Self-reported depressive symptoms following treatment with corticosteroids and sedative-hypnotics.

OBJECTIVE To evaluate associations between exposure to corticosteroids or sedative-hypnotic medications and incident self-reported depressive symptoms in medical inpatients. METHOD The study utilized a prospective cohort design, focusing on acute depressive symptoms developing soon after medication exposure. The incidence of self-reported depressive symptoms was evaluated using a modified version of the Center for Epidemiological Studies Depression Rating Scale (CES-D). The incidence of depressive symptoms in subjects newly exposed to corticosteroids and sedative-hypnotics was compared to that of a nonexposed comparison cohort. RESULTS The incidence of self-reported depressive symptoms was elevated in subjects newly exposed to corticosteroids (Risk Ratio = 3.10), although the association did not attain statistical significance (p = .07). The risk ratio for sedative-hypnotic exposure was 4.18, a statistically significant finding (p = .02). As expected, incident self-reported depressive symptoms were also associated with several psychosocial variables. However, the data did not suggest that the observed associations between drug exposures and depressive symptoms were due to confounding by psychosocial or illness-related variables.