Can methodological features account for patient-treatment matching findings in the alcohol field?

Objective: Despite enthusiasm for the potential of matching patients to alc ohol treatments to improve outcomes, consistent findings have not emerged. This review considers the extent to which methodological factors may accoun t for the pattern of findings from research on Patient x Alcohol Treatment interactions. Method: We focused on 55 studies that compared more than one type of alcohol treatment and included formal statistical tests for interac tions. We examined four predictors of the number of significant interaction s found in the 55 studies: (1) the number of statistical tests for interact ions conducted (2) the average number of participants, (3) whether or not p articipants were randomized to treatment and (4) the proportion of tested i nteractions that were hypothesis- or rationale-driven, as opposed to explor atory. Results: Only the number of statistical tests for interactions predi cted the number of patient-treatment interactions identified per study (zer o-order r = 0.47; r(2) = 0.22). A substantial number of tests for interacti ons (43) was conducted on average, per study. Only a minority of the studie s (33%) included enough participants to have a reasonable probability (0.80 ) of identifying a medium-sized matching effect. Conclusions: Drawing gener al conclusions regarding matching patients to alcohol treatments is hampere d because Type I error has contributed to the matches identified studies in this area are often underpowered and the combinations of patient and treat ment variables that have been tested are few relative to the numerous possi ble combinations. To be productive, future research will need to focus on p atients at the extremes of matching dimensions and on distinct treatments.