Psychiatry and primary careRecovery from depression, work productivity, and health care costs among primary care patients☆
Introduction
A series of epidemiologic studies over the last 15 years demonstrate a consistent association between depression and lost productivity due to illness. In both the Epidemiologic Catchment Area study [1] and National Comorbidity Survey [2], community residents with depression report five-fold or greater increases in time lost from work compared to those without symptoms of depression. Among primary care patients, current depression is associated with an increase of 2 to 4 disability days per month 3, 4. This association between depression and disability has been consistently demonstrated across a wide range of culture and economic development [5]. Cross-sectional data from two large epidemiologic studies have been used to estimate a monthly productivity loss of approximately $200 to $400 for each worker with major depressive disorder [6].
A parallel series of reports over the last 15 years have demonstrated that depression is associated with significant increases in medical utilization. Various analyses of data from the Epidemiologic Catchment Area survey 7, 8, 9 describe increased use of general medical services among community residents with depressive disorders. Studies using health plan claims 10, 11 data have reported a similar association between use of medical services and measures of either depression or general psychological distress. In two previous studies of primary care patients in a large managed care organization, we reported that depressive disorders (both those detected by screening [12] and those recognized by the primary care physician [13]) were associated with a 50–75% increase in health care costs after adjustment for differences in medical comorbidity. In a study of health care costs among “high utilizers” of general medical services, Henk and colleagues [14] found that depression detected by telephone screening was associated with approximately $1500 in higher medical care costs over one year (a proportional increase of approximately 35 percent). In a sample of Medicare HMO patients, Unutzer and colleagues [15] found that an elevated depressive symptom score was associated with a 30–50% increase in overall health care costs after adjustment for medical comorbidity.
The findings cited above suggest that depression is responsible for a tremendous economic burden on employers and insurers. Most of the studies described above, however, were cross-sectional in design. Depression was assessed during a single time period, and patients with depression were compared to those without. Such a cross-sectional design suffers from significant limitations. An observed association between depression and disability or health care costs might actually reflect the confounding influence of other patient characteristics such as personality disorder, somatoform disorder, or comorbid medical illness. Even if not due to confounding factors, a cross-sectional association between depression and either disability or use of medical service might not necessarily imply that this burden could be reduced. Work disability or excess use of medical services might be limited to patients with the most chronic or treatment-resistant depression. A longitudinal study design (i.e., assessment of depression at multiple time points) can provide stronger evidence that improvement in depression results in increased work productivity or decreased medical utilization.
Previous longitudinal studies provide only limited data regarding the effects of improvement in depression on work disability and health care costs. Longitudinal and experimental studies have found that improvement in depression is associated with significant improvement in self-reported 4, 16, 17 or interviewer-rated [18] measures of daily functioning. Because negative self-assessment is a core symptom of depression, some have suggested that changes in self-reported functioning associated with depression may represent biased self-assessment rather than true disability [19]. Our previous analyses of longitudinal and experimental data 20, 21 suggest an association between depression and more “objective” disability measures (such as work missed due to illness)—but sample sizes were not adequate to draw firm conclusions [4]. Using pooled data from several depression treatment studies, Mintz and colleagues did find that favorable symptomatic outcomes were associated with favorable work outcomes [22]. Wells and colleagues [23] recently have reported that improved depression treatment in primary care may be associated with higher likelihood of continued employment during long-term follow up. We previously attempted to examine the relationship between change in depression and change in health care costs in one of the primary care studies described above, but the sample size was not sufficient to detect even moderately large effects [12].
This report uses a sample of primary care patients beginning antidepressant treatment to examine whether improvement in depression is associated with changes in probability of paid employment, time missed from work due to illness, and overall health care costs. These data were collected as part of a randomized trial comparing effectiveness and cost of fluoxetine and tricyclic antidepressants in primary care practice. The long-term follow-up and longitudinal health plan cost accounting data allow us to examine the longitudinal relationship between recovery from depression and economic impact from the perspective of the insurer and employer.
Section snippets
Methods
Study methods are described in detail in earlier publications 24, 25 and will be summarized here. Patients were enrolled from selected primary care clinics of Group Health Cooperative (GHC) of Puget Sound, a staff-model health maintenance organization serving approximately 400,000 members. The study protocol was approved by GHC’s Human Subjects Review Committee. At participating clinics, all primary care physicians were asked to refer any adult patient beginning antidepressant treatment for
Results
In this report we limit our analyses to patients satisfying DSM-IIIR criteria for major depressive episode at the baseline assessment. Of the 358 patients in this group, 290 (81%) completed the 12-month assessment so that depression status at 12 months (our primary independent variable) could be determined. Of these 290 eligible patients, the numbers completing subsequent follow-up assessments were 273 (94%) at 18 months and 260 (90%) at 24 months. Participation in follow-up interviews (at 12,
Discussion
In this sample of primary care patients beginning depression treatment, recovery from depression was associated with significant increases in probability of paid employment, and reductions in time lost from work due to illness. Better clinical outcome was also associated with a marginally significant reduction in health care costs, but this relationship was not apparent until the second year of follow-up. Clinical outcome at one year (persistent depression, improved, remitted) showed an
References (52)
- et al.
Cost-effectiveness comparisons using “real world” randomized trialsthe case of new antidepressant drugs
J Clin Epidemiol
(1995) - et al.
Telephone assessment of depression severity
J Psychiatr Res
(1993) - et al.
Agreement between face-to-face and telephone-administered versions of the depression section of the NIMH Diagnostic Interview Schedule
J Psychiatric Res
(1988) - et al.
A chronic disease score from automated pharmacy data
J Clin Epidemiol
(1992) The logged dependent variable, heteroscedasticity, and the retransformation problem
J Health Econ
(1998)- et al.
Outcomes of recognized and unrecognized depression in an international primary care study
Gen Hosp Psychiatry
(1999) Amplification, somatization, and the somatoform disorders
Psychosomatics
(1992)- et al.
Depression, disability days and days lost from work in a prospective epidemiologic survey
JAMA
(1990) - et al.
The impact of psychiatric disorders on work loss days
Psychol Med
(1997) - Spitzer R, Kroenke K, Linzer M, Hahn SR, Williams JBW, deGruy FV, Brody D, Davies M: Health-related quality of life in...
Disability and depression among high utilizers of health care
Arch Gen Psychiatry
Common mental disorders and disability across cultures
JAMA
Depression in the workplaceEffects on short-term disability
Health Affairs
Service utilization and social morbidity associated with depressive symptoms in the community
JAMA
Psychiatric disorder and functional somatic symptoms as predictors of health care use
Psychiatric Med
Distress syndromes, illness behavior, access to care and medical utilization in a defined population
Medical Care
The effects of psychological distress and psychological well-being on use of medical services
Medical Care
Health care costs associated with depressive and anxiety disorders in primary care
Am J Psychiatry
Health care costs of primary care patients with recognized depression
Arch Gen Psychiatry
Medical costs attributed to depression among patients with a history of high medical expenses in a health maintenance organization
Arch Gen Psychiatry
Depressive symptoms and the cost of health services in HMO patients age 65 and overa four-year prospective study
JAMA
Impact of improved depression treatment in primary care on daily functioning and disability
Psychol Med
Treating depressed primary care patients improves their physical, mental, and social functioning
Arch Intern Med
Depression, anxiety, and social disability show synchrony of change in primary care patients
Am J Public Health
Subjective bias in reports of poor work adjustment in depressed patients
Acta Psychiatr Scand
Cited by (0)
- ☆
Supported by a grant from Lilly Research Laboratories and by NIMH Grant #MH51338.