The impact of cognitive deficit on self-reported car crashes in ultra-octogenarian population: data of an Italian population-based study.

OBJECTIVE: To examine the usefulness of specific neurocognitive tests for predicting the crash involvement in ultra-octogenarian population. METHODS: A total of 800 subjects (mean age 82.4 + 3.1 years) underwent a battery of neuropsychological tests. Global intellectual functioning was assessed using the Mini Mental State Examination, mental flexibility and information processing speed were assessed using the Trail Making Test parts A and B (TMT-A and TMT-B), long-term memory was evaluated with the short story, and visuo-spatial skills were tested with Clock Drawing Test. One year after this evaluation, 343 (43%) participants have been interviewed by a telephone call to know if they were currently driving and if they had a car crash during this period. RESULTS: Two hundred ninety-seven subjects had their driving license renewed and completed the follow-up 1 year after. Data shows that less than 11% of this group had a car crash during the first year of observation (Crash Involved). Older subjects involved in a car crash showed significant worse performances on TMT-B (TMT-B pathological Crash Involved vs. Noncrash Involved 47% vs. 27%; p = 0.02) and on short story (short story pathological Crash Involved vs. Noncrash Involved 19% vs. 5%; p = 0.02). CONCLUSIONS: Trail Making test B and short story have been demonstrated to provide a predictive value of driving performance of older people. Therefore, we suggest that a simple and standardized battery of neuropsychological tests, lasting about 30 min and administered by an experienced staff, is a good diagnostic instrument for risk prevention of driving activity of older drivers.