Medication use and major depressive syndrome in a community population.

A variety of medications have been reported to cause depression as a side effect. This study evaluated cross-sectional associations between a variety of medications and a syndrome resembling major depression. A sample of 2,542 subjects were selected using the Mitofsky-Waksberg random digit dialing (RDD) procedure. Major depression was evaluated using a short form version of the Composite International Diagnostic Interview (CIDI). Current medication use was recorded during the same telephone interview. Statistical methods accounting for clustering and unequal selection probabilities were employed. Most medications were not associated with major depression: these included beta-blockers, angiotensin-converting enzyme (ACE) inhibitors, lipid-lowering agents, digoxin, and diuretics. Calcium channel blockers were significantly associated with major depression, but only in a subset of young subjects, and these tended to be seriously ill individuals taking multiple medications. Opiate analgesics were associated with major depression, but only in male subjects. Corticosteroids were significantly associated with major depression in a logistic regression model that adjusted for age and gender. The cross-sectional nature of this study precludes causal inference about the observed associations. With the exception of the association of major depression with corticosteroid use, convincing associations with other medications were not observed. It is possible that medication-induced depressive episodes lead to changes in exposure status (such as discontinuation of the offending medications) such that the associations are not apparent in cross-sectional data. If this interpretation is correct, these data suggest that the problem of medication-induced depression is being managed effectively at the clinical level and is not a substantial public health problem.