Red Light, Green Light: Activity Fragmentation and All-Cause Mortality in United States Adults.

PURPOSE: Physical-activity epidemiology treats how much people move as the health-relevant exposure. We asked whether activity fragmentation, the probability that movement gives way to rest, predicts all-cause mortality independently of activity volume, and how the two compare as predictors. METHODS: In a prospective cohort analysis, we pooled the 2003-2004 and 2005-2006 cycles of the National Health and Nutrition Examination Survey, which used the same hip-worn accelerometer and protocol, and linked them to mortality through 2019. Adults aged 20 years or older with at least four valid wear days were eligible (N = 5,973). Fragmentation was the active-to-sedentary transition probability. Survey-weighted Cox models adjusted for volume, demographic and socioeconomic characteristics, smoking, alcohol, hypertension, diabetes and cardiovascular history. RESULTS: Over a median 14.4 years, 1,429 participants died. Each standard-deviation increase in fragmentation carried a 19% higher hazard of death (hazard ratio 1.19, 95% confidence interval 1.09-1.30). Volume remained protective in the same model (0.83, 0.75-0.92). Fragmentation added more to a model containing volume than volume added to one containing fragmentation. Higher education predicted more fragmented movement; income did not. CONCLUSIONS: Activity fragmentation carries mortality information that activity volume does not, and is the stronger of the two predictors.