ABSTRACT
Objective
Examine the bidirectional relationships between within-person day-to-day fluctuations in physical activity (PA) and sleep characteristics among people with heart failure (HF) and insomnia.
Participants
Ninety-seven community-dwelling adults [median age 61.9 (interquartile range 55.3,70.9) years, female 41%] with stable HF and insomnia (insomnia severity index >7).
Methods
This sub-study longitudinally analyzed 15 consecutive days and nights of wrist actigraphy recordings, that were collected for baseline data prior to participation in a randomized controlled trial of cognitive behavioral therapy for insomnia. We used two-level mixed models of within- (daily) and between-participants variation to predict daytime PA counts/minutes from sleep variables (total sleep time, sleep efficiency) and predict sleep variables from PA.
Results
PA counts/minutes were low compared to prior cohorts that did not have HF (209 (166,259)) and negatively associated with NYHA class (standardized coefficient βs = −0.14, p < .01), age (βs = −0.13, p = .01), comorbidities (βs = −0.19, p < .01), and body mass index (βs = −0.12, p = .04). After adjustment for all significant covariates, the within-participant association of total sleep time with next-day PA was estimated to be positive among participants with NYHA class II–IV HF (βs = 0.09, p = .01), while the within-participant association of PA with same-night total sleep time was estimated to be positive among participants aged ≥60 years (βs = 0.10, p = .03).
Conclusions
Depending upon age and HF class, daytime PA was associated with longer same-night sleep and/or longer sleep was associated with greater next-day PA. Among those with more advanced HF, realistic sleep improvements were associated with clinically meaningful PA gains the next day.
Acknowledgments
We thank the investigators and staff of the parent study that provided these data: James W. Darden IV, Jessica Kelly-Hauser, Beeba Mathew, Patrick A. Richardson, Edward C. Gaiser, Eileen Condon, Laura Andrews, Lisa Finoia, and Joanne Iennaco.
Disclosure statement
No potential conflict of interest was reported by the authors.
Data availability
Data sharing is not applicable to this article as no new data were created or analyzed in this sub-study. The data from the parent study are pending publication.