Over a median 8.0 years, 1,634 of 65,028 German National Cohort (NAKO) adults aged 19-74 with hip accelerometry died. In Cox models, activity volume at the 75th versus 25th percentile had a hazard ratio of 0.73. Adjusted for volume, intensity of the day's most active 1 and 120 minutes gave hazard ratios of 0.58 and 0.69. Within continuous bouts they weakened to 0.71 and 0.83, and links lasted to 120 minutes only for mostly intermittent accumulation. In continuous bouts, intensity was linked to lower mortality at low volume, but at high volume only for longer bouts. Activity was measured over one baseline week, and hip-worn devices may miss cycling, swimming and upper-body movement.
Why it is interesting: Most mortality evidence on activity relies on self-report or intensity cut-points. This study separates how much people move from how hard and in what pattern, within one large device-measured cohort.
Abstract — the authors' own words, CC BY
Background
Greater physical activity volume is associated with lower mortality, but how activity intensity and its accumulation throughout the day relate to mortality beyond volume remains unclear. Methods
We prospectively analyzed hip-worn accelerometer-derived physical activity in 65,028 participants aged 19-74 years from the German National Cohort (NAKO). Using multivariable-adjusted Cox models, hazard ratios (HR) and 95% confidence intervals (CI) for all-cause mortality were estimated. Models of daily intensity and daily accumulation (most active 1-120 min of the day, sporadically distributed or within continuous bouts) were additionally adjusted for total activity volume. Results
During a median follow-up of 8.0 years, 1,634 deaths occurred. Physical activity volume showed a linear inverse association with mortality (HR comparing 75th vs. 25th percentile = 0.73 [95% CI = 0.67-0.80]). Independent of activity volume, higher activity intensity during the most active minutes of the day was inversely associated with mortality, with stronger associations at shorter than at longer durations (e.g., HR = 0.58 [95% CI = 0.53-0.63] for the most active 1 min vs. HR = 0.69 [95% CI = 0.65-0.74] for 120 min). Associations were attenuated when these minutes were accumulated in continuous bouts (e.g., HR = 0.71 [95% CI = 0.66-0.77] for 1 min vs. HR = 0.83 [95% CI = 0.76-0.91] for 120 min). In analyses stratified by overall activity volume, higher activity intensity within continuous bouts was associated with lower mortality risk at low activity volume, whereas associations at high activity volume were absent at shorter but inverse at longer bout durations. Conclusions
Beyond total activity volume, intensity patterns of activity accumulation were associated with mortality risk. Inverse associations were strongest at shorter durations and for intermittent rather than continuous accumulation. Higher activity intensity was more strongly associated with lower mortality among individuals with low activity volume, particularly at shorter bout durations. Together, these findings support a multidimensional view of physical activity in which volume, intensity, and accumulation pattern provide complementary information for mortality risk.
Cite this item
BibTeX
Free sign-up. A digest of the new papers twice a week, grouped by domain.