Longitudinal Associations of Rest-Activity Rhythm Profiles with Perceived Physical Fatigability

TL;DR

This study shows that disrupted rest-activity rhythms are linked to increased physical fatigue over time in older adults.

Contribution

The study identifies specific rest-activity rhythm profiles and their longitudinal associations with perceived physical fatigability in older adults.

Findings

The Low-Fragmented rest-activity rhythm group had higher baseline physical fatigability and a steeper increase over time.

Adjusting for fitness and step count confirmed the association between disrupted rhythms and worsening fatigue.

RAR profiles were identified using clustering, revealing distinct activity patterns linked to fatigue levels.

Abstract

Rest-activity rhythms (RARs), 24-hour activity and sleep patterns, are cross-sectionally associated with fatigability, a prognostic marker of functional decline. Participants (N = 821, age=76.4±5.0 years) from the Study of Muscle, Mobility and Aging had perceived physical fatigability measured using the validated 10-item Pittsburgh Fatigability Scale (PFS; 0-50, higher score=greater fatigability) at baseline and up to three follow-up visits over 2.6±0.4 years. RARs were derived from 7-day wrist-worn accelerometry using the extended cosine model (parametric: amplitude, acrophase, alpha, beta) or non-parametric approaches (interdaily stability, intradaily variability, relative amplitude). Three RAR profiles (Synchronized, Concentrated High-Activity, Low-Fragmented) were identified using the k-means clustering algorithm. Compared to Synchronized, the Concentrated High-Activity group had the highest amplitude (5,176 vs. 2,258 counts/min) and earliest acrophase (11:54AM vs. 2:18PM). The Low-Fragmented group vs. Synchronized had the lowest amplitude (1,462 vs. 2,258 counts/min), interdaily stability (0.46 vs. 0.63), and highest intradaily variability (1.10 vs. 0.81). PFS score varied by RAR groups: Synchronized=15.2±8.5, Concentrated High-Activity=14.2±7.9, and Low-Fragmented=17.3±8.7, p = 0.004. Using linear mixed models, the Low-Fragmented group had higher perceived physical fatigability at baseline (β = 1.30, p = 0.046) and a steeper increase in PFS score over time (β = 0.46, p = 0.03) vs. Synchronized, adjusted for demographics, health conditions, lifestyle factors, and sleep efficiency. After further adjustment for cardiorespiratory fitness (VO2peak from treadmill testing) and daily step count, the Low-Fragmented group showed a significantly larger increase in PFS score over time (β = 0.46, p = 0.03; β = 0.49, p = 0.02), although the baseline association was attenuated. Our findings reveal that disrupted RARs are independently associated with persistently worsening perceived physical fatigability.