Association of Accelerometry Assessed Physical Activity with Mortality Among a National Sample of United States Adults

Purpose: This study described the association of accelerometer-determined physical activity and sedentary behavior with all-cause mortality using a nationally representative sample from the United States.

Methods: Using 2003-04 National Health and Nutrition Examination Survey (NHANES) data, 4000 adults at least 18 years wore an ActiGraph accelerometer for one week, providing at least 3 compliant days of wear (>=8 hours/day). Cutpoints originally applied to NHANES defined moderate to vigorous physical activity (MVPA; >=2020 counts/minute) and sedentary behavior (<100 counts/minute). Mortality was verified through 2006 by National Death Index. Cox proportional hazards models provided adjusted hazard ratios (AHR, presented with 95% confidence intervals). All estimates were weighted to reflect the United States population.

Results: During an average of 2.8 years of follow-up, 157 deaths occurred. The sample averaged 312.1 counts/minute, with 22.7 minutes/day in MVPA, with 30% occurring in MVPA bouts (>=10 minutes allowing up to 2 minutes below the threshold). Sedentary behavior was 8.2 hours/day, with 43% occurring in sedentary bouts (>=30 minutes with at least 80% of minutes below the threshold). Higher average counts/minute were associated with lower mortality risk: highest quartile 0.16 (0.04, 0.55), 3rd quartile 0.25 (0.10, 0.61), 2nd quartile 0.38 (0.21, 0.71), and 1st quartile (referent). Similar significant reductions in risk of death were observed for higher MVPA and MVPA bouts. More time in sedentary behavior was associated with higher mortality risk: highest quartile 3.52 (1.34, 9.24), 3rd quartile 2.08 (0.81, 5.35), 2nd quartile 1.74 (0.66, 4.58), and 1st quartile (referent). Similar significant increases in risk of death were observed for longer time in sedentary bouts.

Conclusions: Higher overall physical activity and MVPA were associated with a reduction in mortality risk, while higher sedentary behavior was associated with an increase in mortality risk. Longer follow-up time will provide more robust measures of association and allow sensitivity analyses to exclude the immediate follow-up period.