Self-reported sleep duration and falls in older adults.

Whether the association between sleep duration and accidental falls is independent of sleep quality is uncertain. It is also unclear if this association varies with age and sex among older adults. Data were collected through telephone interviews of 1542 community-dwelling individuals aged ≥68 years in Spain. Analyses were performed with logistic regression and adjusted for the main confounders, including lifestyle, health status, comorbidity and sleep quality, as measured by nighttime and daytime sleep complaints. Analyses were also stratified by age and sex. In total, 449 (29.1%) participants fell in the previous year; of these, 57.7% had one fall and 42.3% had recurrent (≥2) falls. As compared with those who usually slept 7-8 h, those sleeping ≥11 h were more likely to suffer recurrent falls [odds ratios (OR) 2.75; 95% confidence intervals (CI) 1.32-5.62]. In the stratified analysis, there was no association between sleep duration and falls in those aged ≤75 years and in men. In contrast, among those aged >75 years, the percentage of individuals with ≥1 falls was higher in those sleeping ≤5 h (OR 2.15; 95% CI 1.20-3.87) and ≥11 h (OR 2.34; 95% CI 1.17-4.68) than in those sleeping 7-8 h; also, women who slept ≥11 h were more likely to fall ≥1 times (OR 3.89; 95% CI 1.74-8.69). In conclusion, long sleep duration was associated with higher frequency of falls in older adults, even after adjustment for a wide range of sleep quality indicators. The association is clearer in women and the oldest adults. Assessment of sleep duration can help identify older adults who might benefit from interventions aimed at fall prevention.