|| Checking for direct PDF access through Ovid
To estimate the relative efficacy of alcohol use disorder treatments, the authors meta-analyzed studies that directly compared 2 bona fide psychological treatments. The authors accommodated problems with the inclusion of multiple treatment comparisons by randomly assigning a positive/negative sign to the effect size derived from each comparison and then estimating the extent to which effect sizes were heterogeneous. The authors' primary hypothesis was that the variability in effect sizes of bona fide psychological treatments for alcohol use disorders that were directly compared would be zero. For both alcohol measures and measures of abstinence, analyses indicate that effects were homogenously distributed about zero (I2 = 10.61, 0.00, respectively), indicating that different treatment comparisons yielded a common effect size that was not significantly different from zero. Analyses also indicate that allegiance accounted for a significant portion of variability in differences between treatments. Implications for the treatment of alcohol use disorders as well as research on the mechanisms responsible for the benefit of treatment are discussed.