Background: Evidence on sedentary behavior and cardiovascular disease (CVD) is largely based on self-reported sedentary time. Furthermore, how sedentary time is accumulated (in longer vs. shorter bouts) may be related to CVD risk but has not been tested.
Methods: Women (n=5638, mean age=79±7) with no history of myocardial infarction or stroke wore accelerometers for 4-7 days and were followed for up to 4 years for incident CVD. Hazard ratios (HR) and 95% confidence intervals (CIs) for CVD and coronary heart disease (CHD) events were estimated across quartiles of sedentary time and mean sedentary bout duration using Cox proportional hazard models adjusting for covariates. Separate models evaluated associations after adding moderate-to-vigorous physical activity (MVPA) and possible mediators: body mass index, diabetes, hypertension, systolic blood pressure, fasting glucose, HDL-cholesterol, and triglycerides. We then tested whether mean bout duration was associated with increased risk for CVD and CHD among women with above median sedentary time (≥10hr/day).
Results: Covariate-adjusted HRs for CVD and CHD increased across quartiles of both sedentary time and mean bout duration (Table). All CHD associations remained significant but attenuated after adjustment for possible mediators. After adjustment for MVPA, highest vs. lowest quartile HRs (CI) for CHD were 1.6 (0.7-3.4; p-trend = .08) for sedentary time and 1.8 (0.9-3.5; p-trend = .047) for mean bout duration. Among women with high sedentary time, the HRs (CI) comparing the 75th vs. 25th percentile of mean bout duration were 1.05 (0.95-1.15) for CVD and 1.16 (1.01-1.34) for CHD.
Conclusions: Both sedentary time and mean bout duration showed independent, dose-response associations with increased risk of CVD and CHD events in older women. Among women with high sedentary time, longer mean bout duration was associated with higher CHD risk. Taken together, this provides evidence that both total sedentary time and the way it is accumulated are predictive of incident CHD.