Prospective associations between changes in physical activity and sedentary time and subsequent lean muscle mass in older English adults: the EPIC-Norfolk cohort study.

Acknowledgements: We are grateful to our study participants for giving their time to the EPIC-Norfolk study. We thank the principal investigators of the EPIC-Norfolk study, who are Nick Day, Sheila Bingham, Kay-Tee Khaw, and Nick Wareham. We thank the EPIC-Norfolk field epidemiology, IT, and data management teams for running the study. We would also like to thank Tom White, Kate Westgate and the Physical Activity Technical Team (MRC Epidemiology Unit) for their assistance with processing the accelerometry data used in the present analyses. ; Funder: NIHR Senior Investigator ; BACKGROUND: The longitudinal associations between physical behaviours and lean muscle mass indices need to be better understood to aid healthy ageing intervention development. METHODS: We assessed physical behaviours (total physical activity, moderate-to-vigorous physical activity (MVPA), light physical activity, total sedentary time and prolonged sedentary bout time) for 7 days using hip-worn accelerometers. We also assessed domain-specific physical behaviours (walking, cycling, gardening and housework time) with self-report questionnaires at baseline (2006-2011) and follow-up (2012-2016) in the European Prospective Investigation into Cancer (EPIC)-Norfolk study. We assessed body composition using dual-energy X-ray absorptiometry (DEXA) at follow-up in 1535 participants (≥ 60 years at baseline). From this, we derived appendicular lean muscle mass (ALM) indices (% relative ALM = (ALM/total body weight)*100), body mass index (BMI)-scaled ALM (ALM/BMI, kg/kg/m2) and height-scaled ALM (ALM/height2, kg/m2)). We evaluated the prospective associations of both baseline and change in physical behaviours with follow-up muscle mass indices using multivariable linear regression. RESULTS: Over 5.5 years (SD 14.8) follow-up, higher baseline accelerometer-measured physical activity and lower sedentary time were associated with higher subsequent relative ALM and BMI-scaled ALM, but not height-scaled ALM (e.g. 0.02% higher subsequent relative ALM per ...