Methods used in studies of drink-drive control efforts: a meta-analysis of the literature from 1960 to 1991.

We searched the drink-drive control literature over the past three decades, finding over six thousand documents. After detailed review of the abstracts and papers, 125 studies contained separate empirical evaluations of the effects of 12 DWI control policies and enforcement efforts (administrative license suspension, illegal per se, implied consent, preliminary breath test, mandatory jail sentence, mandatory community service, mandatory license suspension, limits on plea bargaining, mandatory fines, selective enforcement patrols, regular police patrols, and sobriety checkpoints). The 125 studies contained 664 distinct analyses that formed the basis for meta-analysis. All of the DWI control efforts were associated with reductions in drink-driving and traffic crashes. The DWI control literature is limited by the preponderance of weak study designs and reports that often fail to include basic data required for meta-analysis. Because of the poor quality of much extant research, we were limited to simple gain scores or percent change estimates in the current study. Further research that does not include appropriate research designs and analytic methods will be of limited utility. We recommend that all future reports include effect estimates and standard error estimates, minimum data required for effective meta-analysis.