Comparing the predictive validity of DUI risk screening instruments: development of validation standards

Aims. This study compares the predictive efficacy of driving under the infl uence (DUI) screening instruments validated in previous studies, illustrate s how variations in base rates of failure and selection ratios affect concl usions concerning the efficacy of different instruments, and develops evalu ation standards to ensure valid comparisons of risk prediction instruments. Design. The study: (I) examines a sample of 4815 DUI offenders to illustra te how variations in base rates of failure and selection ratios affect trad itional measures of predictive efficacy, (2) uses such measures to compare the predictive efficacy of 10 instruments validated in previous studies, an d (3) demonstrates the use of a measure of predictive efficacy which is rel atively insensitive to the aforementioned variations. Findings. While three instruments Examined at specific cut-points consistently ranked highest on several measures of predictive efficacy, use of different evaluation stand ards produced substantively different conclusions regarding the efficacy of different instruments. Based on the analyses, standards for validation of risk prediction instruments were developed. Conclusions. The findings illus trate how failure to use equivalent standards have led to erroneous conclus ions concerning the relative predictive efficacy of different risk predicti on instruments. The standards developed in this study should facilitate equ ivalent comparisons of the predictive efficacy of risk prediction instrumen ts.