Roadwise review has limited congruence with actual driving performance of aging drivers

The Roadwise Review (RR) CD-ROM has been proposed as one way for older drivers to self-evaluate the skills that support safe driving. To better gauge the utility of RR for that purpose, further validation of RR vis-a-vis measures of driving performance is needed. We carried out this study to provide such validation. We correlated the results of the standard Useful Field of View subtest 2 (UFOV-2) and Trail Making Test (TMT) parts A and B to the corresponding RR subsections on visual information processing and visual search. We further examined the correlations between on-road performance (based on demerit points) and RR results (absolute scores and number of problems identified). The correlations between UFOV-2, TMT and the relevant RR sections ranged from .46 to .61. The correlations between demerit points and RR absolute values ranged from -.40 to .50. The correlations between demerit points and the number of problems identified by RR ranged from -.24 to .45. Using a "pass/fail" approach we also failed to find a relationship with the number of problems. These results indicate limited convergence between findings obtained with RR and actual performance using standardized approaches. Such results emphasize the importance of developing a rigorous data synthesis process before attempting knowledge translation of material aimed at older drivers. (C) 2011 Elsevier Ltd. All rights reserved.