OBJECTIVES:

To determine the validity of a road test performed by stroke patients in Belgium and to reestablish its reliability.

DESIGN:

Prospective study of a predriving evaluation.

SETTING:

University hospital in Belgium.

PARTICIPANTS:

Thirty-eight patients with sequelae of first-ever stroke.

INTERVENTIONS:

Not applicable.

MAIN OUTCOME MEASURES:

Performance in the Stroke Driver Screening Assessment (SDSA) and on a road test.

RESULTS:

Interrater reliability of the road test subitems was moderate to substantial (weighted kappa range, .44-.78). Item-per-item reliability varied from moderately high (intraclass correlation coefficient [ICC]=.63) to very high (ICC=.87). The reliability of the overall performance in the road test was very high (ICC=.83). For the criterion validity of the road test, 78.9% of the subjects were correctly classified when the judgments of the principal evaluator were compared with outcomes of the SDSA. Agreement in classification between the principal evaluator and a state-registered evaluator's judgments was 81.6%. The sensitivity and specificity of the agreement were very high (80.6%) and perfect (100%), respectively.

CONCLUSIONS:

The road test is a reliable and valid test of driving ability after stroke.