Accelerometer-Derived Rest–Activity Rhythms and Sleep Disturbance in Older Adults: A Three-Wave Longitudinal Study
Abstract
Purpose: To examine associations between accelerometer-derived rest–activity indicators and sleep disturbance among older adults. Methods: This analysis included 1,373 person-wave observations from 569 National Health and Aging Trends Study participants. Multilevel models separated within-person and between-person components of the least active five-hour period (L5), coefficient of variation, and behavioral fragmentation. Relative amplitude (RA) was included at the between-person level in the primary model and examined separately in sensitivity analyses. Results: In primary analyses, no rest–activity indicator was associated with sleep disturbance, including when sleep medication use was excluded; within-person estimates were small (95% CI bounds within ±0.031). Weak between-person associations for L5 and RA emerged only in exploratory, model-dependent analyses. Pain, depressive symptoms, chronic disease burden, and educational attainment were associated with sleep disturbance. Conclusion: Rest–activity indicators showed no clear association with self-reported sleep disturbance, and small between-person estimates were model-dependent. Health-related and socioeconomic factors warrant attention in population-based sleep research.
Related
- Wrist-worn accelerometer-derived activity fragmentation and cause-specific mortality in US adults: a population-based prospective cohort study
- Accelerometer-derived ‘weekend warrior’ physical activity, genetic susceptibility, and the associated risk of osteoarthritis
- Accelerometer-derived moderate-to-vigorous-intensity physical activity: sex differences in health benefits and health equivalence between the sexes: a prospective cohort study
- Accelerometer-derived 24-hour behavior composition and incident clinically diagnosed sleep apnoea: a prospective cohort study
- Changes in accelerometer-measured physical activity and sedentary time among young people in Belgium between 2014-15 and 2022-23
BibTeX
@misc{lee2026accelerometerderived,
title = {{Accelerometer-Derived Rest–Activity Rhythms and Sleep Disturbance in Older Adults: A Three-Wave Longitudinal Study}},
author = {Kyungmi Lee and Jennifer B. Levin},
howpublished = {PsyArXiv},
year = {2026},
month = nov,
doi = {10.31234/osf.io/bwe7f_v1},
url = {https://osf.io/preprints/psyarxiv/bwe7f_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Accelerometer-Derived Rest–Activity Rhythms and Sleep Disturbance in Older Adults: A Three-Wave Longitudinal Study
AU - Kyungmi Lee
AU - Jennifer B. Levin
PY - 2026
DA - 2026/11/05
PB - PsyArXiv
DO - 10.31234/osf.io/bwe7f_v1
UR - https://osf.io/preprints/psyarxiv/bwe7f_v1/
AB - Purpose: To examine associations between accelerometer-derived rest–activity indicators and sleep disturbance among older adults. Methods: This analysis included 1,373 person-wave observations from 569 National Health and Aging Trends Study participants. Multilevel models separated within-person and between-person components of the least active five-hour period (L5), coefficient of variation, and behavioral fragmentation. Relative amplitude (RA) was included at the between-person level in the primary model and examined separately in sensitivity analyses. Results: In primary analyses, no rest–activity indicator was associated with sleep disturbance, including when sleep medication use was excluded; within-person estimates were small (95% CI bounds within ±0.031). Weak between-person associations for L5 and RA emerged only in exploratory, model-dependent analyses. Pain, depressive symptoms, chronic disease burden, and educational attainment were associated with sleep disturbance. Conclusion: Rest–activity indicators showed no clear association with self-reported sleep disturbance, and small between-person estimates were model-dependent. Health-related and socioeconomic factors warrant attention in population-based sleep research.
ER -