Abstract P132: Self-reported Recreational Physical Activity And Incident Venous Thromboembolism Risk In The Women’s Health Initiative

Abstract

Introduction: The association between physical activity (PA) and incident venous thromboembolism (VTE) is inconsistent and incompletely characterized. Hypothesis: Higher amounts of self-reported total PA will be associated with a lower risk of incident VTE. Methods: Eligible women were Women’s Health Initiative (WHI) Hormone Therapy (HT) trial and Observational Study (OS) participants without prior VTE, able to walk at least one block, with baseline PA data (n=110,631). At multiple timepoints, women reported their current usual frequency and duration of walking and low, moderate, and strenuous intensity recreational PA, which were used to estimate energy expended. VTE, including deep vein thrombosis and pulmonary embolism, were validated in the WHI-HT and self-reported in the WHI-OS. Multivariable-adjusted (Table) Cox Proportional Hazards models estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for the association of time-varying total PA and incident VTE risk, with women censored at death, loss to follow-up, and three years after their final PA measure. Results: At baseline, women were 64 years of age on average. Over an average of 9.9 years of follow-up, 1,403 women experienced incident VTE. In primary confounder-adjusted models, compared with women who reported no PA, higher tertiles of total PA were associated with a lower risk of incident VTE (high vs. no PA HR=0.60; 95% CI: 0.51, 0.71; p-trend<0.001) (Table). Further adjustment for potential mediators on the causal pathway (physical function, self-reported health, body mass index [BMI], cardiovascular-related factors) attenuated but did not eliminate this association (high vs. no PA HR=0.81; 95% CI: 0.68, 0.97; p-trend=0.017). Conclusions: Compared to no PA, higher tertiles of total PA were associated with a lower risk of incident VTE, even after adjusting for potential mediators reflecting physical function, health, BMI, and cardiovascular risk. Our results add to the apparent health benefits of even moderate amounts of recreational PA.