This study failed to demonstrate a significant difference between acamprosate and placebo in terms of the number of days of cumulative abstinence, but clinically this outcome is less important than maintaining complete abstinence or increasing the proportion of abstinent days. Nonetheless, the apparent lack of efficacy of acamprosate in this regard requires further consideration. Both groups demonstrated a high proportion of days of cumulative abstinence during the treatment period (73.9% for acamprosate and 74.2% for placebo). This may be explained by our rigorous efforts to establish abstinence status in order to ensure data integrity, which may have given extra impetus and motivation to the participants. All of the subjects enrolled in this study had cooperative attendants. The subjects consented to keeping a diary and had the ability to make entries in their diaries. These conditions are associated with favorable prognosis after treatment and may have contributed to the high proportion of days of cumulative abstinence in both groups. However, in terms of the rate of complete abstinence, the primary endpoint of this study, a statistically significant difference was found between the 2 groups because a subject who drank alcohol even once was excluded from the abstinent cases even if the proportion of days of cumulative abstinence was high. We used biological markers including GGTP, AST, and ALT to validate the drinking status of the subjects ascertained by investigators. This procedure is considered to increase further the reliability of the drinking outcome in this study.