[Skip to Content]
[Skip to Content Landing]
Original Contributions
March 18, 1992

Service Utilization and Social Morbidity Associated With Depressive Symptoms in the Community

Author Affiliations

From the Department of Psychiatry, Columbia University College of Physicians and Surgeons (Drs Johnson and Weissman); Division of Clinical and Genetic Epidemiology, New York State Psychiatric Institute (Drs Johnson and Weissman); and Department of Psychiatry, Cornell University Medical Center and The Payne-Whitney Clinic, New York Hospital (Dr Klerman), New York, NY.

JAMA. 1992;267(11):1478-1483. doi:10.1001/jama.1992.03480110054033

Objective.  —To estimate service utilization and social morbidity in the community associated with depressive symptoms. Estimates were made using an epidemiologic measure, population attributable risk. Population attributable risk is a compound measure reflecting both the morbid risk to an individual with a disorder and the prevalence of the disorder in the community.

Design.  —Epidemiologic survey.

Participants.  —Eighteen thousand five hundred seventy-one adults in the Epidemiologic Catchment Area Study interviewed from a complex random sample in five US communities.

Outcome Measures.  —Suicide attempts, use of psychoactive medications, self-reported physical and emotional health, time lost from work, and general medical services or use of emergency departments for emotional problems.

Results.  —Major depression-dysthymia (lifetime prevalence, 6.1%) and depressive symptoms (lifetime prevalence, 23.1%) were associated with increased service utilization and social morbidity as measured by the outcome variables. On a population basis, however, as much or more service burden and impairment was associated with depressive symptoms as with the clinical conditions of depression or dysthymia. The equal association results from the greater prevalence of depressive symptoms. Population attributable risk percentages associated with depressive symptoms (not disorder) were as follows: emergency department use (11.8%) or medical consultations for emotional problems (21.5%); use of tranquilizers (14.6%), sleeping pills (21.0%), or antidepressants (22.2%); fair or poor self-reported emotional health (15.3%); days lost from work (17.8%); and suicide attempts (25.0%).

Conclusions.  —Estimates of population attributable risk indicated that physicians actually provided services to more persons with depressive symptoms than to persons with formally defined conditions of depressive disorders. Subclinical depression, as a consequence of high prevalence, is a clinical and public health problem. Attention to diagnostic and treatment issues is indicated.(JAMA. 1992;267:1478-1483)