Objective: We sought to determine whether depressive symptoms assessed during hospitalization predicted alcohol use and alcohol-related problems during the subsequent year among medically ill patients.
Method: The study sample was a cohort of hospitalized medical patients with unhealthy alcohol use who participated in a randomized controlled trial of an alcohol brief intervention. Depressive symptoms at baseline, assessed with the Center for Epidemiologic Studies-Depression Scale (CES-D), were used to predict alcohol use and problems at 3 and 12 months.
Results: Of the 341 patients enrolled, 90% (men = 220; women = 88) provided data on at least one follow-up time point during the subsequent year. Gender-stratified longitudinal Pois-son regression models were fit for each alcohol outcome, adjusting for baseline values of age, physical symptoms, randomization group, alcohol outcome, cocaine use, and socioeconomic indicators. Depressive symptoms were significantly associated with drinks per day (men: incidence rate ratio [IRR] = 1.17 per 10-unit increase in CES-D, p < .01; women: IRR = 1.00 per 10-unit increase in CES-D, p = .98) and alcohol-related problems (men: IRR = 1.22 per 10-unit increase in CES-D, p < .001; women: IRR = 1.05 per 10-unit increase in CES-D, p = .39) for men but not for women. They were not significantly associated with the number of days abstinent in men or women.
Conclusions: In hospitalized medical patients with unhealthy alcohol use, depressive symptoms predict subsequent drinks per day and alcohol-related problems among men. These findings suggest that symptoms of depression may be important to consider in treatment planning for male medical patients with unhealthy patterns of drinking.