Sex Differences in the Association Between 24-hour Rest-Activity Rhythms and Frailty Among U.S. Older Adults: Findings From NHANES 2011–2014

BACKGROUND: Little is known as to how rest-activity rhythms are associated with frailty and how this relationship differs by sex. This study examined the relationship between rest-activity rhythms and frailty in a nationally representative sample of U.S. older adults, focusing on the moderating role of sex. METHODS: 2 531 participants aged ≥60 years (females: 55.2%; frail: 5.15% [4.02-6.29]; pre-frail: 33.49% [31.29-35.68]) were included using the 2011-2014 National Health and Nutrition Examination Survey. Nonparametric rest-activity rhythms parameters, including inter-daily stability, intra-daily variability, relative amplitude, most active 10-hour, and least active 5-hour, were estimated from wrist-worn actigraphy data. Frailty status was assessed using a modified version of frailty phenotype (range: 0-5): frail (≥3), pre-frail (1,2), and non-frail (0). Multinomial logistic regression models were used to examine the interest of associations, adjusting for potential confounders. RESULTS: Frail and pre-frail older adults exhibited significantly lower relative amplitude, inter-daily stability, higher intra-daily variability, and phase delay when compared with non-frail older adults (p's < .05). Particularly, older adults with low relative amplitude had significantly greater odds of being frail and pre-frail (aOR [95% confidence intervals]; frailty: 5.60 [2.61-12.04]; pre-frailty: 1.58 [1.13-2.20]). Significant sex-interaction was observed (p < .01), with this association being greater in females than in males (aOR [95% confidence intervals]; females: 7.78 [2.98-20.30] for frailty, 2.31 [1.60-3.32] for pre-frailty; males: 4.48 [1.38-14.54] for frailty, 1.12 [0.61-2.07] for pre-frailty). CONCLUSION: Weakened rest-activity rhythms strength is unfavorably associated with frailty, particularly in females. Rest-activity rhythms may be a useful indicator associated with frailty in older adults, but sex-specific differences should be considered. Further longitudinal research is necessary to investigate the bidirectionality of their association.