Clinical Question

Is baclofen effective in achieving abstinence in patients with alcohol use disorder (AUD)?

Evidence-Based Answer

Baclofen with psychosocial therapy is modestly effective in helping patients with AUD achieve abstinence, increasing abstinent days by approximately 10%. (Strength of Recommendation [SOR]: A, systematic review and meta-analysis of randomized controlled trials and cohort studies.) A total of 40% to 50% of patients can achieve or maintain abstinence at 1 year after treatment initiation. (SOR: B, several small cohort studies.) Baclofen does not show evidence of worsening liver function in patients with cirrhosis.

Evidence Summary

A 2023 Cochrane review evaluated 17 randomized controlled trials (n = 1,818) investigating the use of baclofen in patients with AUD.¹ Studies assessed baclofen compared with placebo, no treatment, and other pharmacologic therapies for achieving alcohol abstinence or reduced consumption in patients with AUD. The mean age of participants was 46.5 years, and 70% were men. Three studies recruited people with severe liver disease. In all but one study, patients received psychosocial therapy. Treatment duration was at least 4 weeks, with a mean duration of 16 weeks. The daily baclofen dosage ranged from 30 to 300 mg. Most studies started at fixed dosages of 5 mg three times per day and increased to a total daily dosage of 30 to 80 mg. When compared with placebo, baclofen was found to increase the percentage of abstinent days in patients taking low-dose daily regimens of 30 mg or less (mean difference [MD] = 10.6%; 95% CI, 0.8 to 20.4; eight studies; 583 participants) and high-dose regimens of 100 mg or more (MD = 11.1%; 95% CI, 4.4 to 17.8; three studies; 465 participants). There were no differences in abstinent days between patients taking baclofen vs. placebo among studies using medium dosages of 30 to 100 mg daily (MD = 7.1%; 95% CI, −3.1 to 17.4; five studies; 225 participants). Compared with placebo, baclofen was found to decrease the risk of relapse in people no longer drinking alcohol (risk ratio = 0.87; 95% CI, 0.77 to 0.99; 12 studies; 1,057 participants). There were no differences in the risk of relapse at the end of treatment between low, medium, and high dosages compared with placebo. Moderate to substantial heterogeneity was noted between the included studies.