Predictability of clinical assessments for driving performance

Problem: As the number of older drivers grows, it is increasingly important to accurately identify at-risk drivers. This study tested clinical assessments predictive of real-time driving performance. Method: Selected assessment tools considered important in the identification of at-risk older drivers represented the domains of vision, cognition, motor performance, and driving knowledge. Participants were administered the battery of assessments followed by an on-road test. A univariate analysis was conducted to identify significant factors (< .05) to be included in a multivariate regression model. Results: Assessments identified as independently associated with driving performance in the regression model included: FACTTM Contrast sensitivity slide-B, Rapid Pace Walk, UFOV® rating, and MMSE total score. Discussion: The domains of vision, cognitive, and motor performance were represented in the predictive model. Summary: Due to the dynamic nature of the driving task, it is not likely that a single assessment tool will identify at risk drivers. Impact on Industry: By standardizing the selection of clinical assessments used in driving evaluations, practitioners should be able to provide services more efficiently, more objectively, and more accurately to identify at-risk drivers.