<jats:title>Abstract</jats:title><jats:sec><jats:title>Background</jats:title><jats:p>Moderate-to-vigorous physical activity (MVPA), light physical activity, sedentary behaviour and sleep have all been associated with cardiovascular disease (CVD). Due to challenges in measuring and analysing movement behaviours, there is uncertainty about how the association with incident CVD varies with the time spent in these different movement behaviours.</jats:p></jats:sec><jats:sec><jats:title>Methods</jats:title><jats:p>We developed a machine-learning model (Random Forest smoothed by a Hidden Markov model) to classify sleep, sedentary behaviour, light physical activity and MVPA from accelerometer data. The model was developed using data from a free-living study of 152 participants who wore an Axivity AX3 accelerometer on the wrist while also wearing a camera and completing a time use diary. Participants in UK Biobank, a prospective cohort study, were asked to wear an accelerometer (of the same type) for seven days, and we applied our machine-learning model to classify their movement behaviours. Using Compositional Data Analysis Cox regression, we investigated how reallocating time between movement behaviours was associated with CVD incidence.</jats:p></jats:sec><jats:sec><jats:title>Findings</jats:title><jats:p>We classified accelerometer data as sleep, sedentary behaviour, light physical activity or MVPA with a mean accuracy of 88% (95% CI: 87, 89) and Cohen’s kappa of 0·80 (95% CI: 0·79, 0·82). Among 87,509 UK Biobank participants, there were 3,424 incident CVD events. Reallocating time from any behaviour to MVPA, or reallocating time from sedentary behaviour to any behaviour, was associated with a lower risk of CVD. For example, for a hypothetical average individual, reallocating 20 minutes/day to MVPA from all other behaviours proportionally was associated with 9% (7%, 10%) lower risk of incident CVD, while reallocating 1 hour/day to sedentary behaviour was associated with 5% (3%, 7%) higher risk.</jats:p></jats:sec><jats:sec><jats:title>Interpretation</jats:title><jats:p>Reallocating time from light physical activity, sedentary behaviour or sleep to MVPA, or reallocating time from sedentary behaviour to other behaviours, was associated with lower risk of incident CVD. Accurate classification of movement behaviours using machine-learning and statistical methods to address the compositional nature of movement behaviours enabled these insights. Public health interventions and guidelines should promote reallocating time to MVPA from other behaviours, as well as reallocating time from sedentary behaviour to light physical activity.</jats:p></jats:sec><jats:sec><jats:title>Funding</jats:title><jats:p>Medical Research Council.</jats:p></jats:sec><jats:sec><jats:title>Research in Context</jats:title><jats:sec><jats:title>Evidence before this study</jats:title><jats:p>Low levels of moderate-to-vigorous physical activity, low levels of light physical activity, high levels of sedentary behaviour and short and long sleep time have been associated with cardiovascular disease risk. Uncertainty remains about how different combinations of behaviours are associated with risk, in part due to challenges in the measurement (e.g. reliance on self-reported measures or crude device-based measures) and analysis of movement behaviours (e.g. taking appropriate account of behaviours making up the 24 hour day). We searched PubMed for studies published up to the end of September 2020 that investigated the association between device-measured moderate-to-vigorous physical activity, light physical activity, sedentary behaviour or sleep and incident cardiovascular disease in adult populations (search terms in <jats:bold>Supplementary Material</jats:bold>). Four studies were identified, which found that lower levels of sedentary behaviour, higher levels of light physical activity and higher levels of moderate-to-vigorous physical activity were associated with lower cardiovascular disease risk. These studies did not consider sleep and used traditional ‘cut-point’ based methods to identify movement behaviours in device data. Most categorised the exposure and adjusted only partially for time in behaviours other than the behaviour of interest.</jats:p></jats:sec><jats:sec><jats:title>Added value of this study</jats:title><jats:p>In this prospective study of 87,509 participants, we used machine-learning methods, developed and validated using a separate free-living dataset, to accurately classify device data as sleep, sedentary behaviour, light physical activity or moderate-to-vigorous physical activity. We used a Compositional Data Analysis approach to investigate how reallocating time between device-measured movement behaviours was associated with cardiovascular disease incidence. We found that, for a hypothetical average individual, reallocating 20 minutes/day to moderate-to-vigorous physical activity from all other behaviours proportionally was associated with 9% (95% CI: 7%, 10%) lower risk of incident cardiovascular disease, whereas reallocating 1 hour/day to sedentary behaviour was associated with 5% (3%, 7%) higher risk.</jats:p></jats:sec><jats:sec><jats:title>Implications of all the available evidence</jats:title><jats:p>Emerging methods, such as machine-learning based behaviour classification and Compositional Data Analysis for epidemiological analysis, can provide new health insights. Using these methods we found that reallocating time to moderate-to-vigorous physical activity from other behaviours was associated with lower cardiovascular disease risk and thus should be encouraged. Reallocating time from sedentary behaviour to other behaviours was also associated with lower cardiovascular risk. To strengthen the evidence for causality, intervention studies should examine the health consequences of reallocating time between movement behaviours.</jats:p></jats:sec></jats:sec>
Journal article
Cold Spring Harbor Laboratory
13/11/2020