Dose–response relationships of accelerometer‐measured sedentary behaviour and physical activity with non‐alcoholic fatty liver disease

Abstract

Although a few studies have confirmed the association of accelerometer-measured sedentary behaviour (SB) and physical activity (PA) with non-alcoholic fatty liver disease (NAFLD), PA intensity and co-dependent daily time-use of movement behaviours are yet to be studied. To cross-sectionally examine the dose-response relationship between accelerometer-measured SB or PA and NAFLD using cubic spline analysis and the interdependence of movement behaviours over 24hours with compositional data analysis. Data were obtained between May 2017 and February 2020 from 1914 people who were not heavy alcohol drinkers using health check-ups at the Meiji Yasuda Shinjuku Medical Center, Tokyo. SB, light-intensity PA (LPA) and moderate- to vigorous-intensity PA (MVPA) were evaluated using a triaxial accelerometer. Fatty liver was diagnosed by ultrasonography. A multivariable-adjusted logistic model showed an inverse association between MVPA and NAFLD (odds ratio [OR] per 600 metabolic equivalents [MET]-min/week=0.85, 95% confidence interval [CI]=0.76-0.96). MVPA showed rapidly decreasing odds of NAFLD to approximately 1800 MET-min/week in cubic spline analysis and then a moderate decline. Although SB was significantly associated with NAFLD in a body mass index (BMI)-unadjusted model, it was not significant in a BMI-adjusted model. In the compositional isotemporal substitution, reallocating 60min/day of SB to MVPA decreased the odds of NAFLD by 22% (OR=0.78, 95% CI: 0.65-0.93), whereas reallocating 60min/day of MVPA to SB increased it by 69% (OR=1.69, 95% CI: 1.12-2.38). The inverse dose-response association between MVPA and NAFLD confirms increased MVPA as a target for the prevention of NAFLD.