• elderly;
  • epidemiology;
  • incident dementia;
  • sleep duration

Background and purpose:  We determined in a population-based study whether sleep duration was associated with an increased risk of incident dementia.

Methods:  In a population-based study in central Spain, participants were evaluated at baseline and 3 years later. Baseline demographic variables were recorded and participants indicated their daily sleep duration as the sum of night-time sleep and daytime napping. The average daily total sleep duration was grouped into five categories: ≤5 (short sleepers), 6, 7 (reference), 8, and ≥9 h (long sleepers). We identified all cases with incident dementia, diagnosed using DSM-IV criteria.

Results:  Three thousand two hundred eighty six participants with baseline information about sleep duration had a median duration of follow-up of 3.2 years. There were 140 incident cases of dementia. The relative risks (RR) for short sleepers and for long sleepers were 2.36 (95% CI = 1.07–5.21, P = 0.03) and 2.40 (95% CI = 1.20–4.81, P = 0.01), respectively. After adjustment for potential confounders, the RR was only marginally increased for short sleepers (1.87, 95% CI = 0.85–4.15, P = 0.12) but remained increased for long sleepers (2.18; 95% CI = 1.09–4.37, P = 0.03).

Conclusions:  Prolonged sleep duration (night-time sleep and daytime napping) may be associated with an increased risk of dementia.