Validation of a Cognitive Self-Assessment Tool Simulating Japan's Official Digital Test for Older Drivers

Background A mandatory tablet-based cognitive function test for older drivers in Japan is employed for formal assessment only, terminating once a passing score is achieved and precluding a complete assessment. To bridge this gap between formal assessment and the need for self-preparation among older drivers, Nippontect Systems Co., Ltd., Japan, developed "MOGI, " a tablet-based application that allows users to experience the entire official test for self-assessment purposes. The objective of this study was to validate "MOGI" by examining its correlation with the Mini-Mental State Examination-Japanese version (MMSE-J). Methods We conducted a cross-sectional study at the Minato City Silver Human Resources Center in Tokyo and among outpatients at the Oyama Orthopedics and Internal Medicine Clinic in Tochigi Prefecture. The required sample size was calculated by assuming a specific correlation coefficient, significance level, and power. Community-dwelling volunteers and individuals clinically diagnosed with mild cognitive impairment (MCI) or mild-to-moderate dementia participated from February 3 to 17, 2025. All diagnoses were made by a neurologist based on criteria from the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition. Participants completed both the "MOGI" application, assessing memory and judgment via cued recall and time orientation tasks, and the MMSE-J. The agreement between automated scoring by "MOGI" and manual scoring by qualified staff was evaluated using the intraclass correlation coefficient (ICC). Spearman's rank correlation was used to examine the relationship between "MOGI" and MMSE-J scores, and differences in "MOGI" scores among MMSE-J-based groups (≥28, 24-27, and ≤23) were evaluated. Results The required sample size was 37, assuming a 0.5 correlation coefficient, 5% significance level, and 90% power. A total of 42 participants, including 17 male and 25 female participants, were included in the final analysis; their mean age was 76.4±8.2 years. Excellent agreement was observed between the automated and manual scoring systems (ICC = 0.97, 95% CI: 0.94-0.98). A significant, strong positive correlation was observed between the "MOGI" total score and the MMSE-J score (ρ = 0.64, p<0.001). "MOGI" also demonstrated excellent discriminative ability, with significant differences in scores among the three MMSE-J-based groups (p<0.001 among the three groups; p<0.05 for ≥28 vs. 24-27; p<0.001 for ≥28 vs. ≤23; p<0.01 for 24-27 vs. ≤23). Conclusion "MOGI" exhibits robust validity as a cognitive assessment tool, supported by a reliable automated scoring system. By providing a comprehensive assessment experience unavailable in the official test, "MOGI" serves as a valuable complementary tool for practice, self-monitoring, and a more nuanced understanding of one's cognitive function, potentially contributing to the early detection of cognitive decline.