Journal of Sport and Health Science · Apr 14, 2026

Bidirectional reallocations of sitting time, physical activity, and sleep in relation to risk of all-cause mortality and cardiovascular disease: An analysis of the Prospective Urban Rural Epidemiology (PURE)-China cohort study

By Huang Y, Hao J, Hu B et al.

BACKGROUND: Prolonged sedentary time is associated with adverse outcomes, but evidence in low-exposure ranges remains limited. The health effects of reallocating time between sitting, physical activity, and sleep have not been comprehensively examined. METHODS: We analyzed 41,733 adults (mean age = 50.6 years) from the Prospective Urban Rural Epidemiology (PURE)-China cohort, recruited between 2005 and 2009 and followed for a median of 11.9 years. Sitting, physical activity, and sleep were assessed using validated questionnaires. The primary outcome was a composite of all-cause mortality and major cardiovascular events. Cox frailty and isotemporal substitution models were applied. RESULTS: Median sitting time was 3.0 h/day (interquartile range (IQR): 1.7-4.6). Sitting showed a J-shaped association with outcomes, with lowest risk around 4 h/day. Both low (<2 h/day) and high (≥6 h/day) sitting were associated with higher risk. Among participants sitting ≥4 h/day, replacing 30 min of sitting with physical activity was associated with a 3%-4% lower risk of the composite outcome, with stronger associations observed for all-cause mortality (6%-7%). In contrast, among those sitting <4 h/day, reallocating 30 min of physical activity or prolonged sleep to sitting was associated a 4%-6% lower risk of the composite outcome and a 4%-10% lower risk of all-cause mortality. CONCLUSION: Sitting time demonstrates context-dependent associations with health. The finding that moderate sitting was protective in highly active individuals reflects a potential "sitting paradox". Our study highlights the bidirectional effects of reallocating sitting, activity, and sleep, underscoring the need for more context-specific guidance.

doi.org/10.1016/j.jshs.2026.101140Open accessPubMed 41985744Populations

AgingResearch.News