Psychometric prediction of driving performance among the disabled.

A battery of psychometric and performance tests was administered to 25 subjects who were classified into able-bodied (n = 8), brain-injured (n = 10), and spinal cord injured groups (n = 7). All disabled subjects were regarded by their referring rehabilitation therapists as potential candidates for driver assessment. Data were analyzed to identify which measures were useful in differentiating among the groups and predicting driving performance. Results indicated that psychometric measures can be useful in predicting driving performance among disabled drivers. Nearly all the measures in the battery were significant predictors of driving ability, and some were highly predictive. The best was the oral version of the Symbol Digit Modalities test, which by itself accounted for 70% of the variance of the full-sized-vehicle driving score. Adding a second variable into a two-step multiple regression further increased the correlation between predictors and driving, accounting for almost 80% of the variance in driving score. These findings support the feasibility of developing a simple test battery to determine which disabled candidates are ready for in-vehicle assessment, and which candidates are not ready without further evaluation of cognitive and performance skills.