Wrist-worn Accelerometer Data Research Articles

Association of Recess Provision With Accelerometer-Measured Physical Activity and Sedentary Time in a Representative Sample of 6- to 11-Year-Old Children in the United States.

To assess the association between the amount of recess provision and children's accelerometer-measured physical activity (PA) levels. Parents/guardians of 6- to 11-year-olds (n = 451) in the 2012 National Youth Fitness Survey reported recess provision, categorized as low (10-15min; 31.9%), medium (16-30min; 48.0%), or high (>30min; 20.1%). Children wore a wrist-worn accelerometer for 7days to estimate time spent sedentary, in light PA, and in moderate to vigorous PA using 2 different cut points for either activity counts or raw acceleration. Outcomes were compared between levels of recess provision while adjusting for covariates and the survey's multistage, probability sampling design. Children with high recess provision spent less time sedentary, irrespective of type of day (week vs weekend) and engaged in more light or moderate to vigorous PA on weekdays than those with low recess provision. The magnitude and statistical significance of effects differed based on the cut points used to classify PA (eg,4.7 vs 11.9 additional min·d-1 of moderate to vigorous PA). Providing children with >30minutes of daily recess, which exceeds current recommendations of ≥20minutes, is associated with more favorable PA levels and not just on school days. Identifying the optimal method for analyzing wrist-worn accelerometer data could clarify the magnitude of this effect.

Longitudinal associations of meeting the WHO physical activity guidelines and physical fitness, from preschool to childhood.

There is a well-established relationship between physical activity (PA) and physical fitness in children, being the latter an important marker for present and future health; however, there is still insufficient knowledge for the transition from the preschool age to early childhood. Therefore, this study in Swedish children aimed to investigate the estimated effect of meeting the aerobic component of the PA guidelines at 4 and/or 9 years of age on physical fitness measured at 9 years of age. PA was assessed using a wrist-worn ActiGraph accelerometer and identical data processing in 217 healthy children in Sweden (114 boys and 103 girls). Physical fitness test included cardiorespiratory (20 m shuttle run test), motor (4 × 10 m shuttle run), and muscular fitness (hand grip strength and long jump). A linear mixed model was run, investigating the interaction between meeting the PA guidelines and time (either 4 or 9 years of age) and each fitness component (at 4 and 9). Interactions by sex were also checked. Meeting the PA guidelines consistently (at 4 and 9 years) was significantly associated to better performance in physical fitness parameters for motor fitness (-0.76 s, p < 0.001) and lower body muscular fitness (+4.6 cm; p < 0.001) at 9 years. There was an interaction between meeting the PA guidelines and time point, for cardiorespiratory fitness (+4.58 laps; p < 0.001). This study shows that meeting the PA guidelines at 4 and 9 years of age is associated to higher physical fitness at 9 years of age.

Impaired 24-h activity patterns are associated with an increased risk of Alzheimer’s disease, Parkinson’s disease, and cognitive decline

BackgroundSleep-wake regulating circuits are affected during prodromal stages in the pathological progression of both Alzheimer’s disease (AD) and Parkinson’s disease (PD), and this disturbance can be measured passively using wearable devices. Our objective was to determine whether accelerometer-based measures of 24-h activity are associated with subsequent development of AD, PD, and cognitive decline.MethodsThis study obtained UK Biobank data from 82,829 individuals with wrist-worn accelerometer data aged 40 to 79 years with a mean (± SD) follow-up of 6.8 (± 0.9) years. Outcomes were accelerometer-derived measures of 24-h activity (derived by cosinor, nonparametric, and functional principal component methods), incident AD and PD diagnosis (obtained through hospitalization or primary care records), and prospective longitudinal cognitive testing.ResultsOne hundred eighty-seven individuals progressed to AD and 265 to PD. Interdaily stability (a measure of regularity, hazard ratio [HR] per SD increase 1.25, 95% confidence interval [CI] 1.05–1.48), diurnal amplitude (HR 0.79, CI 0.65–0.96), mesor (mean activity; HR 0.77, CI 0.59–0.998), and activity during most active 10 h (HR 0.75, CI 0.61–0.94), were associated with risk of AD. Diurnal amplitude (HR 0.28, CI 0.23–0.34), mesor (HR 0.13, CI 0.10–0.16), activity during least active 5 h (HR 0.24, CI 0.08–0.69), and activity during most active 10 h (HR 0.20, CI 0.16–0.25) were associated with risk of PD. Several measures were additionally predictive of longitudinal cognitive test performance.ConclusionsIn this community-based longitudinal study, accelerometer-derived metrics were associated with elevated risk of AD, PD, and accelerated cognitive decline. These findings suggest 24-h rhythm integrity, as measured by affordable, non-invasive wearable devices, may serve as a scalable early marker of neurodegenerative disease.

Device-Measured Physical Activity in 3506 Individuals with Knee or Hip Arthroplasty.

Hip and knee arthroplasty aims to reduce joint pain and increase functional mobility in patients with osteoarthritis; however, the degree to which arthroplasty is associated with higher physical activity is unclear. The current study sought to assess the association of hip and knee arthroplasty with objectively measured physical activity. This cross-sectional study analyzed wrist-worn accelerometer data collected in 2013-2016 from UK Biobank participants (aged 43-78 yr). Multivariable linear regression was performed to assess step count, cadence, overall acceleration, and activity behaviors between nonarthritic controls, end-stage arthritic, and postoperative cohorts, controlling for demographic and behavioral confounders. From a cohort of 94,707 participants with valid accelerometer wear time and complete self-reported data, electronic health records were used to identify 3506 participants having undergone primary or revision hip or knee arthroplasty and 68,389 nonarthritic controls. End-stage hip or knee arthritis was associated with taking 1129 fewer steps per day (95% confidence interval (CI), 811-1447; P < 0.001) and having 5.8 fewer minutes per day (95% CI, 3.0-8.7; P < 0.001) of moderate-to-vigorous activity compared with nonarthritic controls. Unilateral primary hip and knee arthroplasties were associated with 877 (95% CI, 284-1471; P = 0.004) and 893 (95% CI, 232-1554; P = 0.008) more steps than end-stage osteoarthritic participants, respectively. Postoperative unilateral hip arthroplasty participants demonstrated levels of moderate-to-vigorous physical activity and daily step count equivalent to nonarthritic controls. No difference in physical activity was observed between any cohorts in terms of overall acceleration, or time spent in daily light activity, sedentary behavior, or sleep. Hip and knee arthroplasties are associated with higher levels of physical activity compared with participants with end-stage arthritis. Unilateral hip arthroplasty patients, in particular, demonstrate equivalence to nonarthritic peers at more than 1 yr after surgery.

Circadian rest-activity rhythms and cognitive decline and impairment in older Chinese adults: A multicohort study with prospective follow-up

BackgroundThe associations between rest-activity rhythms and cognitive health are inconclusive. The potential changes in rest-activity rhythms in older people with mild cognitive impairment (MCI) remains unclear. This study aimed to examine the association between rest-activity rhythms and cognitive health across different outcome measures in older Chinese people. MethodsA total of 710 community-dwelling participants (average age 81.1 ± 5.2 years) from two cohort studies. Wrist-worn accelerometer data was used to estimate the circadian rest-activity rhythms at baseline. Cognitive function was assessed and clinical diagnosis was made at baseline and follow-up. The two-way Analysis of Co-variance was used to compare the differences in rest-activity rhythms across participants with cognitively normal, MCI and Alzheimer's disease (AD). Logistic regression models were used to investigate the association between rest-activity rhythms and incidence of cognitive decline and impairment in a 4-year prospective follow-up of cognitively normal individuals. ResultsThere was a progressive trend of lower relative amplitude and higher activity level during the least active 5 h across participants with cognitively normal, MCI and Alzheimer's disease (AD). Among the cognitively normal participants, lower relative amplitude at baseline was associated with a greater risk of cognitive decline (per 1 SD decrease, odds ratio 1.66 [95 %CI 1.13–2.45]) and increased incidence of MCI or AD (per 1 SD decrease, 1.68 [1.12–2.50]). ConclusionsLower relative amplitude could potentially serve as a robust biomarker of cognitive decline and impairment. Further studies could evaluate the potential benefits of interventions associated with rest-activity relative amplitude to prevent or delay the progression of AD.

Sedentary Behavior and Incident Dementia Among Older Adults.

Sedentary behavior is associated with cardiometabolic disease and mortality, but its association with dementia is unclear. To investigate whether accelerometer-assessed sedentary behavior is associated with incident dementia. A retrospective study of prospectively collected data from the UK Biobank including 49 841 adults aged 60 years or older without a diagnosis of dementia at the time of wearing the wrist accelerometer and living in England, Scotland, or Wales. Follow-up began at the time of wearing the accelerometer (February 2013 to December 2015) and continued until September 2021 in England, July 2021 in Scotland, and February 2018 in Wales. Mean daily sedentary behavior time (included in the primary analysis) and mean daily sedentary bout length, maximum daily sedentary bout length, and mean number of daily sedentary bouts (included in the secondary analyses) were derived from a machine learning-based analysis of 1 week of wrist-worn accelerometer data. Incident all-cause dementia diagnosis from inpatient hospital records and death registry data. Cox proportional hazard models with linear and cubic spline terms were used to assess associations. A total of 49 841 older adults (mean age, 67.19 [SD, 4.29] years; 54.7% were female) were followed up for a mean of 6.72 years (SD, 0.95 years). During this time, 414 individuals were diagnosed with incident all-cause dementia. In the fully adjusted models, there was a significant nonlinear association between time spent in sedentary behavior and incident dementia. Relative to a median of 9.27 hours/d for sedentary behavior, the hazard ratios (HRs) for dementia were 1.08 (95% CI, 1.04-1.12, P < .001) for 10 hours/d, 1.63 (95% CI, 1.35-1.97, P < .001) for 12 hours/d, and 3.21 (95% CI, 2.05-5.04, P < .001) for 15 hours/d. The adjusted incidence rate of dementia per 1000 person-years was 7.49 (95% CI, 7.48-7.49) for 9.27 hours/d of sedentary behavior, 8.06 (95% CI, 7.76-8.36) for 10 hours/d, 12.00 (95% CI, 10.00-14.36) for 12 hours/d, and 22.74 (95% CI, 14.92-34.11) for 15 hours/d. Mean daily sedentary bout length (HR, 1.53 [95% CI, 1.03-2.27], P = .04 and 0.65 [95% CI, 0.04-1.57] more dementia cases per 1000 person-years for a 1-hour increase from the mean of 0.48 hours) and maximum daily sedentary bout length (HR, 1.15 [95% CI, 1.02-1.31], P = .02 and 0.19 [95% CI, 0.02-0.38] more dementia cases per 1000 person-years for a 1-hour increase from the mean of 1.95 hours) were significantly associated with higher risk of incident dementia. The number of sedentary bouts per day was not associated with higher risk of incident dementia (HR, 1.00 [95% CI, 0.99-1.01], P = .89). In the sensitivity analyses, after adjustment for time spent in sedentary behavior, the mean daily sedentary bout length and the maximum daily sedentary bout length were no longer significantly associated with incident dementia. Among older adults, more time spent in sedentary behaviors was significantly associated with higher incidence of all-cause dementia. Future research is needed to determine whether the association between sedentary behavior and risk of dementia is causal.

Comparison Between Hip And Wrist Worn Accelerometer Data In Pre-kindergarten Children During Indoor Free-play

Comparison Between Hip And Wrist Worn Accelerometer Data In Pre-kindergarten Children During Indoor Free-play

Replacing device-measured sedentary time with physical activity is associated with lower risk of coronary heart disease regardless of genetic risk.

Excess sedentary time (ST) is recognized as an important modifiable risk factor for coronary heart disease (CHD). However, whether the associations of genetic susceptibility with CHD incidence can be modified by replacing wearable-device-measured ST with physical activity (PA) is unknown. To examine the associations of wearable-device-measured ST replaced by PA with incident CHD across strata of genetic susceptibility. This study included 77,500 White British (57% female) with valid wrist-worn accelerometry and without prevalent CHD/stroke from UK Biobank. Genetic susceptibility to CHD was quantified through weighted polygenic risk scores for CHD based on 300 single-nucleotide polymorphisms. Wrist-worn accelerometer data were used to derive ST, light PA, and moderate-to-vigorous PA (MVPA). Reallocation of 60min/day of ST into the same amount of MVPA was associated with approximately 9% lower relative risk of CHD for all participants and across strata of genetic risk: replacement of 1min/day of ST associated with <1% lower relative risk of CHD. No evidence of interaction (p: 0.784) was found between genetic risk and ST for CHD risk. Reallocating 60min/day of ST into the same MVPA time was associated with greater absolute CHD risk reductions at high genetic risk (0.27%) versus low genetic risk (0.15%). Replacing any amount of ST with an equal amount of MVPA time is associated with a lower relative risk of CHD, irrespective of genetic susceptibility to CHD. Reductions in CHD absolute risk for replacing ST with MVPA are greater at high genetic risk versus low genetic risk.

Abstract P1045: Replacement Of Wearable-device-measured Sedentary Time By Physical Activity, Genetic Susceptibility, And Risk Of Coronary Heart Disease

Backgrounds: Excess sedentary time (ST) is recognized as an important modifiable risk factor for coronary heart disease (CHD). However, whether the associations of genetic susceptibility with CHD incidence can be modified by replacing wearable-device-measured ST with physical activity (PA) is unknown. This study aims to examine the combined associations of genetic susceptibility to CHD, and wearable-device-measured ST replaced by moderate-to-vigorous PA (MVPA) or light PA in relation to incident CHD. Methods and Findings: This study included 75,545 white British (57% female) with valid wrist-worn accelerometer data and without prevalent CHD/stroke from UK Biobank. Genetic susceptibility to CHD was quantified through weighted polygenic risk scores for CHD based on 300 single-nucleotide polymorphisms. Wrist-worn accelerometer data were used to derive ST, light PA and MVPA. The hazard ratio of CHD was 1.36 (95% CI: 1.03-1.80) for every-30-minute/day-increase in ST relative to other movement behaviors, after adjusting for genetic risk and confounders. Reallocation of any amount of ST (as little as 1 minute/day) into the same MVPA time (but not light PA) was associated with lower CHD hazards overall and within strata of genetic risk: no evidence of interaction between genetic risk and ST (p-value: 0.739). However, differences in absolute risk of CHD for ST replaced by MVPA were greater at high genetic risk versus low genetic risk. Limitations of our study include no causal evidence derived, limited generalizability to individuals of other ethnic groups or with sub-clinical symptoms, and potential residual confounding. Conclusions: Replacing any amount of ST with an equal amount of MVPA time is associated with a lower relative risk of developing CHD at all levels of genetic susceptibility to CHD. Potential reductions in absolute risk of CHD through replacement of ST by MVPA in equivalent amounts may be greater in individuals at high genetic risk of CHD.

Association between accelerometer-measured amplitude of rest–activity rhythm and future health risk: a prospective cohort study of the UK Biobank

The health effects of rest-activity rhythm are of major interest to public health, but its associations with health outcomes remain elusive. We aimed to examine the associations between accelerometer-measured rest-activity rhythm amplitude and health risks among the general UK population. We did a prospective cohort analysis of UK Biobank participants aged 43-79 years with valid wrist-worn accelerometer data. Low rest-activity rhythm amplitude was defined as the first quintile of relative amplitude; all other quintiles were classified as high rest-activity rhythm amplitude. Outcomes of interest were defined using International Classification of Diseases 10th Revision codes and consisted of incident cancer and cardiovascular, infectious, respiratory, and digestive diseases, and all-cause and disease-specific (cardiovascular, cancer, and respiratory) mortality. Participants with a current diagnosis of any outcome of interest were excluded. We assessed the associations between decreased rest-activity rhythm amplitude and outcomes using Cox proportional hazards models. Between June 1, 2013, and Dec 23, 2015, 103 682 participants with available raw accelerometer data were enrolled. 92 614 participants (52 219 [56·4%] women and 40 395 [42·6%] men) with a median age of 64 years (IQR 56-69) were recruited. Median follow-up was 6·4 years (IQR 5·8-6·9). Decreased rest-activity rhythm amplitude was significantly associated with increased incidence of cardiovascular diseases (adjusted hazard ratio 1·11 [95% CI 1·05-1·16]), cancer (1·08 [1·01-1·16]), infectious diseases (1·31 [1·22-1·41]), respiratory diseases (1·26 [1·19-1·34]), and digestive diseases (1·08 [1·03-1·14]), as well as all-cause mortality (1·54 [1·40-1·70]) and disease-specific mortality (1·73 [1·34-2·22] for cardiovascular diseases, 1·32 [1·13-1·55] for cancer, and 1·62 [1·25-2·09] for respiratory diseases). Most of these associations were not modified by age older than 65 years or sex. Among 16 accelerometer-measured rest-activity parameters, low rest-activity rhythm amplitude had the strongest or second- strongest associations with nine health outcomes. Our results suggest that low rest-activity rhythm amplitude might contribute to major health outcomes and provide further evidence to promote risk-modifying strategies associated with rest-activity rhythm to improve health and longevity. National Natural Science Foundation of China and China Postdoctoral Science Foundation.

Rest–activity rhythm as a clinical biomarker: what are the next steps?

Rest–activity rhythm as a clinical biomarker: what are the next steps?

Developing better digital health measures of Parkinson's disease using free living data and a crowdsourced data analysis challenge.

One of the promising opportunities of digital health is its potential to lead to more holistic understandings of diseases by interacting with the daily life of patients and through the collection of large amounts of real-world data. Validating and benchmarking indicators of disease severity in the home setting is difficult, however, given the large number of confounders present in the real world and the challenges in collecting ground truth data in the home. Here we leverage two datasets collected from patients with Parkinson's disease, which couples continuous wrist-worn accelerometer data with frequent symptom reports in the home setting, to develop digital biomarkers of symptom severity. Using these data, we performed a public benchmarking challenge in which participants were asked to build measures of severity across 3 symptoms (on/off medication, dyskinesia, and tremor). 42 teams participated and performance was improved over baseline models for each subchallenge. Additional ensemble modeling across submissions further improved performance, and the top models validated in a subset of patients whose symptoms were observed and rated by trained clinicians.

AccNet24: A deep learning framework for classifying 24-hour activity behaviours from wrist-worn accelerometer data under free-living environments

ObjectiveAlthough machine learning techniques have been repeatedly used for activity prediction from wearable devices, accurate classification of 24-hour activity behaviour categories from accelerometry data remains a challenge. We developed and validated a deep learning-based framework for classifying 24-hour activity behaviours from wrist-worn accelerometers. MethodsUsing an openly available dataset with free-living wrist-based raw accelerometry data from 151 participants (aged 18–91 years), we developed a deep learning framework named AccNet24 to classify 24-hour activity behaviours. First, the acceleration signal (x, y, and z-axes) was segmented into 30-second nonoverlapping windows, and signal-to-image conversion was performed for each segment. Deep features were automatically extracted from the signal images using transfer learning and transformed into a lower-dimensional feature space. These transformed features were then employed to classify the activity behaviours as sleep, sedentary behaviour, and light-intensity (LPA) and moderate-to-vigorous physical activity (MVPA) using a bidirectional long short-term memory (BiLSTM) recurrent neural network. AccNet24 was trained and validated with data from 101 and 25 randomly selected participants and tested with the remaining unseen 25 participants. We also extracted 112 hand-crafted time and frequency domain features from 30-second windows and used them as inputs to five commonly used machine learning classifiers, including random forest, support vector machines, artificial neural networks, decision tree, and naïve Bayes to classify the 24-hour activity behaviour categories. ResultsUsing the same training, validation, and test data and window size, the classification accuracy of AccNet24 outperformed the accuracy of the other five machine learning classification algorithms by 16%–30% on unseen data. ConclusionAccNet24, relying on signal-to-image conversion, deep feature extraction, and BiLSTM achieved consistently high accuracy (>95 %) in classifying the 24-hour activity behaviour categories as sleep, sedentary, LPA, and MVPA. The next generation accelerometry analytics may rely on deep learning techniques for activity prediction.

Examination Of Cardiovascular Disease Risk Factors And Objective Physical Activity At A South African Company

PURPOSE: Ukumela impilo is a randomized controlled trial (PACTR201911656014962) investigating the effectiveness of adjustable sit-stand desks (environmental change) on sedentary behaviour and cardiovascular disease risk factors in South African office workers. The target sample (n = 120) will be randomized equally into control and experimental groups. Early baseline findings to understand anthropometric and cardiometabolic factors will be reported. METHODS: This was a cross-sectional study. Participants were full-time employed office workers of a credit and information management company in South Africa. Standard protocols were followed for the collection of simple body composition, point-of-care non-fasting glucose, glycated haemoglobin (HbA1C) and lipid profile. Wrist-worn accelerometer data were collected on 7 days of wear time, measuring sleep and physical activity. RESULTS: We have recruited 37 participants with accelerometer data thus far, 65% (n = 24) are female. The mean BMI was 27.7 ± 6.6 kg/m2 while 29.7% (n = 11) were obese. The majority (73%) had elevated blood pressures (BP) (systolic BP ≥130 and/or diastolic BP ≥80 mmHg), while 13.5% (n = 5) participants had HbA1C levels ≥6.5%. The mean total cholesterol levels were 4.03 ± 0.82 mmol/L-1and only 10.8% (n = 4) had cardioprotective HDL levels ≥1.55 mmol/L-1. The participants spent 13.4 ± 21.7 minutes/day in moderate-to-vigorous physical activity (MVPA) and slept for 7 ± 1.3 hours/night. Univariate analysis showed that MVPA was inversely correlated with total cholesterol (r = -0.33, p = 0.046). CONCLUSIONS: These preliminary results show obvious concerns around cardiovascular disease risk factors in office workers that can benefit from physical activity. Future investigation seeks to understand their sedentary behaviour patterns and examine the effectiveness of environmental changes to interrupt extended sitting time during work hours.

Exercise intensity of active video gaming in cerebral palsy: hip- versus wrist-worn accelerometer data

ABSTRACT Objective The aim of this study was to compare exercise intensity of active video games (AVGs) between hip- and wrist-worn accelerometer data in cerebral palsy (CP). Methods Twenty children and adolescents (9.35 ± 3.71 years) with CP performed two exercise sessions, completing a standardized series of AVGs. Exercise intensity was collected, while one accelerometer was fitted to wrist and hip in separate, counterbalanced sessions. Results Accelerometer counts per minute and cut-points determined were significantly different between the wrist- and hip-worn outputs (p < .001). Metabolic equivalents (METs) of performing AVGs exceeded the three METs moderate intensity threshold in wrist-worn (3.12 ± 0.86) accelerometer and hip-worn data tend to underestimate intensity (1.16 ± 0.08). Conclusions Previous studies showed METs required to perform AVGs were related to moderate intensity (3–6 METs) in CP with mild deficits. Wrist-worn accelerometer, exceeding 3 METs, seem to have higher accuracy in measuring exercise intensity of AVGs than hip-worn.

Detecting Sleep and Nonwear in 24-h Wrist Accelerometer Data from the National Health and Nutrition Examination Survey.

Estimating physical activity, sedentary behavior, and sleep from wrist-worn accelerometer data requires reliable detection of sensor nonwear and sensor wear during both sleep and wake. This study aimed to develop an algorithm that simultaneously identifies sensor wake-wear, sleep-wear, and nonwear in 24-h wrist accelerometer data collected with or without filtering. Using sensor data labeled with polysomnography ( n = 21) and directly observed wake-wear data ( n = 31) from healthy adults, and nonwear data from sensors left at various locations in a home ( n = 20), we developed an algorithm to detect nonwear, sleep-wear, and wake-wear for "idle sleep mode" (ISM) filtered data collected in the 2011-2014 National Health and Nutrition Examination Survey. The algorithm was then extended to process original raw data collected from devices without ISM filtering. Both algorithms were further validated using a polysomnography-based sleep and wake-wear data set ( n = 22) and diary-based wake-wear and nonwear labels from healthy adults ( n = 23). Classification performance (F1 scores) was compared with four alternative approaches. The F1 score of the ISM-based algorithm on the training data set using leave-one-subject-out cross-validation was 0.95 ± 0.13. Validation on the two independent data sets yielded F1 scores of 0.84 ± 0.60 for the data set with sleep-wear and wake-wear and 0.94 ± 0.04 for the data set with wake-wear and nonwear. The F1 score when using original, raw data was 0.96 ± 0.08 for the training data sets and 0.86 ± 0.18 and 0.97 ± 0.04 for the two independent validation data sets. The algorithm performed comparably or better than the alternative approaches on the data sets. A novel machine-learning algorithm was designed to recognize wake-wear, sleep-wear, and nonwear in 24-h wrist-worn accelerometer data that are applicable for ISM-filtered data or original raw data.

Importance of Overall Activity and Intensity of Activity for Cardiometabolic Risk in Those with and Without a Chronic Disease.

Higher levels of physical activity are associated with lower cardiometabolic risk. However, the relative contribution of overall activity and the intensity of activity are unclear. Our aim was to determine the relative contribution of overall activity and intensity distribution of activity to cardiometabolic risk in a cross-sectional analysis of apparently healthy office workers and in people with one or more chronic disease. Clustered cardiometabolic risk score was calculated from mean arterial pressure, high-density lipoprotein cholesterol, triglycerides and HbA1c. Open-source software (GGIR) was used to generate average acceleration and intensity gradient from wrist-worn accelerometer data for two data sets: office-workers who did not have a self-reported medical condition ( n = 399, 70% women) and adults with one or more chronic disease ( n = 1137, 34% women). Multiple linear regression analyses were used to assess the relative contribution of overall activity and intensity of activity to cardiometabolic risk. When mutually adjusted, both overall activity and intensity of activity were independently associated with cardiometabolic risk in the healthy group ( P < 0.05). However, for the CD group, although mutually adjusted associations for average acceleration were significantly associated with cardiometabolic risk ( P < 0.001), intensity was not. In healthy individuals, cardiometabolic risk was lower in those with high overall activity and/or intensity of activity, and who also undertook at least 10 min brisk walking. In those with a chronic disease, risk was lower in those who undertook at least 60 min slow walking. These findings suggest interventions aiming to optimize cardiometabolic health in healthy adults could focus on increasing both intensity and amount of physical activity. However, in those with chronic disease, increasing the amount of activity undertaken, regardless of intensity, may be more appropriate.

Are Machine Learning Models on Wrist Accelerometry Robust against Differences in Physical Performance among Older Adults?

Sufficient physical activity (PA) reduces the risk of a myriad of diseases and preserves physical capabilities in later life. While there have been significant achievements in mapping accelerations to real-life movements using machine learning (ML), errors continue to be common, particularly for wrist-worn devices. It remains unknown whether ML models are robust for estimating age-related loss of physical function. In this study, we evaluated the performance of ML models (XGBoost and LASSO) to estimate the hallmark measures of PA in low physical performance (LPP) and high physical performance (HPP) groups. Our models were built to recognize PA types and intensities, identify each individual activity, and estimate energy expenditure (EE) using wrist-worn accelerometer data (33 activities per participant) from a large sample of participants (n = 247, 57% females, aged 60+ years). Results indicated that the ML models were accurate in recognizing PA by type and intensity while also estimating EE accurately. However, the models built to recognize individual activities were less robust. Across all tasks, XGBoost outperformed LASSO. XGBoost obtained F1-Scores for sedentary (0.932 0.005), locomotion (0.946 0.003), lifestyle (0.927 0.006), and strength flexibility exercise (0.915 0.017) activity type recognition tasks. The F1-Scores for recognizing low, light, and moderate activity intensity were (0.9320.005), (0.840 0.004), and (0.869 0.005), respectively. The root mean square error for EE estimation was 0.836 0.059 METs. There was no evidence showing that splitting the participants into the LPP and HPP groups improved the models’ performance on estimating the hallmark measures of physical activities. In conclusion, using features derived from wrist-worn accelerometer data, machine learning models can accurately recognize PA types and intensities and estimate EE for older adults with high and low physical function.

A Personalized Smartphone-Delivered Just-in-time Adaptive Intervention (JitaBug) to Increase Physical Activity in Older Adults: Mixed Methods Feasibility Study.

BackgroundJust-in-time adaptive interventions (JITAIs) provide real time in-the-moment behavior change support to people when they need it most. JITAIs could be a viable way to provide personalized physical activity (PA) support to older adults in the community. However, it is unclear how feasible it is to remotely deliver a PA intervention through a smartphone to older adults or how acceptable they would find a JITAI targeting PA in everyday life.ObjectiveThe aims of this study are to describe the development of JitaBug, a personalized smartphone-delivered JITAI designed to support older adults to increase or maintain their PA level, assess the feasibility of conducting an effectiveness trial of the JitaBug intervention, and explore the acceptability of JitaBug among older adults in a free-living setting.MethodsThe intervention was developed using the Behavior Change Wheel and consisted of a wearable activity tracker (Fitbit) and a companion smartphone app (JitaBug) that delivered goal-setting, planning, reminders, and JITAI messages to encourage achievement of personalized PA goals. Message delivery was tailored based on time of day, real time PA tracker data, and weather conditions. We tested the feasibility of remotely delivering the intervention with older adults in a 6-week trial. Data collection involved assessment of PA through accelerometery and activity tracker, self-reported mood and mental well-being through ecological momentary assessment, and contextual information on PA through voice memos. Feasibility outcomes included recruitment capability and adherence to the intervention, intervention delivery in the wild, appropriateness of data collection methodology, adverse events, and participant satisfaction.ResultsOf the 46 recruited older adults (aged 56-72 years), 31 (67%) completed the intervention. The intervention was successfully delivered as intended; 87% (27/31) of the participants completed the intervention independently; 94% (2247/2390) of the PA messages were successfully delivered; 99% (2239/2261) of the Fitbit and 100% (2261/2261) of the weather data calls were successful. Valid and usable wrist-worn accelerometer data were obtained from 90% (28/31) of the participants at baseline and follow-up. On average, the participants recorded 50% (7.9/16, SD 7.3) of the voice memos, 38% (3.3/8, SD 4.2) of the mood assessments, and 50% (2.1/4, SD 1.6) of the well-being assessments through the app. Overall acceptability of the intervention was very good (23/30, 77% expressed satisfaction). Participant feedback suggested that more diverse and tailored PA messages, app use reminders, technical refinements, and an improved user interface could improve the intervention and make it more appealing.ConclusionsThis study suggests that a smartphone-delivered JITAI is an acceptable way to support PA in older adults in the community. Overall, the intervention is feasible; however, based on user feedback, the JitaBug app requires further technical refinements that may enhance use, engagement, and user satisfaction before moving to effectiveness trials.

The Dilemma of Analyzing Physical Activity and Sedentary Behavior with Wrist Accelerometer Data: Challenges and Opportunities.

Physical behaviors (e.g., physical activity and sedentary behavior) have been the focus among many researchers in the biomedical and behavioral science fields. The recent shift from hip- to wrist-worn accelerometers in these fields has signaled the need to develop novel approaches to process raw acceleration data of physical activity and sedentary behavior. However, there is currently no consensus regarding the best practices for analyzing wrist-worn accelerometer data to accurately predict individuals’ energy expenditure and the times spent in different intensities of free-living physical activity and sedentary behavior. To this end, accurately analyzing and interpreting wrist-worn accelerometer data has become a major challenge facing many clinicians and researchers. In response, this paper attempts to review different methodologies for analyzing wrist-worn accelerometer data and offer cutting edge, yet appropriate analysis plans for wrist-worn accelerometer data in the assessment of physical behavior. In this paper, we first discuss the fundamentals of wrist-worn accelerometer data, followed by various methods of processing these data (e.g., cut points, steps per minute, machine learning), and then we discuss the opportunities, challenges, and directions for future studies in this area of inquiry. This is the most comprehensive review paper to date regarding the analysis and interpretation of free-living physical activity data derived from wrist-worn accelerometers, aiming to help establish a blueprint for processing wrist-derived accelerometer data.