Background: Cardiovascular disease (CVD) remains the leading cause of death in the US. Although physical activity (PA) has been inversely associated with the risk of CVD, few studies have examined whether sedentary behaviors modify such association. Our goal was to examine associations of leisure time PA with risk of CVD and mortality and the role of sedentary behavior as potential effect modifier among US veterans.
Methods: We analyzed self-reported data on leisure time PA, television watching, and time spent on the computer among 438,364 participants of the Veterans Affairs Million Veteran Program from 2011 to 2018. We calculated metabolic equivalent of task-hours per week (MET-h/week) for each person and used electronic health record data to ascertain CVD.
Results: Mean age was 64.6 ± 12.6 years and 92% were men. During a mean follow up of 3.3 years, we observed 22,942 new cases of CVD and 48,325 deaths. There was an inverse relation of leisure time PA with CVD and total mortality [HR: 0.96 (0.95-0.97) and 0.91 (0.90-0.92) per 2 MET-h/week increment for CVD and total mortality, respectively]. The associations of PA with both incident CVD and mortality were stronger in participants who spent more time watching television or on computer (all p values for interaction < 0.01). No interaction of PA with time spent on video game was observed (p>0.05).
Conclusions: Leisure time PA is inversely associated with risk of CVD and mortality among US veterans and such relations were stronger in participants who spent more time watching television or on computer.
Keywords: Physical activity; cardiovascular disease; epidemiology; risk factors.