Combined Pharmacotherapies and Behavioral Interventions for Alcohol Dependence: The COMBINE Study: A Randomized Controlled Trial

CONTEXT Alcohol dependence treatment may include medications, behavioral therapies, or both. It is unknown how combining these treatments may impact their effectiveness, especially in the context of primary care and other nonspecialty settings. OBJECTIVES To evaluate the efficacy of medication, behavioral therapies, and their combinations for treatment of alcohol dependence and to evaluate placebo effect on overall outcome. DESIGN, SETTING, AND PARTICIPANTS Randomized controlled trial conducted January 2001-January 2004 among 1383 recently alcohol-abstinent volunteers (median age, 44 years) from 11 US academic sites with Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, diagnoses of primary alcohol dependence. INTERVENTIONS Eight groups of patients received medical management with 16 weeks of naltrexone (100 mg/d) or acamprosate (3 g/d), both, and/or both placebos, with or without a combined behavioral intervention (CBI). A ninth group received CBI only (no pills). Patients were also evaluated for up to 1 year after treatment. MAIN OUTCOME MEASURES Percent days abstinent from alcohol and time to first heavy drinking day. RESULTS All groups showed substantial reduction in drinking. During treatment, patients receiving naltrexone plus medical management (n = 302), CBI plus medical management and placebos (n = 305), or both naltrexone and CBI plus medical management (n = 309) had higher percent days abstinent (80.6, 79.2, and 77.1, respectively) than the 75.1 in those receiving placebos and medical management only (n = 305), a significant naltrexone × behavioral intervention interaction (P = .009). Naltrexone also reduced risk of a heavy drinking day (hazard ratio, 0.72; 97.5% CI, 0.53-0.98; P = .02) over time, most evident in those receiving medical management but not CBI. Acamprosate showed no significant effect on drinking vs placebo, either by itself or with any combination of naltrexone, CBI, or both. During treatment, those receiving CBI without pills or medical management (n = 157) had lower percent days abstinent (66.6) than those receiving placebo plus medical management alone (n = 153) or placebo plus medical management and CBI (n = 156) (73.8 and 79.8, respectively; P