This study has several limitations to consider. First, the drop-out rate is significant, as is often the case in clinical trials in substance dependence. However, to put our results in context, the treatment completion rates reported in a meta-analysis of randomized controlled trials involving 6,111 outpatients with alcohol dependence were 52.7% for placebo and 57.8% for acamprosate, which is directly comparable to our treatment completion rate of 56%.43 Furthermore, our mean duration of study participation was 9.1 weeks of a 12-week study, which is a clinically relevant period of drug exposure for assessing treatment effects. Concerns about potential bias introduced by drop-outs are mitigated by a lack of differential drop-out between groups, and by consistency across outcomes that include the assumption of missing at random and response variables derived from data collected on study without assumption for 96.7% of participants. The validity of results is supported by pre-clinical12 and human laboratory studies13 of gabapentin effects on models of protracted abstinence and by clinical proof-of-concept studies from different groups.14, 17, 20, 21
