Discriminant Validity of the Standardized On-Road Assessment for Drivers (SOAD) Among Stroke Patients in Japan

Introduction: On-road tests are considered the gold standard for evaluating real-world driving skills. However, their reliability and validity remain inadequately established, particularly under varying legal and road conditions across countries.

Aim: This study investigates the discriminant validity of the closed-course version of the Standardized On-Road Assessment for Drivers (SOAD) in Japan.

Methods: This study was conducted in five Japanese rehabilitation hospitals and affiliated driving schools. The participants consisted of 108 brain-injured individuals (mean age: 50.0 years) undergoing driving assessments. The inclusion criteria focused on physician-referred patients diagnosed with brain injuries. The SOAD closed-course test, consisting of 40 items, was compared with off-road cognitive assessments, including the Mini-Mental State Examination Japanese Version (MMSE-J), Trail Making Test Japanese Version (TMT-J), Rey-Osterrieth Complex Figure Test, Stroke Drivers Screening Assessment Japanese version (J-SDSA), and Kohs Block Design Test. Spearman's correlation was used to evaluate discriminant validity, distinguishing driving-specific skills from cognitive functions.

Results: Weak to moderate correlations were found between SOAD and off-road tests, supporting the discriminant validity of SOAD. Among off-road tests, the J-SDSA dot time correlated most frequently with SOAD items, followed by MMSE-J and TMT-J. The highest correlation coefficient (-0.38) was observed between the J-SDSA dot error and a specific SOAD item.

Conclusion: These results show that SOAD demonstrates strong discriminant validity as a closed-course on-road assessment tool for brain-injured individuals and measures unique aspects of driving skills not captured by cognitive tests.