Disclosures: Mohammed M. Alshehri, PT, MSc: Nothing to disclose

Objective: To examine the association between sleep efficiency (SE) variability and number of falls, and to identify the cutoff score for SE variability that predict multiple falls (>2 falls) in general population.

Design: Retrospective cross-sectional.

Setting: Midlife in the United States-2 database, 2004-2009, a multisite longitudinal study.

Participants: Data from 342 participants (mean age 53.2511.29, 63.5% females) were included.

Interventions: N/A

Main Outcome Measures: The SE is the ratio between total sleep time and total bed time multiply by 100, which was measured using actigraph for 7 nights. SE variability was calculated by coefficient of variance = (7 nights standard deviation of SE divided by 7 nights mean of SE) multiply by 100. Self-reported falls in the past 12 months was obtained as a continuous variable. Number of falls was used for multiple negative binomial regression analysis. Fallers categories (i.e. non-fallers or single fallers versus multiple fallers) were used to determine the cutoff score for SE variability using area under the curve (AUC) and receiver operator characteristics. Youden index was calculated and largest Youden index was chosen to determine the cutoff score. Alpha was set at 0.05.

Results: Mean value for SE variability was 10.65% (ranged from 1% to 69%). Number of falls was significantly associated with increased SE variability (incidence rate ratio = 1.02, 95% confidence interval [1.01-1.04], P=.027) after controlling for age, gender, body mass index, depression and total number of chronic conditions. The results found that the cutoff score for SE variability that differentiate multiple falls (>2 falls) was 7.38% (sensitivity 0.67; specificity 0.58), and an AUC of 0.61.

Conclusions: This study found that increasing the SE variability was associated with number of falls in general population. The cutoff score of 7.38% for SE variability differentiates multiple fallers from single/non-fallers. High SE variability could be screened and treated for people who are at risk of fall.

Level of Evidence: Level III

« Back to AAPM&R Annual Assembly 2019

PM&R Meeting Abstracts - https://pmrjabstracts.org/abstract/the-association-between-sleep-efficiency-variability-and-falls-in-general-population/