Accelerometer-Assessed Physical Activity in People with Type 2 Diabetes: Accounting for Sleep when Determining Associations with Markers of Health.

High physical activity levels during wake are beneficial for health, while high movement levels during sleep are detrimental to health. Our aim was to compare the associations of accelerometer-assessed physical activity and sleep disruption with adiposity and fitness using standardized and individualized wake and sleep windows. People (N = 609) with type 2 diabetes wore an accelerometer for up to 8 days. Waist circumference, body fat percentage, Short Physical Performance Battery (SPPB) test score, sit-to-stands, and resting heart rate were assessed. Physical activity was assessed via the average acceleration and intensity distribution (intensity gradient) over standardized (most active 16 continuous hours (M16h)) and individualized wake windows. Sleep disruption was assessed via the average acceleration over standardized (least active 8 continuous hours (L8h)) and individualized sleep windows. Average acceleration and intensity distribution during the wake window were beneficially associated with adiposity and fitness, while average acceleration during the sleep window was detrimentally associated with adiposity and fitness. Point estimates for the associations were slightly stronger for the standardized than for individualized wake/sleep windows. In conclusion, standardized wake and sleep windows may have stronger associations with health due to capturing variations in sleep durations across individuals, while individualized windows represent a purer measure of wake/sleep behaviors.

Self-supervised learning of accelerometer data provides new insights for sleep and its association with mortality.

Background: Sleep is essential to life. Accurate measurement and classification of sleep/wake and sleep stages is important in clinical studies for sleep disorder diagnoses and in the interpretation of data from consumer devices for monitoring physical and mental well-being. Existing non-polysomnography sleep classification techniques mainly rely on heuristic methods developed in relatively small cohorts. Thus, we aimed to establish the accuracy of wrist-worn accelerometers for sleep stage classification and subsequently describe the association between sleep duration and efficiency (proportion of total time asleep when in bed) with mortality outcomes. Methods: We developed and validated a self-supervised deep neural network for sleep stage classification using concurrent laboratory-based polysomnography and accelerometry data from three countries (Australia, the UK, and the USA). The model was validated within-cohort using subject-wise five-fold cross-validation for sleep-wake classification and in a three-class setting for sleep stage classification wake, rapid-eye-movement sleep (REM), non-rapid-eye-movement sleep (NREM) and by external validation. We assessed the face validity of our model for population inference by applying the model to the UK Biobank with 100,000 participants, each of whom wore a wristband for up to seven days. The derived sleep parameters were used in a Cox regression model to study the association of sleep duration and sleep efficiency with all-cause mortality. Findings: After exclusion, 1,448 participant nights of data were used to train the sleep classifier. The difference between polysomnography and the model classifications on the external validation was 34.7 minutes (95% limits of agreement (LoA): -37.8 to 107.2 minutes) for total sleep duration, 2.6 minutes for REM duration (95% LoA: -68.4 to 73.4 minutes) and 32.1 minutes (95% LoA: -54.4 to 118.5 minutes) for NREM duration. The derived sleep architecture estimate in the UK Biobank sample showed good face validity. Among 66,214 UK Biobank participants, 1,642 mortality events were observed. Short sleepers (<6 hours) had a higher risk of mortality compared to participants with normal sleep duration (6 to 7.9 hours), regardless of whether they had low sleep efficiency (Hazard ratios (HRs): 1.69; 95% confidence intervals (CIs): 1.28 to 2.24 ) or high sleep efficiency (HRs: 1.42; 95% CIs: 1.14 to 1.77). Interpretation: Deep-learning-based sleep classification using accelerometers has a fair to moderate agreement with polysomnography. Our findings suggest that having short overnight sleep confers mortality risk irrespective of sleep continuity.

Association between Chronotype and Physical Behaviours in Adolescent Girls.

The aim of this study was to (1) describe accelerometer-assessed physical behaviours by chronotype, and (2) examine the association between chronotype and accelerometer-assessed physical behaviours in a cohort of adolescent girls. Chronotype (single question) and physical behaviours (GENEActiv accelerometer on the non-dominant wrist) were assessed in 965 adolescent girls (13.9 ± 0.8 years). Linear mixed-effects models examined the relationships among chronotype and physical behaviours (time in bed, total sleep time, sleep efficiency, sedentary time, overall, light and moderate-to-vigorous physical activity) on weekdays and weekend days. Over the 24 h day, participants spent 46% sedentary, 20% in light activity, 3% in moderate-to-vigorous physical activity, and 31% in 'time in bed'. Seventy percent of participants identified as 'evening' chronotypes. Compared to evening chronotypes, morning chronotypes engaged in less sedentary time (10 min/day) and had higher overall physical activity (1.3 mg/day, ~30 min of slow walking) on weekdays. Most girls identified as evening chronotypes with a large proportion of their day spent sedentary and a small amount in physical activities which may be exacerbated in evening chronotypes on weekdays. The results maybe be important for programmes aiming to promote physical activity in adolescent girls.

Effects of sleep disturbance on dyspnoea and impaired lung function following hospital admission due to COVID-19 in the UK: a prospective multicentre cohort study.

BACKGROUND: Sleep disturbance is common following hospital admission both for COVID-19 and other causes. The clinical associations of this for recovery after hospital admission are poorly understood despite sleep disturbance contributing to morbidity in other scenarios. We aimed to investigate the prevalence and nature of sleep disturbance after discharge following hospital admission for COVID-19 and to assess whether this was associated with dyspnoea. METHODS: CircCOVID was a prospective multicentre cohort substudy designed to investigate the effects of circadian disruption and sleep disturbance on recovery after COVID-19 in a cohort of participants aged 18 years or older, admitted to hospital for COVID-19 in the UK, and discharged between March, 2020, and October, 2021. Participants were recruited from the Post-hospitalisation COVID-19 study (PHOSP-COVID). Follow-up data were collected at two timepoints: an early time point 2-7 months after hospital discharge and a later time point 10-14 months after hospital discharge. Sleep quality was assessed subjectively using the Pittsburgh Sleep Quality Index questionnaire and a numerical rating scale. Sleep quality was also assessed with an accelerometer worn on the wrist (actigraphy) for 14 days. Participants were also clinically phenotyped, including assessment of symptoms (ie, anxiety [Generalised Anxiety Disorder 7-item scale questionnaire], muscle function [SARC-F questionnaire], dyspnoea [Dyspnoea-12 questionnaire] and measurement of lung function), at the early timepoint after discharge. Actigraphy results were also compared to a matched UK Biobank cohort (non-hospitalised individuals and recently hospitalised individuals). Multivariable linear regression was used to define associations of sleep disturbance with the primary outcome of breathlessness and the other clinical symptoms. PHOSP-COVID is registered on the ISRCTN Registry (ISRCTN10980107). FINDINGS: 2320 of 2468 participants in the PHOSP-COVID study attended an early timepoint research visit a median of 5 months (IQR 4-6) following discharge from 83 hospitals in the UK. Data for sleep quality were assessed by subjective measures (the Pittsburgh Sleep Quality Index questionnaire and the numerical rating scale) for 638 participants at the early time point. Sleep quality was also assessed using device-based measures (actigraphy) a median of 7 months (IQR 5-8 months) after discharge from hospital for 729 participants. After discharge from hospital, the majority (396 [62%] of 638) of participants who had been admitted to hospital for COVID-19 reported poor sleep quality in response to the Pittsburgh Sleep Quality Index questionnaire. A comparable proportion (338 [53%] of 638) of participants felt their sleep quality had deteriorated following discharge after COVID-19 admission, as assessed by the numerical rating scale. Device-based measurements were compared to an age-matched, sex-matched, BMI-matched, and time from discharge-matched UK Biobank cohort who had recently been admitted to hospital. Compared to the recently hospitalised matched UK Biobank cohort, participants in our study slept on average 65 min (95% CI 59 to 71) longer, had a lower sleep regularity index (-19%; 95% CI -20 to -16), and a lower sleep efficiency (3·83 percentage points; 95% CI 3·40 to 4·26). Similar results were obtained when comparisons were made with the non-hospitalised UK Biobank cohort. Overall sleep quality (unadjusted effect estimate 3·94; 95% CI 2·78 to 5·10), deterioration in sleep quality following hospital admission (3·00; 1·82 to 4·28), and sleep regularity (4·38; 2·10 to 6·65) were associated with higher dyspnoea scores. Poor sleep quality, deterioration in sleep quality, and sleep regularity were also associated with impaired lung function, as assessed by forced vital capacity. Depending on the sleep metric, anxiety mediated 18-39% of the effect of sleep disturbance on dyspnoea, while muscle weakness mediated 27-41% of this effect. INTERPRETATION: Sleep disturbance following hospital admission for COVID-19 is associated with dyspnoea, anxiety, and muscle weakness. Due to the association with multiple symptoms, targeting sleep disturbance might be beneficial in treating the post-COVID-19 condition. FUNDING: UK Research and Innovation, National Institute for Health Research, and Engineering and Physical Sciences Research Council.

The effect of COVID rehabilitation for ongoing symptoms Post HOSPitalisation with COVID-19 (PHOSP-R): protocol for a randomised parallel group controlled trial on behalf of the PHOSP consortium.

INTRODUCTION: Many adults hospitalised with COVID-19 have persistent symptoms such as fatigue, breathlessness and brain fog that limit day-to-day activities. These symptoms can last over 2 years. Whilst there is limited controlled studies on interventions that can support those with ongoing symptoms, there has been some promise in rehabilitation interventions in improving function and symptoms either using face-to-face or digital methods, but evidence remains limited and these studies often lack a control group. METHODS AND ANALYSIS: This is a nested single-blind, parallel group, randomised control trial with embedded qualitative evaluation comparing rehabilitation (face-to-face or digital) to usual care and conducted within the PHOSP-COVID study. The aim of this study is to determine the effectiveness of rehabilitation interventions on exercise capacity, quality of life and symptoms such as breathlessness and fatigue. The primary outcome is the Incremental Shuttle Walking Test following the eight week intervention phase. Secondary outcomes include measures of function, strength and subjective assessment of symptoms. Blood inflammatory markers and muscle biopsies are an exploratory outcome. The interventions last eight weeks and combine symptom-titrated exercise therapy, symptom management and education delivered either in a face-to-face setting or through a digital platform ( www.yourcovidrecovery.nhs.uk ). The proposed sample size is 159 participants, and data will be intention-to-treat analyses comparing rehabilitation (face-to-face or digital) to usual care. ETHICS AND DISSEMINATION: Ethical approval was gained as part of the PHOSP-COVID study by Yorkshire and the Humber Leeds West Research NHS Ethics Committee, and the study was prospectively registered on the ISRCTN trial registry (ISRCTN13293865). Results will be disseminated to stakeholders, including patients and members of the public, and published in appropriate journals. Strengths and limitations of this study • This protocol utilises two interventions to support those with ongoing symptoms of COVID-19 • This is a two-centre parallel-group randomised controlled trial • The protocol has been supported by patient and public involvement groups who identified treatments of symptoms and activity limitation as a top priority.

Validation of an automated sleep detection algorithm using data from multiple accelerometer brands.

To evaluate the criterion validity of an automated sleep detection algorithm applied to data from three research-grade accelerometers worn on each wrist with concurrent laboratory-based polysomnography (PSG). A total of 30 healthy volunteers (mean [SD] age 31.5 [7.2] years, body mass index 25.5 [3.7] kg/m2 ) wore an Axivity, GENEActiv and ActiGraph accelerometer on each wrist during a 1-night PSG assessment. Sleep estimates (sleep period time window [SPT-window], sleep duration, sleep onset and waking time, sleep efficiency, and wake after sleep onset [WASO]) were generated using the automated sleep detection algorithm within the open-source GGIR package. Agreement of sleep estimates from accelerometer data with PSG was determined using pairwise 95% equivalence tests (±10% equivalence zone), intraclass correlation coefficients (ICCs) with 95% confidence intervals and limits of agreement (LoA). Accelerometer-derived sleep estimates except for WASO were within the 10% equivalence zone of the PSG. Reliability between data from the accelerometers worn on either wrist and PSG was moderate for SPT-window duration (ICCs ≥ 0.65), sleep duration (ICCs ≥ 0.54), and sleep onset (ICCs ≥ 0.61), mostly good for waking time (ICCs ≥ 0.80), but poor for sleep efficiency (ICCs ≥ 0.08) and WASO (ICCs ≥ 0.08). The mean bias between all accelerometer-derived sleep estimates worn on either wrist and PSG were low; however, wide 95% LoA were observed for all sleep estimates, apart from waking time. The automated sleep detection algorithm applied to data from Axivity, GENEActiv and ActiGraph accelerometers, worn on either wrist, provides comparable measures to PSG for SPT-window and sleep duration, sleep onset and waking time, but a poor measure of wake during the sleep period.

Stepping towards More Intuitive Physical Activity Metrics with Wrist-Worn Accelerometry: Validity of an Open-Source Step-Count Algorithm.

Stepping-based targets such as the number of steps per day provide an intuitive and commonly used method of prescribing and self-monitoring physical activity goals. Physical activity surveillance is increasingly being obtained from wrist-worn accelerometers. However, the ability to derive stepping-based metrics from this wear location still lacks validation and open-source methods. This study aimed to assess the concurrent validity of two versions (1. original and 2. optimized) of the Verisense step-count algorithm at estimating step-counts from wrist-worn accelerometry, compared with steps from the thigh-worn activPAL as the comparator. Participants (n = 713), across three datasets, had >24 h continuous concurrent accelerometry wear on the non-dominant wrist and thigh. Compared with activPAL, total daily steps were overestimated by 913 ± 141 (mean bias ± 95% limits of agreement) and 742 ± 150 steps/day with Verisense algorithms 1 and 2, respectively, but moderate-to-vigorous physical activity (MVPA) steps were underestimated by 2207 ± 145 and 1204 ± 103 steps/day in Verisense algorithms 1 and 2, respectively. In summary, the optimized Verisense algorithm was more accurate in detecting total and MVPA steps. Findings highlight the importance of assessing algorithm performance beyond total step count, as not all steps are equal. The optimized Verisense open-source algorithm presents acceptable accuracy for derivation of stepping-based metrics from wrist-worn accelerometry.

Device-assessed sleep and physical activity in individuals recovering from a hospital admission for COVID-19: a multicentre study.

BACKGROUND: The number of individuals recovering from severe COVID-19 is increasing rapidly. However, little is known about physical behaviours that make up the 24-h cycle within these individuals. This study aimed to describe physical behaviours following hospital admission for COVID-19 at eight months post-discharge including associations with acute illness severity and ongoing symptoms. METHODS: One thousand seventy-seven patients with COVID-19 discharged from hospital between March and November 2020 were recruited. Using a 14-day wear protocol, wrist-worn accelerometers were sent to participants after a five-month follow-up assessment. Acute illness severity was assessed by the WHO clinical progression scale, and the severity of ongoing symptoms was assessed using four previously reported data-driven clinical recovery clusters. Two existing control populations of office workers and individuals with type 2 diabetes were comparators. RESULTS: Valid accelerometer data from 253 women and 462 men were included. Women engaged in a mean ± SD of 14.9 ± 14.7 min/day of moderate-to-vigorous physical activity (MVPA), with 12.1 ± 1.7 h/day spent inactive and 7.2 ± 1.1 h/day asleep. The values for men were 21.0 ± 22.3 and 12.6 ± 1.7 h /day and 6.9 ± 1.1 h/day, respectively. Over 60% of women and men did not have any days containing a 30-min bout of MVPA. Variability in sleep timing was approximately 2 h in men and women. More severe acute illness was associated with lower total activity and MVPA in recovery. The very severe recovery cluster was associated with fewer days/week containing continuous bouts of MVPA, longer total sleep time, and higher variability in sleep timing. Patients post-hospitalisation with COVID-19 had lower levels of physical activity, greater sleep variability, and lower sleep efficiency than a similarly aged cohort of office workers or those with type 2 diabetes. CONCLUSIONS: Those recovering from a hospital admission for COVID-19 have low levels of physical activity and disrupted patterns of sleep several months after discharge. Our comparative cohorts indicate that the long-term impact of COVID-19 on physical behaviours is significant.

Effect of exercise on sleep and bi-directional associations with accelerometer-assessed physical activity in men with obesity.

This study examined the effect of exercise training on sleep duration and quality and bidirectional day-to-day relationships between physical activity (PA) and sleep. Fourteen inactive men with obesity (age: 49.2 ± 7.9 years, body mass index: 34.9 ± 2.8 kg/m2) completed a baseline visit, 8-week aerobic exercise intervention, and 1-month post-intervention follow-up. PA and sleep were assessed continuously throughout the study duration using wrist-worn accelerometry. Generalised estimating equations were used to examine associations between PA and sleep. Sleep duration increased from 5.2 h at baseline to 6.6 h during the intervention period and 6.5 h at 1-month post-intervention follow-up (p < 0.001). Bi-directional associations showed that higher overall activity volume and moderate-to-vigorous physical activity (MVPA) were associated with earlier sleep onset time (p < 0.05). Later timing of sleep onset was associated with lower overall volume of activity, most active continuous 30 min (M30CONT), and MVPA (p < 0.05). Higher overall activity volume, M30CONT, and MVPA predicted more wake after sleep onset (WASO) (p < 0.001), whereas greater WASO was associated with higher overall volume of activity, M30CONT, and MVPA (p < 0.001). An aerobic exercise intervention increased usual sleep duration. Day-to-day, more PA predicted earlier sleep onset, but worse sleep quality and vice versa. Novelty: Greater levels of physical activity in the day were associated with an earlier sleep onset time that night, whereas a later timing of sleep onset was associated with lower physical activity the next day in men with obesity. Higher physical activity levels were associated with worse sleep quality, and vice versa.

Effectiveness of the 'Girls Active' school-based physical activity programme: A cluster randomised controlled trial.

BACKGROUND: Globally, adolescent girls' physical activity (PA) levels are low. The 'Girls Active' secondary school-based programme, developed by the Youth Sport Trust, aims to increase PA in adolescent girls. This paper explores the effectiveness of the 'Girls Active' school-based PA programme. METHODS: A random sample of girls aged 11-14 from 20 secondary schools (Midlands, UK) participated in a two-arm cluster randomised controlled trial. Ten schools received Girls Active and 10 continued with usual practice. Measurements were taken at baseline, seven- and 14-month follow-up. PRIMARY OUTCOME: wrist-worn accelerometer measured moderate- to vigorous-intensity PA (MVPA). SECONDARY OUTCOMES: overall PA, light PA, sedentary time, body composition, and psychosocial outcomes. Generalised estimating equations, adjusted for school cluster and potential confounders, were used and A priori subgroup analysis was undertaken. Micro-costing and cost-consequence analyses were conducted using bespoke collection methods on programme delivery information. Outcomes for the cost-consequence analysis were health related quality of life measured by the Child Health Utility-9D and service use. RESULTS: Overall, 1752 pupils participated, 1211 (69.1%) provided valid 14-month accelerometer data. No difference in MVPA (mins/day; 95% confidence intervals) was found at 14 months (1.7; -0.8 to 4.3), there was at seven months (2.4; 0.1 to 4.7). Subgroup analyses showed significant intervention effects on 14-month in larger schools (3.9; 1.39 to 6.09) and in White Europeans (3.1; 0.60 to 6.02) and in early maturers (5.1; 1.69 to 8.48) at seven months. The control group did better in smaller schools at 14-months (-4.38; -7.34 to -1.41). Significant group differences were found in 14-month identified motivation (-0.09; -0.18 to -0.01) and at seven months in: overall PA (1.39 mg/day; 0.1 to 2.2), after-school sedentary time (-4.7; -8.9 to -0.6), whole day (5.7; 1.0 to 10.5) and school day (4.5; 0.25 to 8.75) light PA, self-esteem. Small, statistically significant, differences in some psychosocial variables favoured control schools. Micro-costing demonstrated that delivering the programme resulted in a range of time and financial costs at each school. Cost-consequence analysis demonstrated no effect of the programme for health related quality of life or service use. CONCLUSIONS: Compared with usual practice, 'Girls Active' did not affect 14-month MVPA. TRIAL REGISTRATION: ISRCTN10688342.

Effect of breakfast omission and consumption on energy intake and physical activity in adolescent girls: a randomised controlled trial.

It is not known if breakfast consumption is an effective intervention for altering daily energy balance in adolescents when compared with breakfast omission. This study examined the acute effect of breakfast consumption and omission on free-living energy intake (EI) and physical activity (PA) in adolescent girls. Using an acute randomised cross-over design, forty girls (age 13·3 (sd 0·8) years, BMI 21·5 (sd 5·0) kg/m2) completed two, 3-d conditions in a randomised, counter-balanced order: no breakfast (NB) and standardised (approximately 1962 kJ) breakfast (SB). Dietary intakes were assessed using food diaries combined with digital photographic records and PA was measured via accelerometry throughout each condition. Statistical analyses were completed using repeated-measures ANOVA. Post-breakfast EI was 483 (sd 1309) kJ/d higher in NB v. SB (P=0·025), but total daily EI was 1479 (sd 11311) kJ/d higher in SB v. NB (P<0·0005). Daily carbohydrate, fibre and protein intakes were higher in SB v. NB (P<0·0005), whereas daily fat intake was not different (P=0·405). Effect sizes met the minimum important difference of ≥0·20 for all significant effects. Breakfast manipulation did not affect post-breakfast macronutrient intakes (P≥0·451) or time spent sedentary or in PA (P≥0·657). In this sample of adolescent girls, breakfast omission increased post-breakfast free-living EI, but total daily EI was greater when a SB was consumed. We found no evidence that breakfast consumption induces compensatory changes in PA. Further experimental research is required to determine the effects of extended periods of breakfast manipulation in young people.