RESUMEN
BACKGROUND: No papers have examined the relationship between socio-demographic characteristics and cognitive performance in oldest old subjects (i.e, > = 80 years old) asking for driving license renewal. We hypothesize that, even in this highly functioning population, age, sex, and education influence cognitive performance, expressed as total or single domain (raw) test scores. This research question allows to describe, identify, and preserve independence of subjects still able to drive safely. METHODS: We examined cross-sectionally a cohort of > = 80 years old subjects (at enrollment) asking for driving license renewal in the Milan area, Italy, 2011-2017. The analysis was restricted to 3378 first and 863 second visits where individual's cognitive performance was evaluated. According to the study protocol, the Mini Mental State Examination (MMSE) test was administered at the first visit for driving license renewal and the Montreal Cognitive Assessment (MoCA) test at the second visit, following an additional renewal request. Ordinary least squares regression models were fitted at either time points. In each model, we included age, sex, and education as independent variables, whereas the dependent variable was total or single domain score for either test. In total, we fitted 15 regression models to assess our research hypothesis. RESULTS: The median subject in our sample reached the maximum scores on domains targeting operational and tactical abilities implied in safe driving, but had sub-optimal scores in the long-term memory domain included among the strategic abilities. In multiple models, being > = 87 (versus 80- < 86 years old) significantly decreased the mean total and memory scores of MMSE, but not those of the MoCA. Females (versus males) had significantly higher mean total and long-term memory scores of either tests, but not other domains. Mean total and single domain scores increased for increasing education levels for either tests, with increments for high school graduates being ~ 2 of those with (at most) a junior high school diploma. CONCLUSIONS: Sex and education, as well as age to a lesser extent, predict cognitive functioning in our oldest old population, thus confirming that concepts like cognitive reserve and successful ageing are valuable constructs in the identification of older subjects still able to drive.