A strategic driving maneuver that predicts on-road outcomes in adults with multiple sclerosis

Abstract

ObjectiveThe purpose of this study is threefold: (1) To compare and determine the differences in higher-order cognitive functions and driving errors underlying the strategic driving maneuver between persons with Multiple Sclerosis (PwMS) who pass versus fail the University of Western Ontario (UWO) on-road assessment; (2) To quantify the bivariate correlational relationships between higher-order cognitive functions and driving errors of strategic driving maneuvers; and (3) To determine the predictive validity of higher-order cognitive functions and driving errors on pass versus fail outcomes. MethodsThis prospective study included licensed drivers (N = 35; mean age = 50 years) with either relapsing-remitting or progressive MS, and cognitive impairment in processing speed and at least one of memory or executive function, but low physical disability. Participants completed the Minimal Assessment of Cognitive Function in MS and Useful Field of View™ before the UWO on-road assessment. The on-road assessment incorporated a strategic driving maneuver that required higher-order cognitive functions (i.e., executive function) to assess, initiate, plan, reason, decide, and problem-solve, to select a route from various alternatives and reach a goal-directed destination. The on-road assessment included pass versus fail outcomes. The strategic driving maneuver included the number and total of seven types of driving errors as outcomes. ResultsSeven of thirty-five participants failed the on-road assessment. No significant differences existed on clinical assessment measures between PwMS who passed versus failed. When compared to PwMS who passed the on-road assessment, PwMS who failed made significantly more lane maintenance (mean = 1.1 versus 0.3, p = .02), and speed regulation errors (mean = 0.7 versus 0.04, p = .03) in the strategic driving maneuver. Poorer delayed visuospatial recall moderately correlated with a higher number of speed regulation errors (ρ = −0.37, p = .03). Final regression models included two univariate models. The first model included lane maintenance errors (OR = 0.18, p = .009, 95% CI = [0.05, 0.66]), and the second model included speed regulation errors (OR = 0.04, CI = [0.003, 0.44]), as sole predictors of pass versus fail outcomes in PwMS. An optimal cut-point of one or more lane maintenance errors validly predicted 78 percent (p = .02) of pass versus fail outcomes, with 71 percent sensitivity, 79 percent specificity, and 23 percent misclassification rate. An optimal cut-point of one or more speed regulation errors validly predicted 77 percent (p = .03) of pass versus fail outcomes, with 57 percent sensitivity, 96 percent specificity, and 11 percent misclassification rate. ConclusionsThe strategic driving maneuver can validly predict 77–78 percent of pass versus fail outcomes in PwMS, through lane maintenance or speed regulation errors. Either driving error may supplement on-road outcomes to help determine pass versus fail outcomes in PwMS.