Studies in Europe and the U.S. have indicated that baclofen reduces alcohol intake in a variety of animal models and, in humans, reduces drinking and craving and enhances abstinence. Our group, based on fi ndings from a preliminary open label study (n = 12), have conducted a randomized, placebo-controlled 12-week trial comparing 10 mg t.i.d. po of baclofen to placebo in 80 alcohol dependent outpatients. In conjunction with medication or placebo, participants received nine counseling sessions using BRENDA — a supportive, compliance enhancement therapy. Newspaper and radio advertisements were employed to recruit participants. Along with several other inclusion/exclusion criteria, participants were required to average at least two heavy drinking days/week in the month prior to screening. Randomization was balanced by gender (about 40% of participants were women). Approximately 75% of subjects completed
the protocol. Reasons for drop-outs included time confl icts, minor adverse events, and lack of efficacy. To date, participants have demonstrated a robust decrease in heavy drinking although, at this point in time, the blind has not been broken. Differential effects of baclofen versus placebo on drinking and other factors will be conducted upon study completion estimated for Spring, 2007. The primary analysis will examine medication effect on percent days of heavy drinking. Secondary analyses will include medication effects on time between heavy drinking days, percent days of abstinence during treatment and treatment effects on craving, anxiety,
and depressed mood.