ABSTRACT
Objective: To determine the test–retest reliability of a computerized preroad screen of prerequisite skills required for safe driving. Subjects & Methods: Driveable Screen administered to 39 individuals referred for driving evaluation due to cognitive decline or stroke. The Screen provides three recommendations—driving cessation, indeterminate, no evidence of reduced competence. Predicted probability of failing the road test is also generated. Results: Reliability according to the recommendation was substantial for the entire group (K = 0.654; CI = 0.459–0.848) and subgroup with stroke (K = 0.742; CI = 0.540–0.944). The numbers of participants with cognitive decline was too small to allow subgroup analyses. Of the nine participants with test–retest disagreement in recommendation, Test 2 was always better (McNemar's X2 = 9.00, p = 0.01). Excellent reliability was found on probability of failing the Road Test (Intraclass Correlation Coefficient = 0.908; CI = 0.832–0.950). The test–retest reliability of the subtests ranged from fair to substantial. Conclusions: The Screen demonstrates sufficient test–retest reliability to encourage its use. However, the finding of significantly better scores on Test 2 for a portion of participants raises concern regarding its validity with repeat testing.