ABSTRACT

Objective : The study aims to predict the risk factors during driving situations, and verify the usability of cognitionassessment tools that may be used to check the level of danger when driving for the safety of older drivers withMCI. Methods : A meta-analysis was performed on 17 RCT studies conducted for driving and cognition assessments byclassifying the “MCI” and “normal” groups. A quantitative analysis was also conducted based on the statisticalheterogeneity, size effect, sensitivity, and publication bias for every cognition assessment tool and risk factor. Results : A Jadad assessment was conducted to assess whether the selected studies were of high quality. In termsof the effective size of each tool, TMT-A was 0.44, TMT-B was 0.54, and UFOV-sub-test 2 was 0.52, whichcan be interpreted as having a “large size effect.” For the size effect, the driving error was 0.83, the vehicle speedwas –0.17, the reaction time was 0.70, the braking was 0.47, and the turning was – 0.81. Therefore, the drivingerror and turn were interpreted as having a “large size effect”, and the reaction time and braking were interpretedas having a “medium size effect” and “small size effect,” respectively. Conclusion : TMT-A, TMT-B, and UFOV–sub-test 2 are useable to verify the driving danger of MCI drivers. Inaddition, driving error is useable as additional information for safe driving.