Background Multimorbidity – the co-existence of two or more long-term conditions in an individual – is a growing problem in Malawi, as the population transitions from predominantly young and rural to older and more urban. People living with multimorbidity experience lower quality of life and shorter life expectancies. Managing multiple long-term conditions also poses a significant challenge for Malawi’s health system. Both physical activity and sleep have been shown, in other settings, to influence the development of multimorbidity, but no large-scale objective measurement of these behavioural determinants has yet been carried out in Malawi. Objectives To describe 24-hour free-living movement behaviour, comprising physical activity, inactivity, and sleep in an urban and a rural population in Malawi, and quantify the associations of these behaviours with multimorbidity prevalence. Methods This will be a cross-sectional study, nested within a larger cross-sectional study of long-term conditions in an urban and a rural Health and Demographic Surveillance System in Malawi. A random sample of 4,000 pre-selected participants will be invited to wear an Axivity AX3 accelerometer on their dominant wrist for one week. Raw accelerometery data will be processed to produce 24-hour movement behaviour metrics of physical activity, inactivity, and sleep. Inverse probability weighting will be applied to account for selection procedures. Multimorbidity will be assessed through a combination of self-reported physical and mental long-term conditions and objective measurements, including blood pressure and fasting blood glucose. In addition, this study will add hand-held spirometry to assess for obstructive airways disease. Associations between accelerometer-derived metrics and multimorbidity will be established through multivariable regression, compositional data analysis, and latent class analysis. Discussion This study will provide the largest description of device-measured 24-hour movement behaviours in Malawi to date, and the first description of the associations of these behaviours with multimorbidity in this population.
Warner et al. (Tue,) studied this question.