Do accelerometer-based physical activity patterns differentially affect cardiorespiratory fitness? A growth mixture modeling approach

Abstract

Findings on the association between cardiorespiratory fitness (CRF) and moderate-to-vigorous physical activity (MVPA) may be distorted if patterns of accumulated MVPA over a week exist but are ignored. Our aim was to identify MVPA patterns and to associate them to CRF. Two hundred twenty-four 40-75-year-old adults wore accelerometers for 7 days. CRF was measured by peak oxygen uptake (V'O_2,peak) assessed on a cycle ergometer via standardized cardiopulmonary exercise testing. Growth mixture modeling indicated four MVPA patterns: "low/stable" (57%, Mean MVPA time (M) = 21 min day^-1), "medium/stable" (20%, M = 46 min day^-1), "medium/weekend high" (14%, M = 47 min day^-1), and "high/weekend low" (9%, M = 71 min day^-1). V'O_2,peak was higher for persons with "high/weekend low" and "medium/weekend high" patterns compared to "low/stable" and "medium/stable" (p values < 0.001). The same total amount of MVPA may have greater benefit if performed on fewer days during the week but with a longer duration than if performed every day but with a lower duration.

Keywords: Adults; Aerobic capacity; Latent class; Motion sensor; Oxygen uptake; Unobserved heterogeneity.