Average treatment effects on alcohol use among alcohol users (n = 23).

Reported baseline alcohol use score of eight or above on AUDIT; model weighted to account for loss to follow-up, and robust standard error estimators used to account for clustering by counselor. “Pre-post change” is the change from pre-intervention to post-intervention.IQR, interquartile range.Average treatment effects on alcohol use among alcohol users (n = 23).在酒精使用障碍识别测试（Alcohol Use Disorders Identification Test，简称AUDIT）中报告基线酒精使用评分≥8分；模型经加权以处理失访问题，并采用稳健标准误估计量以校正咨询师层面的聚类效应。"干预前后变化"指干预前至干预后的得分变化。四分位距（interquartile range，简称IQR）。酒精使用者群体中酒精使用情况的平均治疗效应（n=23）。