Association of wearable-device-measured fragmented sedentary behavior with cardiovascular disease, mortality, and life expectancy.

Abstract

Objectives: It remains unclear whether the degree of fragmentation modifies the adverse health outcomes associated with prolonged sedentary behavior. This study aims to investigate the associations of objectively measured sedentary patterns with incident cardiovascular disease (CVD), mortality, and life expectancy.

Methods: This prospective cohort study included 89 010 UK Biobank participants (median age 63.5 years, 56.5% women) who provided 3 to 7 valid days of wrist accelerometer data. Daily sedentary time and number of sedentary breaks were the metrics of interest. Four overall sedentary behavior patterns were identified by joint classification of sedentary time and number of sedentary breaks. Outcomes were all-cause mortality, CVD incidence and mortality, CVD-free life expectancy, and total life expectancy. Multivariable Cox proportional hazards models were used to evaluate the associations of sedentary behavior metrics with all-cause mortality and CVD incidence, whereas Fine-Gray competing-risk models were used to evaluate their associations with cardiovascular mortality, with non-CVD death treated as a competing event. Restricted cubic splines were used to assess potential nonlinear associations. Multi-state survival analyses were performed to estimate CVD-free life expectancy. Isotemporal substitution models were used to estimate the effects of replacing sedentary time with moderate-to-vigorous physical activity (MVPA).

Results: During a median follow-up of 8.0 years, 3 730 deaths occurred, including 1 223 CVD deaths and 8 883 incident CVD cases. Total sedentary time showed J-shaped dose-response associations with all-cause and cardiovascular mortality and a linear association with increased CVD incidence. The number of sedentary breaks showed L-shaped associations with risks of all-cause mortality and CVD mortality. Compared to the least sedentary group, participants with prolonged sedentary time (≥10 h/d) and a low degree of fragmentation (sedentary break <12 bouts/d) had a 51.5% higher risk of all-cause mortality [hazard ratio (HR)=1.515, 95% confidence interval (CI) 1.380 to 1.663], a 55.3% higher risk of cardiovascular mortality (subdistribution HR=1.553, 95% CI 1.325 to 1.820), and a 10.0% higher risk of incident CVD (HR=1.100, 95% CI 1.030 to 1.174). At age 50, the loss of life expectancy was primarily observed in CVD-free years (men: -2.770 years, 95% CI -3.392 to -2.148; women: -3.113 years, 95% CI -3.794 to -2.432). Notably, a higher frequency of sedentary breaks attenuated the increased mortality risk and the loss of CVD-free life expectancy associated with prolonged sedentary time. Nevertheless, replacing sedentary behavior with MVPA reduced the risk of mortality and CVD incidence among all sedentary participants, regardless of levels of fragmentation.

Conclusions: Prolonged sedentary time and its uninterrupted accumulation patterns are associated with greater risks of mortality and CVD, alongside reductions in CVD-free life expectancy. The increased mortality risk and the loss of CVD-free years may be mitigated by frequent interruptions to sedentary behavior.