Examining reactivity to the measurement of physical activity and sedentary behavior among women in midlife with elevated risk for cardiovascular disease

Objectives To estimate the extent of reactivity to measurement of physical activity (PA) and sedentary behavior among women in midlife with elevated risk for cardiovascular disease (CVD). Design Secondary analysis of a 10-day observational study of PA and sedentary behavior. Main Outcome Measures PA (steps, minutes of light PA, total minutes of moderate-to- vigorous PA [MVPA]) and percent time in sedentary behavior per day were assessed using ActiGraph GT3X tri-axial accelerometers in 75 women in midlife with elevated CVD risk (e.g. hypertension; M-Age = 51.61, M-BMI = 34.02 kg/m(2)). Two-level multilevel models were used to test for evidence of reactivity, with the addition of random effects to test for evidence of individual differences in observed trends. Results All outcomes showed linear trends across days (ps < 0.001), though this masked what appeared to be meaningful dropoff after Day 1 or Day 2 (with little difference between subsequent days; srs ranging from 0.15 to 0.32). The random effect was significant only for percent time in sedentary behavior (chi(2)[1] = 10.40, p = 0.02). Conclusions Consistent small to medium effects were found for all PA and sedentary behavior outcomes, underscoring the importance of considering measurement reactivity in populations with elevated CVD risk.