This review considers the novel drug treatments that have been suggested to help prevent relapse or attenuate drinking in people with alcohol problems. The evidence from randomized controlled trials for the efficacy of some of the main candidates: acamprosate, naltrexone, bromocriptine, selective serotonin re-uptake inhibitors and buspirone, was examined. Important methodological problems which may have introduced bias were detected in many of the trials. These included failure to test the integrity of the double blind, excluding or estimating outcome in early withdrawals and the comparison of groups on multiple outcome measures with selective reporting of results. In addition, the generalizability of some studies was limited by the procedures used for sample selection. In view of the potential adverse effects of drug treatment it is concluded that the evidence is not strong enough to support the introduction of any of these substances into routine clinical practice at present. The review also emphasizes the importance of methodological rigour to maximize objectivity in treatment evaluation research.