Reallocating time from device-measured sleep, sedentary behaviour or light physical activity to moderate-to-vigorous physical activity is associated with lower cardiovascular disease risk

Abstract

BackgroundModerate-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.MethodsWe 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.FindingsWe 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.InterpretationReallocating 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.FundingMedical Research Council.

Evaluating the Use of Digital Biomarkers to Test Treatment Effects on Cognition and Movement in Patients with Lewy Body Dementia

“…This reduction indicates that the actigraphy-measured reduction in daytime sleep can not only predict treatment-induced increases in wakefulness but can also detect changes in wakefulness earlier and with a smaller sample size than its clinical assessment counterpart. Although the actigraphy-measured sleep appeared to be more sensitive than clinical measurements, it is important to note that sleep was measured using an actigraphybased algorithm [22,32,33]; in future studies, the use of devices that measure other physiological measures related to sleep, such as heart rate variability, may be helpful to further differentiate and improve the accuracy of sleep detection.…”

pyActigraphy: Open-source python package for actigraphy data visualization and analysis

“…Over the last years, there have been efforts to create open-source analysis tools; from packages to simply calculate the IS, IV and RA variables (nparACT [49]) to softwares allowing the preprocessing of large accelerometer data sets and the extraction of the sleep timing and duration (e.g. bioban-kAccelerometerAnalysis [6,[50][51][52], GGIR [53], OMGUI [54]). However, to our knowledge, there is no comprehensive open-source analysis package for actigraphy data that would allow users to read various data format, perform the necessary data cleaning as well as more advanced data analysis within a single framework in the python ecosystem.…”

Digital health technologies and machine learning augment patient reported outcomes to remotely characterise rheumatoid arthritis

“…In this study we trained an activity recognition model on the free-living Capture-24 dataset to estimate daily activity patterns in the weaRAble-PRO population. Leveraging the latest advances in self-supervised learning (SSL) allowed our model to be pretrained on 100,000 participants with 700,000 days of diverse, unlabelled wearable sensor data in the UK Biobank 26 , which combined with HMM temporal smoothing, significantly improved activity prediction compared to our previous established RF-HMM based methods 27,29 . Our SSL DCNN+HMM model enabled a more robust and fine-grained estimation of daily activity patterns beyond traditional acceleration magnitude levels 13,14 , which we proposed could allow a richer characterisation of PA and sleep in RA.…”

“…Activity labels were then summarised into a small number of free-living behaviour labels, defining activity classes in Capture-24. There are two major labelling conventions used within Capture-24 that the model was trained to predict, defined as broad activity: {sleep, sedentary, light physical activity, moderate-to-vigorous physical activity (MVPA)} 28,29 ; and fine-grained activity: {sleep, sitting/standing, mixed, vehicle, walking, bicycling} 27 . HAR model predictions are essentially independentmeaning that the sequence of activities over each 30 second epoch incorporates no temporal information epoch-to-epoch, for instance how the previous epoch prediction affects the current, or next, activity prediction.…”