical providers, those who were treated by psychiatrists were more likely to meet criteria for DSM-IV diagnoses of major depressive disorder, bipolar disorder, panic disorder, social phobia, and posttraumatic stress PSYCHIATRIST.COM 1067 1066 Mojtabai and Olfson Table 1. Sociodemographic Characteristics of National Comorbidity Survey Replication Participants Treated With Antidepressants Prescribed by Psychiatrists and by General Medical Providersa Prescribed by Psychiatrists (N = 255) N % Prescribed by General Medical Providers (N = 673) N % Bivariate Binary Logistic Regression for Comparison of Groups Odds Ratio 95% CI p Value Characteristic Age, yb 18–24 20 9.0 50 8.8 1.00 Reference 25–34 46 16.1 111 13.2 1.19 0.57 to 2.47 35–44 64 26.3 149 21.9 1.17 0.60 to 2.31 45–54 73 29.0 161 26.3 1.08 0.54 to 2.14 55–64 37 12.9 98 13.8 0.91 0.40 to 2.08 ≥ 65 15 6.6 103 16.1 0.40 0.20 to 0.80 Genderc Female 176 63.6 507 73.5 1.00 Reference Male 79 36.4 166 26.5 1.59 1.15 to 2.19 Race/ethnicityc Non-Hispanic white 207 83.2 572 85.8 1.00 Reference Hispanic 15 6.0 44 8.0 0.78 0.34 to 1.81 Non-Hispanic black 19 6.3 40 4.3 1.53 0.82 to 2.86 Other 14 4.5 17 2.0 2.36 1.06 to 5.26 Marital statusd Married/living as married 117 44.6 373 54.3 1.00 Reference Divorced/separated/widowed 85 30.8 190 28.1 1.34 0.82 to 2.17 Never married 53 24.6 110 17.6 1.70 1.04 to 2.78 Education, ye 0–11 33 12.7 105 16.8 1.00 Reference 12 75 31.7 195 32.3 1.30 0.72 to 2.34 13–15 81 30.5 225 30.2 1.33 0.69 to 2.58 ≥ 16 66 25.1 148 20.8 1.60 0.79 to 3.23 Family income (part 2 sample)f Low 68 28.8 117 25.3 1.00 Reference Low average 48 18.8 135 21.9 0.75 0.37 to 1.53 High average 68 29.7 199 33.3 0.78 0.45 to 1.35 High 54 22.6 118 19.5 1.01 0.53 to 1.93 g,h Urbanicity Metropolitan 110 41.6 245 31.8 1.00 Reference Other urban 96 32.8 253 31.0 0.81 0.56 to 1.16 Nonurban 49 25.6 175 37.2 0.52 0.35 to 0.77 Chronic physical conditions (part 2 sample)i Present 154 63.5 362 64.0 1.00 Reference Absent 93 36.5 231 36.0 0.98 0.71 to 1.36 Health insurance status (part 2 sample) Any insurance 218 88.4 551 91.8 1.00 Reference No insurance 28 11.6 42 8.2 1.47 0.76 to 2.84 Preauthorization for referralj Not required 77 38.9 216 41.8 1.00 Reference Required 135 61.1 315 58.2 1.13 0.74 to 1.72 a All percentages are weighted by the National Comorbidity Survey Replication sampling weights. b Overall test: F = 2.77; df = 5,38; p = .031. c Overall test: F = 2.30; df = 3,40; p = .092. d Overall test: F = 2.35; df = 2,41; p = .108. e Overall test: F = 0.72; df = 3,40; p = .547. f Overall test: F = 0.27; df = 3,40; p = .848. g “Metropolitan” consists of large, core metropolitan counties with a population ≥ 1,000,000; “other urban” consists of medium and lesser metropolitan counties with a population < 1,000,000; and “nonurban” consists of all nonmetropolitan counties. h Overall test: F = 6.94; df = 2,41; p = .003. i Chronic medical conditions included arthritis, hypertension, diabetes, heart disease, asthma, and other lung disease. j This question was asked to people who reported having any type of health insurance. Symbol: … = no data. disorder. There was no significant difference between the groups in the prevalence of dysthymia or generalized anxiety disorder. Participants who took antidepressants prescribed by psychiatrists experienced more psychological distress as measured by the K6 (Table 2). Consistent with differences in diagnoses, there were also significant group differences with regard to present- 1067 1068 … .630 .634 .832 .823 .011 … .006 … .554 .176 .037 … .232 .034 … .383 .387 .189 … .425 .366 .963 … .242 .002 … .898 … .248 … .571 ing problems across the 2 groups of participants (Table 2). Compared with participants who were treated by general medical providers, those who were treated by psychiatrists were more likely to present with complaints of mood and anxiety symptoms, including sadness, manic mood, anger/irritability, panic, suicidal thoughts, and poor concentration. PSYCHIATRIST.COM J Clin Psychiatry 69:7, July 2008 National Patterns in Antidepressant Treatment Table 2. Diagnoses, Psychological Distress, and Presenting Complaints of National Comorbidity Survey Replication Participants Treated With Antidepressants Prescribed by Psychiatrists and by General Medical Providersa Prescribed by Psychiatrists (N = 255) N % Prescribed by General Medical Providers (N = 673) N % Characteristic Lifetime psychiatric diagnoses Mood disorders Major depressive disorder 120 46.4 243 35.3 Dysthymia 28 10.9 46 7.4 Bipolar disorder 42 16.9 36 5.4 Any mood disorder 166 64.8 283 41.7 Anxiety disorders Panic disorder 64 23.8 96 14.7 Generalized anxiety disorder 39 14.6 95 14.1 Social phobia 105 40.2 165 22.8 Posttraumatic stress disorder 71 26.0 102 16.3 Any anxiety disorder 121 43.7 226 34.8 Any mood or anxiety disorder 202 76.8 385 57.6 Psychological distress (K6 score)b 0–12 161 63.5 507 82.1 13–24 86 36.5 105 17.9 Presenting problems Mood and anxiety symptoms Sadness/depression/crying 196 78.2 392 58.0 Manic mood 22 9.3 20 2.6 Anger or irritability 38 16.9 54 7.8 Nerves/anxiety 89 36.0 237 36.1 Panic 37 14.9 63 8.4 Suicidal thoughts 14 5.4 11 1.7 Physical symptoms Low energy 27 10.4 56 7.9 Poor appetite 10 4.5 18 2.9 Poor sleep 57 22.3 136 19.0 Little or no sexual functioning 8 3.7 9 1.3 Physical pain 5 3.5 49 7.4 Cognitive symptom Poor concentration 34 14.4 41 5.8 Poor memory 14 5.2 21 2.7 Role functioning problems Marital problems 7 2.8 12 1.8 Not getting along with others 6 1.9 9 1.7 Poor work performance 3 0.9 10 1.4 Alcohol/drug problems 2 1.9 6 0.8 a All percentages are weighted by the National Comorbidity Survey Replication sampling weights. b Part 2 sample. Symbol: … = no data. There were few differences between the groups with regard to type of antidepressant medication prescribed (Table 3). Participants treated with antidepressants by psychiatrists were significantly more likely than those who were treated by general medical providers to be prescribed fluoxetine and significantly less likely to be prescribed amitriptyline. Otherwise, there were no significant differences with regard to type of medication among groups. Selective serotonin reuptake inhibitors (SSRIs) were the most commonly prescribed class of antidepressants overall, prescribed to an equal percentage of participants treated by psychiatrists and general medical providers (Table 3). The percentage of participants treated with tricyclic antidepressants was also very similar across groups (Table 3). Of note, only 1 participant reported having taken a monoamine oxidase inhibitor. J Clin Psychiatry 69:7, July 2008 Bivariate Binary Logistic Regression for Comparison of Groups Odds Ratio 95% CI p Value 1.73 1.62 3.54 2.58 1.27 to 2.34 0.90 to 2.90 2.09 to 5.98 1.81 to 3.67 .001 .105 < .001 < .001 1.84 1.06 2.35 1.94 1.46 2.43 1.16 to 2.92 0.68 to 1.66 1.70 to 3.26 1.34 to 2.80 1.02 to 2.07 1.61 to 3.69 .011 .785 < .001 .001 .036 < .001 1.00 2.64 Reference 1.62 to 4.30 … < .001 2.60 3.91 2.42 1.00 1.90 3.42 1.86 to 3.62 2.60 to 5.88 1.50 to 3.90 0.71 to 1.40 1.29 to 2.80 1.52 to 7.70 < .001 < .001 .001 .979 .002 .004 1.36 1.58 1.22 2.84 0.45 0.77 to 2.39 0.55 to 4.52 0.79 to 1.88 0.75 to 10.72 0.15 to 1.34 .280 .383 .356 .120 .148 2.76 1.98 1.38 to 5.49 0.93 to 4.22 .005 .076 1.59 1.09 0.66 2.45 0.51 to 4.92 0.33 to 3.57 0.15 to 2.96 0.40 to 14.91 .412 .881 .576 .321 However, there were more differences across groups with regard to doses of individual medications. Participants treated by psychiatrists with sertraline, paroxetine, citalopram, and trazodone received higher doses of these medications than those treated by general medical providers (Table 3). The doses of these medications compared with dose ranges recommended in the American Psychiatric Association’s Practice Guideline for the Treatment of Patients With Major Depressive Disorder20 are also depicted in Figure 1, which reveals that a higher percentage of individuals treated by psychiatrists compared with general medical providers received at least the minimum antidepressant dose of these medications or reached the maximum dose. The differences across groups with regard to current status, the number of days that the participant forgot to PSYCHIATRIST.COM 1069 1068 1069 1070 Prescribed by Psychiatrists (N = 255) 25th–75th Median Percentile Prescribed by General Medical Providers (N = 673) 25th–75th Median Percentile .007 .001 .251 .042 … … .793 .503 .312 … … … … … .143 .238 … .044 .689 … 1.32 to 5.13 1.69 to 5.65 0.67 to 4.41 1.04 to 7.35 … … 0.41 to 3.15 0.20 to 23.77 0.23 to 60.08 … … … … … 0.78 to 5.31 0.59 to 7.55 … 1.03 to 7.50 0.03 to 10.98 … Bivariate Ordinal Logistic Regression for Comparison of Groups Odds Ratio 95% CI p Value Doses of Medications (mg/day) Characteristic Medication type SSRIs Sertraline 57 20.5 157 23.3 0.85 0.56 to 1.29 .429 100 50–150 50 50–100 2.61 Paroxetine 45 16.5 133 21.2 0.73 0.51 to 1.05 .085 30 20–40 20 10–30 3.09 Fluoxetine 50 21.2 98 14.9 1.54 1.01 to 2.56 .047 20 20–40 20 20–40 1.72 Citalopram 32 14.7 61 9.8 1.58 0.91 to 2.71 .099 40 20–40 20 20–30 2.77 Fluvoxamine 1 0.1 4 0.8 1.51 0.01 to 3.11 .214 300 … 100 50–100 … Any SSRI 168 66.6 433 66.6 1.00 0.74 to 1.34 1.000 … … … … … Tricyclics Amitriptyline 11 4.5 66 9.5 0.45 0.21 to 0.96 .039 40 25–50 50 25–75 1.14 Doxepin 8 3.0 13 1.4 2.16 0.89 to 5.24 .086 50 25–150 20 20–50 2.15 Nortriptyline 7 2.5 7 0.9 3.04 0.74 to 12.39 .119 50 25–100 37.5 10–50 3.76 b Imipramine 0 0.0 5 0.7 … … … … … 50 25–100 … Clomipramine 2 0.9 0 0.0 …b … … 62.5 25–100 … … … Desipramine 0 0.0 1 0.1 …b … … … … … … … Protriptyline 0 0.0 1 0.3 …b … … … … … … … Any tricyclic 27 10.5 92 12.8 0.80 0.46 to 1.38 .412 … … … … … SNRI/NDRI Bupropion 39 16.0 74 10.7 1.59 0.96 to 2.64 .073 300 250–400 300 150–300 2.03 Venlafaxine 23 8.8 51 7.3 1.23 0.65 to 2.30 .513 150 75–300 150 75–150 2.12 Any SNRI/NDRI 60 23.8 125 17.9 1.43 0.97 to 2.10 .067 … … … … … Other Trazodone 27 10.4 47 6.6 1.64 0.93 to 2.92 .088 100 50–300 50 50–100 2.78 Mirtazapine 7 3.2 7 1.0 3.29 0.91 to 12.24 .069 30 15–30 30 15–30 0.59 b Phenelzine 0 0.0 1 0.1 … … … … … 30 … … a All percentages are weighted by the National Comorbidity Survey Replication sampling weights. b Regression analysis could not be conducted, as 1 group did not have any variability in outcome. Abbreviations: NDRI = norepinephrine dopamine reuptake inhibitor, SNRI = serotonin-norepinephrine reuptake inhibitor, SSRI = selective serotonin reuptake inhibitor. Symbol: … = no data. Prescribed by Psychiatrists (N = 255) N % Medication Types Prescribed by Bivariate Binary General Medical Logistic Regression for Providers (N = 673) Comparison of Groups N % Odds Ratio 95% CI p Value Table 3. Type and Doses of Medications Taken by National Comorbidity Survey Replication Participants Treated With Antidepressants Prescribed by Psychiatrists and by General Medical Providersa Mojtabai and Olfson PSYCHIATRIST.COM J Clin Psychiatry 69:7, July 2008 National Patterns in Antidepressant Treatment N = 29 N = 50 N = 26 N = 43 GEN PSY GEN GEN N = 45 N = 114 PSY GEN N = 50 N = 131 PSY 100 PSY Figure 1. Percentage of Participants in Each Dose Range for Sertraline, Paroxetine, Citalopram, and Trazodone Prescribed by Psychiatrists (PSY) and General Medical Providers (GEN)a 80 Maximum antidepressant dose or more Minimum antidepressant dose or more but less than maximum Less than minimum antidepressant dose % 60 40 20 0 a Sertraline Paroxetine Citalopram Trazodone Doses were compared with dose ranges recommended in the American Psychiatric Association’s Practice Guideline for the Treatment of Patients With Major Depressive Disorder.20 Table 4. Bivariate Analyses of Medication Use Characteristics of National Comorbidity Survey Replication Participants Treated With Antidepressants Prescribed by Psychiatrists and by General Medical Providersa Characteristic Dose in relation to usual range At or above lower bound Below lower bound Stopped medication before 30 days No Yes Continued medication at least 90 days No Yes Current status Still taking medication Stopped medication Prescribed by Psychiatrists (N = 255) N % Prescribed by General Medical Providers (N = 673) N % Bivariate Binary Logistic and Linear Regression Analyses for Comparison of Groups Odds Ratio 95% CI p Value 210 45 83.9 16.1 501 172 74.9 25.1 1.00 0.68 Reference 0.43 to 1.06 … .085 237 17 93.8 6.2 578 88 87.4 12.6 1.00 0.46 Reference 0.28 to 0.74 … .002 40 214 17.6 82.4 179 487 25.8 74.2 1.00 1.63 Reference 1.06 to 2.50 … .027 177 78 68.6 31.4 450 223 68.0 32.1 1.00 1.03 Reference 0.66 to 1.60 … .895 Mean SE Mean SE βb SE p Value 0.30 –0.74 0.47 .122 0.49 0.03 2.18 0.05 0.59 0.06 .001 .459 No. of days forgot to take medication in a typical month or took less 2.09 0.36 2.83 No. of days taking medication in the past 30 days among those who still take medication 28.06 0.42 25.88 Perceived effectiveness of medicationc 2.41 0.06 2.36 a All percentages are weighted by the National Comorbidity Replication sampling weights. b Unstandardized regression coefficient obtained in linear regression. c Scores range from 0 (not at all effective) to 3 (very effective). Symbol: … = no data. take the medication or took a smaller dose in a typical month, and perceived effectiveness of antidepressants were not statistically significant (Table 4). Participants treated by psychiatrists were more likely to receive medications at or above minimum recommended antidepressant doses (aggregated across all medication types), although this difference was only at a trend level (p = .085) and did not reach a statistically significant level. However, differences across groups with regard to continuity of treatment and adherence did reach a statistically significant level. Participants treated with antidepressants by psychiatrists were less likely to stop the medication J Clin Psychiatry 69:7, July 2008 before 30 days and more likely to continue taking the antidepressant for 90 days or longer. Among participants who continued taking the medication, those treated by psychiatrists reported skipping fewer doses in the past 30 days (Table 4). The significant group differences in continuity of antidepressant treatment persisted in multivariate analyses adjusting for all sociodemographic and clinical variables found to be different across groups at a p < .25 level (Table 5). Participants treated by psychiatrists were significantly less likely to discontinue antidepressant treatment before 30 days and more likely to continue PSYCHIATRIST.COM 1071 1070 Mojtabai and Olfson Table 5. Multivariate Analyses of Medication Use Characteristics of National Comorbidity Survey Replication Participants Treated With Antidepressants Prescribed by Psychiatrists and by General Medical Providersa Characteristic Prescribed by General medical provider Psychiatrist Stopped Medication Before 30 Days Adjusted Odds Ratio 95% CI p Value Continued Medication at Least 90 Days Adjusted Odds Ratio 95% CI p Value Reference 0.36 … 0.18 to 0.69 … .003 Reference 1.91 … 1.10 to 3.31 … .022 Age, y 18–24 25–34 35–44 45–54 55–64 ≥ 65 Reference 1.31 0.63 0.54 0.36 0.46 … 0.41 to 4.16 0.19 to 2.12 0.18 to 1.59 0.10 to 1.27 0.11 to 1.94 … .640 .443 .256 .109 .285 Reference 1.51 2.51 1.99 3.37 4.23 … 0.71 to 3.20 1.27 to 4.98 0.91 to 4.34 1.77 to 6.41 1.80 to 9.98 … .279 .010 .082 < .001 .001 Gender Female Male Reference 1.46 … 0.88 to 2.40 … .139 Reference 0.62 … 0.43 to 0.92 … .017 Race/ethnicity Non-Hispanic white Hispanic Non-Hispanic black Other Reference 1.87 2.28 2.40 … 0.77 to 4.54 0.86 to 6.00 0.79 to 7.28 … .163 .094 .121 Reference 0.56 0.32 0.60 … 0.26 to 1.19 0.15 to 0.66 0.21 to 1.74 … .130 .003 .335 Marital status Married/living as married Divorced/separated/widowed Never married Reference 1.00 1.30 … 0.51 to 1.95 0.62 to 2.73 … .998 .477 Reference 0.98 0.80 … 0.58 to 1.67 0.45 to 1.40 … .949 .418 Urbanicity Metropolitan Other urban Nonurban Reference 0.40 0.52 … 0.20 to 0.80 0.31 to 0.88 … .011 .016 Reference 1.29 1.49 … 0.78 to 2.15 0.91 to 2.42 … .314 .110 1.17 0.53 0.77 1.24 0.66 1.62 0.61 to 2.24 0.17 to 1.69 0.25 to 2.38 0.61 to 2.52 0.32 to 1.37 0.88 to 2.97 .635 .278 .639 .545 .257 .118 1.01 1.42 1.57 1.00 0.99 0.82 0.62 to 1.65 0.73 to 2.77 0.63 to 3.92 0.63 to 1.59 0.63 to 1.57 0.52 to 1.31 .967 .299 .330 .999 .973 .407 Psychological distress (K6 score) 0–12 13–24 Reference 1.81 … 0.99 to 3.32 … .055 Reference 0.73 … 0.45 to 1.18 … .194 Presenting problems Sadness/depression/crying Manic mood Anger or irritability Panic Suicidal thoughts Little or no sexual functioning Physical pain Poor concentration Poor memory 0.65 0.28 2.01 0.58 1.77 …b 2.36 0.84 0.41 0.40 to 1.06 0.06 to 1.28 0.72 to 5.64 0.21 to 1.56 0.37 to 8.61 … 0.95 to 5.85 0.21 to 3.34 0.07 to 2.55 .085 .098 .177 .268 .469 … .063 .798 .329 1.60 3.48 0.50 1.84 1.64 0.81 0.51 0.73 1.06 1.05 to 2.44 1.02 to 11.94 0.23 to 1.07 0.95 to 3.57 0.34 to 7.95 0.18 to 3.68 0.21 to 1.22 0.31 to 1.73 0.31 to 3.66 .031 .047 .073 .069 .531 .775 .125 .466 .921 Medication type Paroxetine Fluoxetine Citalopram Fluvoxamine Amitriptyline Doxepin Nortriptyline Bupropion Trazodone Mirtazapine 0.77 0.52 0.99 …b 0.57 2.44 …b 0.78 1.29 …b 0.33 to 1.81 0.20 to 1.35 0.49 to 2.02 … 0.17 to 1.86 0.77 to 7.79 … 0.37 to 1.66 0.49 to 3.41 … .540 .177 .985 … .342 .127 … .510 .605 … 1.44 0.76 0.48 …b 1.08 0.88 1.67 0.71 0.81 1.65 0.76 to 2.73 0.43 to 1.33 0.21 to 1.10 … 0.58 to 1.99 0.33 to 2.39 0.17 to 16.76 0.39 to 1.28 0.39 to 1.71 0.13 to 20.61 .257 .326 .080 … .812 .799 .657 .248 .579 .692 Psychiatric diagnoses Major depression Dysthymia Bipolar disorder Panic disorder Social phobia Posttraumatic stress disorder a All analyses are weighted by the part 2 sample weights. Variable was excluded from the regression model because it predicted outcome perfectly. Symbol: … = no data. b 1071 1072 PSYCHIATRIST.COM J Clin Psychiatry 69:7, July 2008 National Patterns in Antidepressant Treatment antidepressant treatment for 90 days or longer (Table 5). The analysis for the number of days taking the medication in the past 30 days among participants who continued treatment showed a trend level difference in the multivariate model (unstandardized regression coefficient = 1.65, SE = 0.87, t = 1.90, df = 42, p = .064). DISCUSSION The findings of this study should be interpreted in the context of several limitations. First, information on medication use was based on self-report supplemented with information from medication bottle labels. Physician or pharmacy records were not available. However, past research indicates that depressed patients provide accurate reports of their history of antidepressant medication trials and duration of such trials, especially more recent trials.24 Second, the NCS-R data do not distinguish new episodes of medication treatment from episodes that started more than 1 year ago. This limitation may explain the discrepancy in results between this report and an earlier report that found that 42.4% of patients who initiate antidepressants discontinue the medications within 30 days and only 27.6% continue beyond 90 days.21 Furthermore, it is conceivable that short-time users of antidepressants who are overrepresented in the group of participants treated by general medical providers may have different reasons for stopping antidepressants from the long-term users. Thus, combining the short-term and long-term users might have blurred significant differences between these groups with regard to correlates of stopping medications. Third, the survey asked about the health care professional who prescribed the antidepressant medication in the past year. Some participants might have been started on a medication by a psychiatrist and then continued receiving their prescriptions from a primary care provider or vice versa. Fourth, the crosssectional nature of the data limits causal inferences. For example, participants’ knowledge about insurance plan requirements for preauthorization to access a psychiatrist may be the result of having sought such care rather than a determinant of choice between provider types. Individuals who prefer to receive mental health care from psychiatrists may choose plans with freer access to specialists. Fifth, we examined the prevalence of lifetime DSM-IV diagnoses, as some of the patients who have been receiving medications for a year or longer for maintenance treatment of these disorders might not have experienced significant symptoms meeting diagnostic criteria in the past year. However, the results for 12-month diagnoses were consistent with the results for lifetime diagnoses. Overall, 60.6% of participants treated by psychiatrists compared with 42.5% of those treated by general medical providers met the 12-month diagnoses of mood and anxiety disorders included in this study J Clin Psychiatry 69:7, July 2008 (OR = 2.08, 95% CI = 1.54 to 2.80, p < .001). Finally, the study was naturalistic in design, and although multivariate analyses adjusted for several measured differences across the study groups, unmeasured group differences may explain the differences in choice of providers and continuation of medication treatment. In particular, potentially important predictors of choice of provider and continuity of medication treatment such as participants’ motivation to seek mental health treatment and their attitudes toward psychiatric medications16 were not assessed in the NCS-R. Nevertheless, NCS-R data are unique in that they provide relatively recent and detailed information about the use of prescription medications from a representative population sample along with DSM-IV diagnoses based on structured interviews. These strengths make the NCS-R data especially suitable for analyses of recent patterns of antidepressant medication treatment in the U.S. general population. Three important findings emerge from the current analyses. First, antidepressants are commonly prescribed to adults in the United States. During the course of 1 year, 1 in 10 adults report receiving treatment with an antidepressant medication. Second, adults treated with antidepressants by psychiatrists are significantly more likely than those treated by general medical providers to suffer from common DSM-IV mood and anxiety disorders, to present with significant mood and anxiety symptoms, and to report psychological distress. Third, adults treated by psychiatrists are more likely than those treated by general medical providers to continue antidepressant treatment for at least 1 month and less likely to discontinue within 3 months, even after adjusting for differences in diagnoses, presenting problems, and distress. The rate of antidepressant treatment in the United States1 and in other industrialized countries11,25–27 has grown markedly over the past 2 decades. This trend paralleled introduction of newer antidepressant medications with fewer side effects as well as expansion of the indications for antidepressant treatment. In the United States, this trend also coincided with increased demand for and use of mental health services overall,28 changes in the structure and financing of mental health care,5 and increased visibility of antidepressant medications in popular media,6 as well as the growth of direct to consumer advertising of pharmaceuticals.3 These trends have given rise to concerns about the possible overuse of antidepressants, especially in the general medical sector.29,30 Compared with participants treated by general medical providers, those who were treated with antidepressants by psychiatrists were more likely to meet the criteria for common DSM-IV mood and anxiety disorders except for dysthymia and generalized anxiety disorder. They were also more likely to score in the significantly distressed range on the K6. These epidemiologic results are consistent with past clinical research documenting greater severity PSYCHIATRIST.COM 1073 1072 Mojtabai and Olfson of mental health conditions among individuals who seek treatment in the mental health specialty sector compared with the general medical sector15,31 and are consistent with the greater specialized training of psychiatrists. A tendency for general medical providers to refer more severely ill adults to psychiatrists and self-selection of individuals with more severe symptoms into the mental health specialty sector may further contribute to this pattern of illness severity. The differences across groups with regard to presenting problems corroborate differences in diagnoses, as participants treated by psychiatrists reported complaints that were more consistent with severe mood and anxiety disorders. However, in contrast to past research in clinical samples,32,33 in this community sample, individuals treated with antidepressants by general medical providers were not more likely than those treated by psychiatrists to present with physical complaints such as low energy, poor appetite, or pain. In line with previous studies,31,34–36 adults treated by psychiatrists tended to continue treatment for a longer time period. This difference in antidepressant treatment continuity may be attributable to greater skill of psychiatrists in engaging patients in antidepressant treatment37 or to differences between individuals treated by psychiatrists and general medical providers. Although persistence of depressive symptoms and distress appear to motivate patients to continue treatment,21,38 the associations between distress and antidepressant treatment continuity may be quite complex. In our analyses, elevation of a nonspecific psychological distress screen (K6) was weakly associated with early antidepressant discontinuation. However, as the time frame for the K6 rating was not ascertained in relation to antidepressant use, it is not possible to distinguish patients who experienced an increase in distress soon after stopping antidepressant medications from those who had elevated distress before initiating the medication. Furthermore, the K6 was developed to correlate with a wide range of mental disorders,19 some of which, such as nonaffective psychosis and bipolar disorder, may not respond favorably to antidepressant medication treatment. Some past reports have noted that psychiatrists tend to use antidepressants at higher doses than general medical providers.10,11 This practice has been attributed to a reluctance of general medical providers to use the full recommended doses of the older tricyclic antidepressants borne of a concern over their side effects.39 In an era dominated by newer antidepressants, we found that psychiatrists continue to be more likely than general medical providers to prescribe antidepressants and, more specifically, SSRIs at a higher dose and to reach the maximum recommended dose. There was little evidence that the 2 groups differed with respect to selection of specific antidepressants, although general medical providers were significantly more 1073 1074 likely to prescribe amitriptyline, which is more commonly used than other antidepressants for the treatment of chronic pain.40 The high percentage of participants receiving lower than the recommended antidepressant dose of trazodone and the differences across psychiatrists and general medical providers may be due to prescribing this medication at lower doses as a sleep aid.41 Doses as low as 5 to 20 mg of trazodone have been shown to effectively improve sleep.41 In line with our findings, past research15,31,42 has shown that older adults and women are less likely to receive antidepressants from psychiatrists and more likely to receive them from general medical providers. This might be due to greater access of women and older adults to a regular primary care provider or to these providers’ greater willingness to prescribe antidepressants to these groups rather than to refer them to psychiatrists. In our analysis, health insurance, income, and education—factors traditionally associated with access to specialty mental health care—did not influence the choice between psychiatrists and general medical providers. Furthermore, among participants with insurance, the requirement for preauthorization to seek specialist care was not associated with greater reliance on general medical providers for antidepressant medication treatment. A puzzling finding in the multivariate analyses reported in Table 5 was the association of nonmetropolitan residence with lower likelihood of stopping medication before 30 days. The basis of the association remains unclear. However, nonmetropolitan areas are characterized by relatively few per capita mental health service providers43 and lower rates of mental health treatment for depression.44 It is possible that the greater ease of access to mental health services in metropolitan as compared with nonmetropolitan areas contributes to antidepressant initiation among adults with less motivation to continue treatment. Further research on this important topic is clearly needed with stronger measures of geographic access and motivation for antidepressant treatment. CONCLUSION By the early 2000s, 1 of 10 American adults was treated with an antidepressant during the course of a year, and most of the antidepressant treatment was provided within the general medical sector.2,42 As compared with antidepressant treatment provided by psychiatrists, the treatment provided by general medical professionals was less targeted, with a smaller proportion of the treated individuals meeting criteria for common antidepressant indications. Furthermore, the adults treated by general medical providers were typically treated with lower doses of antidepressants and more prone than their counterparts treated by psychiatrists to report early antidepressant PSYCHIATRIST.COM J Clin Psychiatry 69:7, July 2008 National Patterns in Antidepressant Treatment discontinuation. Amid public concern regarding overuse of antidepressants, these differences support calls for vigorous renewed efforts to improve patient selection and enhance antidepressant treatment continuity in general medical care. Drug names: bupropion (Wellbutrin and others), citalopram (Celexa and others), clomipramine (Anafranil and others), desipramine (Norpramin and others), doxepin (Sinequan, Zonalon, and others), fluoxetine (Prozac and others), fluvoxamine (Luvox), imipramine (Tofranil and others), mirtazapine (Remeron and others), nortriptyline (Pamelor, Aventyl, and others), paroxetine (Paxil, Pexeva, and others), phenelzine (Nardil), protriptyline (Vivactil), sertraline (Zoloft and others), venlafaxine (Effexor and others). 21. 22. 23. 24. 25. 26. REFERENCES 27. 1. Olfson M, Marcus SC, Druss B, et al. National trends in the outpatient treatment of depression. JAMA 2002;287:203–209 2. Kessler RC, Demler O, Frank RG, et al. Prevalence and treatment of mental disorders, 1990 to 2003. N Engl J Med 2005;352:2515–2523 3. Kravitz RL, Epstein RM, Feldman MD, et al. Influence of patients’ requests for direct-to-consumer advertised antidepressants: a randomized controlled trial. JAMA 2005;293:1995–2002 4. Greenfield SF, Reizes JM, Muenz LR, et al. Treatment for depression following the 1996 National Depression Screening Day. Am J Psychiatry 2000;157:1867–1869 5. Frank RG, Glied S. Better but Not Well: Mental Health Policy in the United States Since 1950. Baltimore, Md: Johns Hopkins University Press; 2006 6. Kramer PD. Listening to Prozac. New York, NY: Penguin Books; 1997 7. Beardsley RS, Gardocki GJ, Larson DB, et al. Prescribing of psychotropic medication by primary care physicians and psychiatrists. Arch Gen Psychiatry 1988;45:1117–1119 8. Pincus HA, Tanielian TL, Marcus SC, et al. Prescribing trends in psychotropic medications: primary care, psychiatry, and other medical specialties. JAMA 1998;279:526–531 9. Croghan TW, Melfi CA, Dobrez DG, et al. Effect of mental health specialty care on antidepressant length of therapy. Med Care 1999;37: AS20–AS23 10. Kerr MP. Antidepressant prescribing: a comparison between general practitioners and psychiatrists. Br J Gen Pract 1994;44:275–276 11. McManus P, Mant A, Mitchell P, et al. Use of antidepressants by general practitioners and psychiatrists in Australia. Aust N Z J Psychiatry 2003; 37:184–189 12. Mojtabai R. Datapoints: prescription patterns for mood and anxiety disorders in a community sample. Psychiatr Serv 1999;50:1557 13. Olfson M, Klerman GL. The treatment of depression: prescribing practices of primary care physicians and psychiatrists. J Fam Pract 1992;35: 627–635 14. Bambauer KZ, Soumerai SB, Adams AS, et al. Provider and patient characteristics associated with antidepressant nonadherence: the impact of provider specialty. J Clin Psychiatry 2007;68:867–873 15. Cooper-Patrick L, Crum RM, Ford DE. Characteristics of patients with major depression who received care in general medical and specialty mental health settings. Med Care 1994;32:15–24 16. Van Voorhees BW, Cooper LA, Rost KM, et al. Primary care patients with depression are less accepting of treatment than those seen by mental health specialists. J Gen Intern Med 2003;18:991–1000 17. Wittchen HU. Reliability and validity studies of the WHO Composite International Diagnostic Interview (CIDI): a critical review. J Psychiatr Res 1994;28:57–84 18. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Washington, DC: American Psychiatric Association; 1994 19. Kessler RC, Barker PR, Colpe LJ, et al. Screening for serious mental illness in the general population. Arch Gen Psychiatry 2003;60:184–189 20. American Psychiatric Association. Practice Guideline for the Treatment 28. J Clin Psychiatry 69:7, July 2008 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. of Patients With Major Depressive Disorder (revision). Am J Psychiatry 2000;157(suppl 4):1–45 Olfson M, Marcus SC, Tedeschi M, et al. Continuity of antidepressant treatment for adults with depression in the United States. Am J Psychiatry 2006;163:101–108 Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. New York, NY: Wiley; 2000 StataCorp. Stata Statistical Software, Release 9.2. College Station, Tex: Stata Corporation; 2006 Posternak MA, Zimmerman M. How accurate are patients in reporting their antidepressant treatment history? J Affect Disord 2003;75:115–124 Guaiana G, Andretta M, Corbari L, et al. Antidepressant drug consumption and public health indicators in Italy, 1955 to 2000. J Clin Psychiatry 2005;66:750–755 Helgason T, Tomasson H, Zoega T. Antidepressants and public health in Iceland: time series analysis of national data. Br J Psychiatry 2004;184: 157–162 Patten SB, Beck C. Major depression and mental health care utilization in Canada: 1994 to 2000. Can J Psychiatry 2004;49:303–309 Mojtabai R. Trends in contacts with mental health professionals and cost barriers to mental health care among adults with significant psychological distress in the United States: 1997–2002. Am J Public Health 2005; 95:2009–2014 Croghan TW. The controversy of increased spending for antidepressants. Health Aff (Millwood) 2001;20:129–135 Druss BG. Rising mental health costs: what are we getting for our money? Health Aff (Millwood) 2006;25:614–622 Simon GE, Von Korff M, Rutter CM, et al. Treatment process and outcomes for managed care patients receiving new antidepressant prescriptions from psychiatrists and primary care physicians. Arch Gen Psychiatry 2001;58:395–401 Suh T, Gallo JJ. Symptom profiles of depression among general medical service users compared with specialty mental health service users. Psychol Med 1997;27:1051–1063 Williamson PS, Yates WR. The initial presentation of depression in family practice and psychiatric outpatients. Gen Hosp Psychiatry 1989; 11:188–193 Lewis E, Marcus SC, Olfson M, et al. Patients’ early discontinuation of antidepressant prescriptions. Psychiatr Serv 2004;55:494 Robinson RL, Long SR, Chang S, et al. Higher costs and therapeutic factors associated with adherence to NCQA HEDIS antidepressant medication management measures: analysis of administrative claims. J Manag Care Pharm 2006;12:43–54 Weilburg JB, O’Leary KM, Meigs JB, et al. Evaluation of the adequacy of outpatient antidepressant treatment. Psychiatr Serv 2003;54: 1233–1239 Bull SA, Hu XH, Hunkeler EM, et al. Discontinuation of use and switching of antidepressants: influence of patient-physician communication. JAMA 2002;288:1403–1409 Sirey JA, Bruce ML, Alexopoulos GS, et al. Stigma as a barrier to recovery: perceived stigma and patient-rated severity of illness as predictors of antidepressant drug adherence. Psychiatr Serv 2001; 52:1615–1620 Patience D. Dosage of antidepressant medication dispensed in Scotland. Health Bull (Edinb) 1997;55:322–325 Onghena P, Van Houdenhove B. Antidepressant-induced analgesia in chronic non-malignant pain: a meta-analysis of 39 placebo-controlled studies. Pain 1992;49:205–219 Bon OL. Low-dose trazodone effective in insomnia. Pharmacopsychiatry 2005;38:226 Wang PS, Demler O, Olfson M, et al. Changing profiles of service sectors used for mental health care in the United States. Am J Psychiatry 2006;163:1187–1198 Baldwin LM, Patanian MM, Larson EH, et al. Modeling the mental health workforce in Washington State: using state licensing data to examine provider supply in rural and urban areas. J Rural Health 2006;22: 50–58 Lambert D, Agger M, Hartley D. Service use of rural and urban Medicaid beneficiaries suffering from depression: the role of supply. J Rural Health 1999;15:344–355 PSYCHIATRIST.COM 1075 1074 ...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online