Abstract
Time disorientation is commonly observed in dementia, however very little is known about the pathological significance of minor time errors in community-dwelling population. Our objective was to investigate the relationship between time orientation and risk of dementia in a population of older adults. Analyses relies on 8611 dementia-free subjects from the Three-City Study, France. Participants were followed up for 10 years for incident dementia. Time orientation was assessed by asking for the date, the day of the week, the month, the season and the year. At baseline, 905 subjects made at least one error in time orientation. During 57,073 person-years of follow-up, 827 participants developed dementia. After controlling for age, gender and education level, subjects with one error in time had a greater risk of dementia (hazard ratio [HR] 1.44 [1.18-1.77]), while those with at least 2 errors had a more than three-fold increased risk (HR 3.10 [1.98-4.83]). This association was particularly marked for the diagnosis of probable Alzheimer's disease. Time disorientation was associated with an increased risk of dementia in a large population of cognitively normal older people followed during up to 10 years and should not be underestimated in clinical setting.