This study compared the effects of nefazodone, a serotonergic antidepressant, with the opioid antagonist naltrexone, and an inactive placebo in 183 alcohol-dependent subjects receiving weekly relapse prevention psychotherapy. Following a single-blind, placebo lead-in period, subjects were randomly assigned to receive study medication, which they took under double-blind conditions for 11 weeks. Naltrexone treatment was associated with significantly more adverse neuropsychiatric and gastrointestinal effects, poorer compliance, and a greater rate of treatment attrition. There were no reliable between-group differences in drinking behavior. These results indicate that nefazodone is not efficacious for treatment of alcohol dependence. Furthermore, the clinical utility of naltrexone seems to be limited by its adverse effects, a finding that has important implications for efforts to develop medications to treat alcohol dependence.