Driving simulation for evaluation and rehabilitation of driving after stroke

Abstract

Driving is an important activity of daily living. Loss of driving privileges can lead to depression, decreased access to medical care, and increased healthcare costs. The ability to drive is often affected after stroke. In approximately 30% of stroke survivors, it is clear from the onset that driving will no longer be possible. Approximately 33% of survivors will be able to return to driving with little or no retraining, and 35% will require driving-related rehabilitation before they can resume safe driving again. The ability to drive is not routinely evaluated after stroke, and there is no established rehabilitation program for poststroke driving. When driving evaluation does occur, it is not always clear which tests are the most salient for accurately assessing poststroke driving ability. Investigators have examined the efficacy of various methodologies to predict driving performance after stroke and have found mixed results, with each method having unique weaknesses, including poor predictive ability, poor face validity, poor sensitivity or specificity, and limited reliability. Here we review common models of driving to gain insight into why single-construct visual or cognitive off-road measures are inadequate for evaluating driving, a complex and dynamic activity that involves timely interaction of multiple motor, visual, cognitive, and perceptual skills. We also examine the potential for driving simulators to overcome the problems currently faced in the evaluation and rehabilitation of driving after stroke. Finally, we offer suggestions for the future direction of simulator-based poststroke driving evaluation and training.