Back to Search Start Over

Association between circadian physical activity patterns and mortality in the UK Biobank

Authors :
Michael J. Stein
Hansjörg Baurecht
Anja M. Sedlmeier
Julian Konzok
Patricia Bohmann
Emma Fontvieille
Laia Peruchet-Noray
Jack Bowden
Christine M. Friedenreich
Béatrice Fervers
Pietro Ferrari
Marc J. Gunter
Heinz Freisling
Michael F. Leitzmann
Vivian Viallon
Andrea Weber
Publication Year :
2022
Publisher :
Cold Spring Harbor Laboratory, 2022.

Abstract

ImportanceThe benefit of physical activity (PA) for increasing longevity is well-established, however, the impact of diurnal timing of PA on mortality remains poorly understood.ObjectiveTo derive PA patterns and investigate their associations with all-cause mortality.Design, Setting, and ParticipantsThis population-based prospective cohort study analyzed UK Biobank baseline data collected between 2006 and 2010 from adults aged 40 to 79 years in England, Scotland, and Wales. Participants were invited by email to participate in an additional accelerometer study from 2013 to 2015, 7 years (median) after baseline. Participants’ vital status was assessed via linkage with mortality registries through September 2021 (England/Wales) and October 2021 (Scotland). Data analyses were performed in July 2022.ExposureLoading scores of functional principal components (fPCs) obtained from wrist accelerometer-measured activity metrics. The ‘Euclidean norm minus one’ was used as a summary metric of bodily acceleration aggregated to 24 hourly averages across seven days. These timeseries were used for functional principal component analysis (fPCA).Main Outcomes and MeasuresExamination of time-dependent PA patterns obtained using functional principal component analysis in relation to all-cause mortality estimated by multivariable Cox proportional hazard models.ResultsAmong 96,361 participants (56% female), 2,849 deaths occurred during 6.9 (SD 0.9) years of follow-up. Four distinct functional principal components (fPCs) accounted for 96% of the variation of the accelerometry data. The association of fPC1 and mortality was non-linear (pConclusions and RelevanceHigher risks of death were found for patterns denoting lower overall PA and higher late day and nocturnal PA. Conversely, higher levels of PA, distributed continuously, in one, or in two activity peaks during daytime, were inversely associated with lower mortality. Daily timing of PA may have public health implications, as our results suggest that some level of elevated PA during the day and a nighttime rest is associated with longevity.

Details

Database :
OpenAIRE
Accession number :
edsair.doi...........87611ad49f21ca7a8e3bc2285b5976cd