Benzodiazepines appear to be the most effective drugs in treating alcohol withdrawal. Members of the American Society of Addiction Medicine came to this conclusion after reviewing 134 articles on the treatment of alcohol withdrawal, 65 of which were controlled trials covering 42 different drugs. Long-acting benzodiazepines may be more effective but also have the potential for abuse and excessive sedation. Using a structured assessment scale to provide drugs only when symptoms become severe can reduce the length of treatment and the dosage of drugs used., Objective.--To provide an evidence-based practice guideline on the pharmacological management of alcohol withdrawal. Data Sources.--English-language articles published before July 1, 1995, identified through MEDLINE search on "substance withdrawal--ethyl alcohol" and review of references from identified articles. Study Selection.--Articles with original data on human subjects. Data Abstraction.--Stuctured review to determine study design, sample size, interventions used, and outcomes of withdrawal severity, delirium, seizures, completion of withdrawal, entry into rehabilitation, adverse effects, and costs. Data from prospective controlled trials with methodologically sound end points corresponding to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, were abstracted by 2 independent reviewers and underwent meta-analysis. Data Synthesis.--Benzodiazepines reduce withdrawal severity, reduce incidence of delirium (-4.9 cases per 100 patients; 95% confidence interval, -9.0 to -0.7; P = .04), and reduce seizures (-7.7 seizures per 100 patients; 95% confidence interval, -12.0 to -3.5; P = .003). Individualizing therapy with withdrawal scales results in administration of significantly less medication and shorter treatment (P [is less than] .001). [Beta]-Blockers, clonidine, and carbamazepine ameliorate withdrawal severity, but evidence is inadequate to determine their effect on delirium and seizures. Phenothiazines ameliorate withdrawal but are less effective than benzodiazepines in reducing delirium (P = .002) or seizures (P [is less than] .001). Conclusions.--Benzodiazepines are suitable agents for alcohol withdrawal, with choice among different agents guided by duration of action, rapidity of onset, and cost. Dosage should be individualized, based on withdrawal severity measured by withdrawal scales, comorbid illness, and history of withdrawal seizures. [Beta]-Blockers, clonidine, carbamazepine, and neuroleptics may be used as adjunctive therapy but are not recommended as monotherapy. JAMA. 1997;278:144-151