Associations between physical activity and all-cause and cardiovascular mortality in adults with type 2 diabetes mellitus: A prospective cohort study from NHANES 2007–2018

To investigate the dose-response association between physical activity and all-cause and cardiovascular mortality in adults with type 2 diabetes mellitus and the effects of replacing sedentary behavior with physical activity. 4808 adults with type 2 diabetes mellitus were included in NHANES 2007–2018. Cox proportional hazards models were used to calculate hazard ratios and 95% confidence intervals. Isotemporal substitution analyses were further to determine the possible benefit of replacing sedentary time. During a median follow-up of 6.58 years, 902 deaths occurred, including 290 deaths from cardiovascular disease. Compared with the inactive group, the low-active and high-active groups were associated with declined risks of all-cause mortality [HRs (95% CIs) 0.64 (0.50, 0.83); 0.60 (0.50, 0.73), respectively] and cardiovascular mortality [0.50 (0.29, 0.88); 0.54 (0.39, 0.76)), respectively]. Dose-response analysis showed a significant U-shaped curve between physical activity and all-cause and cardiovascular mortality. Replacing 30 min/day of sedentary time with physical activity was substantially linked to a reduced risk of 8–32% mortality. A high level of PA of 40.52 and 31.66 MET-h/week was respectively related to the lowest risk of all-cause and cardiovascular mortality. Replacing sedentary time with physical activity could benefit the type 2 diabetes mellitus population. •Low and high physical activity levels were associated with decreased mortality risk.•A U-shaped curve was observed between physical activity and mortality.•Replacing sedentary time with physical activity reduced mortality risk.•A high physical activity of nearly 40.52 and 31.66 MET-h/week was linked to the lowest all-cause, cardiovascular mortality.