Actigraphy-measured sleep characteristics and risk of falls in older women

Background: Prior studies have suggested that insomnia and self-reported poor sleep are associated with increased risk of falls. However, no previous study, to our knowledge, has tested the independent associations of objectively estimated characteristics of sleep and risk of falls, accounting for the use of commonly prescribed treatments for insomnia. Methods: Study subjects were participants in the Study of Osteoporotic Fractures. In 2978 primarily community-dwelling women 70 years and older (mean age, 84 years), sleep and daytime inactivity were estimated using wrist actigraphy data collected for a minimum of 3 consecutive 24-hour periods (mean duration, 86.3 hours). Fall frequency during the subsequent year was ascertained by a triannual questionnaire. Use of medications was obtained by examiner interview. Results: In multivariate-adjusted models, relative to those with "normal" nighttime sleep duration (>7 to 8 hours per night), the odds of having 2 or more falls in the subsequent year was elevated for women who slept 5 hours or less per night (odds ratio, 1.52; 95% confidence interval, 1.03-2.24). This association was not explained by the use of benzodiazepines. Indexes of sleep fragmentation were also associated with an increased risk of falls. For example, women with poor sleep efficiency (<70% of time in bed spent sleeping) had 1.36-fold increased odds of falling compared with others (odds ratio, 1.36; 95% confidence interval, 1.07-1.74). Conclusion: Short nighttime sleep duration and increased sleep fragmentation are associated with increased risk of falls in older women, independent of benzodiazepine use and other risk factors for falls.