Joint associations of accelero-meter measured physical activity and sedentary time with all-cause mortality: a harmonised meta-analysis in more than 44 000 middle-aged and older individuals.
Level 3 - non-randomized controlled study
Harmonised meta-analysis of prospective cohort studies
PubMed 33239356 · doi:10.1136/bjsports-2020-103270
What was done
A harmonised meta-analysis was conducted using data from nine prospective cohort studies across four countries. The sample included 44,370 middle-aged and older men and women followed for 4.0 to 14.5 years, during which 3,451 deaths occurred (7.8% mortality rate). Moderate-to-vigorous intensity physical activity (MVPA) and sedentary duration were measured objectively with accelerometers. Study-level associations between joint combinations of MVPA and sedentary time and all-cause mortality were analyzed using Cox proportional hazards regression and pooled with random-effects meta-analysis.
What was found
Across cohorts, mean sedentary time ranged from 8.5 to 10.5 hours/day, and MVPA ranged from 8 to 35 minutes/day. Compared to the referent group (highest MVPA, lowest sedentary time), mortality risk increased with lower MVPA and higher sedentary time. In the highest third of MVPA, mortality risk was not statistically different from the referent in the middle (risk increase 16%; 95% CI, 0.87% to 1.54% [reported as 0.87 to 1.54]) and highest (risk increase 40%; 95% CI, 0.87% to 2.26% [reported as 0.87 to 2.26]) thirds of sedentary time. In the lowest third of MVPA, mortality risk was significantly elevated across all sedentary time categories: 65% (95% CI, 1.25 to 2.19) for the lowest sedentary third, 65% (95% CI, 1.24 to 2.21) for the middle third, and 263% (95% CI, 1.93 to 3.57) for the highest third.
Why it matters
Device-measured activity data show that 30 to 40 minutes of daily moderate-to-vigorous physical activity attenuates the excess mortality risk of high sedentary time, a lower threshold than prior estimates based on self-report.
Limits
The included studies were observational cohorts, so residual confounding and reverse causation cannot be fully ruled out. Activity was measured at baseline rather than longitudinally over the 4.0 to 14.5 years of follow-up. Specific device models, wear protocols, cut-points, and demographic breakdowns were not reported in the abstract.
Cited by
- context Prolonged sedentary behavior increases health risks even in individuals who engage in regular exercise.