Association between physical activity and all-cause mortality: A 15-year follow-up using a compositional data analysis

The association between the composition of movement behaviors and mortality risk, acknowledging the composition nature of daily time data, is limited explored. The aim was to investigate how the composition of time spent in sedentary behaviors (SB), light intensity physical activity (LIPA), and moderate-to-vigorous physical activity (MVPA) is associated with all-cause mortality, in a cohort with 15 years follow-up time, using compositional data analysis. Eight hundred fifty-one participants (56% women, mean age 53 years) provided objectively assessed physical activity data using an ActiGraph accelerometer and were followed for 15 years. Association of daily time composition of movement behaviors with risk of mortality was explored using compositional data analysis and hazard ratios (HR) of mortality were estimated based on a cox regression model. A significant (P < .001) positive association between time spent in SB relative to time in other behaviors and a significant (P = .018) negative association between time spent in LIPA relative to time in other behaviors, with all-cause mortality, were found. Substituting time spent in LIPA or MVPA with time in SB increased the hazard for all-cause mortality, with greater effect found for MVPA (20 minutes replacement; HR 1.26, 95% CI 1.04-1.52) than for LIPA (20 minutes replacement; HR 1.06, 95% CI 0.65-1.73). In a public health perspective, it is recommended to substitute SB with either LIPA or MVPA, but for individuals with little time spent in MVPA, the most important message may be to try to maintain that behavior.