Major depression and alcohol problems are common in primary care, yet little is known about the prevalence of alcohol problems in patients with depression or alcohol's effect on depression outcomes. We strove to answer the following questions: How common are alcohol problems in patients with depression? Does alcohol affect the course of depression, response to antidepressant therapy, risk of suicide/death, social functioning and health care utilization? In which alcohol categories and treatment settings have patients with depression and alcohol problems been evaluated? English language studies from MEDLINE, PsychINFO, and Cochrane Controlled Trial Registry were reviewed. Studies were selected using predefined criteria if they reported on the prevalence or effects of alcohol problems in depression. Thirty-five studies met criteria and revealed a median prevalence of current or lifetime alcohol problems in depression of 16% (range 5-67%) and 30% (range 10-60%), respectively. This compares with 7% for current and 16-24% for lifetime alcohol problems in the general population. There is evidence that antidepressants improve depression outcomes in persons with alcohol dependence. Alcohol problems are associated with worse outcomes with respect to depression course, suicide/death risk, social functioning, and health care utilization. The majority of the studies, 34 of 35 (97%), evaluated alcohol abuse and dependence, and 25 of 35 (71%) were conducted in psychiatric inpatients. We conclude that alcohol problems are more common in depression than in the general population, are associated with adverse clinical and health care utilization outcomes, and that antidepressants can be effective in the presence of alcohol dependence. In addition, the literature focuses almost exclusively on patients with alcohol abuse or dependence in psychiatric inpatient settings, and excludes patients with less severe alcohol problems and primary care outpatient settings.