Previous research has examined the association between patient characteristics and low risk (i.e., controlled) drinking outcomes among individuals with alcohol use disorder[11, 12, 14], but most prior research has been limited by small sample sizes and has primarily focused on predicting low risk drinking outcomes following treatment. The current study tested baseline predictors of drinking patterns during treatment among 3,589 patients across three alcohol clinical trials. Of seven distinct patterns of drinking, we identified two low risk drinking patterns: (1) consistent low risk drinking throughout treatment and (2) abstinence early in treatment and a higher probability of low risk drinking during later weeks of treatment. Combined across both patterns, we found over 17% of the sample achieved low risk drinking, in the absence of heavy drinking, by the end of treatment. One of the primary strengths of the current study was the use of an analytic approach that allowed for some deviations from abstinence and low risk drinking in identifying overall patterns of drinking during treatment, whereas many prior studies have defined low risk drinking by never exceeding low risk
