Depression in late life: the use of clinical characteristics to focus screening efforts.

BACKGROUND The objective of the study was to identify clinical characteristics associated with depressive symptoms in late life so that screening could focus on elderly patients most likely to benefit from further evaluation. METHODS We used cross-sectional screening for significant symptoms of depression using the Center for Epidemiologic Studies Depression scale and identification of patients' clinical characteristics from patient interviews and a computerized medical record. The setting was an academic primary care group practice at an urban ambulatory care center. Participants were 1,633 consecutively consenting patients aged 60 and older who visited the center between January and August 1991. Mean age was 70 years; 72% were women, 32% were White, 47% had less than 8 years of education, and 7% had no health insurance. RESULTS There were 251 (15%) patients with significant symptoms of depression. Antidepressants were prescribed to 1 in 7 patients with such symptoms, with amitriptyline being the most commonly prescribed. Bivariate analyses indicated that patients with significant symptoms of depression were more likely to be White, female, without health insurance, and were more likely to have probable alcoholism, mild cognitive loss, and to receive narcotics, histamine H2 antagonists, and/or benzodiazepines. Depressive symptoms were not significantly correlated with age, education, income, or chronic medical conditions. CONCLUSIONS Significant symptoms of depression were common and correlated with several readily available clinical variables. However, these variables lack sufficient discriminatory power to allow for the selective screening of elderly patients most likely to suffer from symptoms of depression. Thus, formal screening for depression among all elderly patients in primary care may be necessary to improve the recognition of this morbid illness.