The association between benzodiazepine use and falls, and the impact of sleep quality on this association: data from the TILDA study

Abstract

Summary Background Benzodiazepines (BZD) are associated with adverse effects, particularly in older adults. Aim This study assesses the association between BZD use and falls, and the impact of sleep quality on this association, in community dwelling adults aged over 50. Design Cross-sectional analysis of data from wave 1 of The Irish Longitudinal Study on Ageing. Methods Participants were classed as BZD users or non-users and asked if they had fallen in the last year, and whether any falls were unexplained. Sleep quality was assessed via self-reported trouble falling asleep, daytime somnolence and early-rising. Logistic regression assessed for an association between BZD use and falls, and the impact of sleep quality on this association was assessed by categorizing based on BZD use and sleep quality variables. Results Of 8175 individuals, 302 (3.69%) reported taking BZDs. BZD use was associated with falls, controlling for confounders [Odds Ratio (OR) 1.40; 1.08, 1.82; P-value 0.012]. There was no significant association between BZDs and unexplained falls, controlling for confounders [OR 1.41; 95% Confidence Interval (CI) 0.95, 2.10; P-value 0.09]. Participants who use BZDs and report daytime somnolence (OR 1.93; 95% CI 1.12, 3.31; P-value 0.017), early-rising (OR 1.93; 95% CI 1.20, 3.11; P-value 0.007) or trouble falling asleep (OR 1.83; 95% CI 1.12, 2.97; P-value 0.015), have an increased odds of unexplained falls. Conclusion BZD use is associated with falls, with larger effect size in those reporting poor sleep quality in community dwelling older adults. Appropriate prescription of medications such as BZDs is an important public health issue.