Wrist-Worn Accelerometer-Brand Independent Posture Classification

Detection of sedentary time and bouts using consumer-grade wrist-worn devices: a hidden semi-Markov model

BACKGROUND

Wrist-worn data from commercially available devices has potential to characterize sedentary time for research and for clinical and public health applications. We propose a model that utilizes heart rate in addition to step count data to estimate the proportion of time spent being sedentary and the usual length of sedentary bouts.

METHODS

We developed and trained two Hidden semi-Markov models, STEPHEN (STEP and Heart ENcoder) and STEPCODE (STEP enCODEr; a steps-only based model) using consumer-grade Fitbit device data from participants under free living conditions, and validated model performance using two external datasets. We used the median absolute percentage error (MDAPE) to measure the accuracy of the proposed models against research-grade activPAL device data as the referent. Bland-Altman plots summarized the individual-level agreement with activPAL.

RESULTS

In OPTIMISE cohort, STEPHEN's estimates of the proportion of time spent sedentary had significantly (p < 0.001) better accuracy (MDAPE [IQR] = 0.15 [0.06-0.25] vs. 0.23 [0.13-0.53)]) and agreement (Bias Mean [SD]=-0.03[0.11] vs. 0.14 [0.11]) than the proprietary software, estimated the usual sedentary bout duration more accurately (MDAPE[IQR] = 0.11[0.06-0.26] vs. 0.42[0.32-0.48]), and had better agreement (Bias Mean [SD] = 3.91[5.67] minutes vs. -11.93[5.07] minutes). With the ALLO-Active dataset, STEPHEN and STEPCODE did not improve the estimation of proportion of time spent sedentary, but STEPHEN estimated usual sedentary bout duration more accurately than the proprietary software (MDAPE[IQR] = 0.19[0.03-0.25] vs. 0.36[0.15-0.48]) and had smaller bias (Bias Mean[SD] = 0.70[8.89] minutes vs. -11.35[9.17] minutes).

CONCLUSIONS

STEPHEN can characterize the proportion of time spent being sedentary and usual sedentary bout length. The methodology is available as an open access R package available from https://github.com/limfuxing/stephen/ . The package includes trained models, but users have the flexibility to train their own models.

Large-scale assessment of physical activity in a population using high-resolution hip-worn accelerometry: the German National Cohort (NAKO)

Large population-based cohort studies utilizing device-based measures of physical activity are crucial to close important research gaps regarding the potential protective effects of physical activity on chronic diseases. The present study details the quality control processes and the derivation of physical activity metrics from 100 Hz accelerometer data collected in the German National Cohort (NAKO). During the 2014 to 2019 baseline assessment, a subsample of NAKO participants wore a triaxial ActiGraph accelerometer on their right hip for seven consecutive days. Auto-calibration, signal feature calculations including Euclidean Norm Minus One (ENMO) and Mean Amplitude Deviation (MAD), identification of non-wear time, and imputation, were conducted using the R package GGIR version 2.10-3. A total of 73,334 participants contributed data for accelerometry analysis, of whom 63,236 provided valid data. The average ENMO was 11.7 ± 3.7 mg (milli gravitational acceleration) and the average MAD was 19.9 ± 6.1 mg. Notably, acceleration summary metrics were higher in men than women and diminished with increasing age. Work generated in the present study will facilitate harmonized analysis, reproducibility, and utilization of NAKO accelerometry data. The NAKO accelerometry dataset represents a valuable asset for physical activity research and will be accessible through a specified application process.

Classification of Free-Living Body Posture with ECG Patch Accelerometers: Application to the Multicenter AIDS Cohort Study

Purpose

As health studies increasingly monitor free-living heart performance via ECG patches with accelerometers, researchers will seek to investigate cardio-electrical responses to physical activity and sedentary behavior, increasing demand for fast, scalable methods to process accelerometer data. We extend a posture classification algorithm for accelerometers in ECG patches when researchers do not have ground-truth labels or other reference measurements (i.e., upright measurement).

Methods

Men living with and without HIV in the Multicenter AIDS Cohort study wore the Zio XT® for up to two weeks (n = 1,250). Our novel extensions for posture classification include (1) estimation of an upright posture for each individual without a reference upright measurement; (2) correction of the upright estimate for device removal and re-positioning using novel spherical change-point detection; and (3) classification of upright and recumbent periods using a clustering and voting process rather than a simple inclination threshold used in other algorithms. As no posture labels exist in the free-living environment, we perform numerous sensitivity analyses and evaluate the algorithm against labelled data from the Towson Accelerometer Study, where participants wore accelerometers at the waist.

Results

On average, 87.1% of participants were recumbent at 4am and 15.5% were recumbent at 1pm. Participants were recumbent 54 minutes longer on weekends compared to weekdays. Performance was good in comparison to labelled data in a separate, controlled setting (accuracy = 96.0%, sensitivity = 97.5%, specificity = 95.9%).

Conclusions

Posture may be classified in the free-living environment from accelerometers in ECG patches even without measuring a standard upright position. Furthermore, algorithms that fail to account for individuals who rotate and re-attach the accelerometer may fail in the free-living environment.

Concurrent Validity of Four Activity Monitors in Older Adults

Sedentary behaviour (SB) and physical activity (PA) have been shown to be independent modulators of healthy ageing. We thus investigated the impact of activity monitor placement on the accuracy of detecting SB and PA in older adults, as well as a novel random forest algorithm trained on data from older persons. Four monitor types (ActiGraph wGT3X-BT, ActivPAL3c VT, GENEActiv Original, and DynaPort MM+) were simultaneously worn on five anatomical sites during ten different activities by a sample of twenty older adults (70.0 (12.0) years; 10 women). The results indicated that collecting metabolic equivalent (MET) data for 60 s provided the most representative results, minimising variability. In addition, thigh-worn monitors, including ActivPAL, Random Forest, and Sedentary Sphere-Thigh, exhibited superior performance in classifying SB, with balanced accuracies ≥ 94.2%. Other monitors, such as ActiGraph, DynaPort MM+, and GENEActiv Sedentary Sphere-Wrist, demonstrated lower performance. ActivPAL and GENEActiv Random Forest outperformed other monitors in participant-specific balanced accuracies for SB classification. Only thigh-worn monitors achieved acceptable overall balanced accuracies (≥80.0%) for SB, standing, and medium-to-vigorous PA classifications. In conclusion, it is advisable to position accelerometers on the thigh, collect MET data for ≥60 s, and ideally utilise population-specific trained algorithms.

Accelerometer Use to Measure Physical Activity in Older Adults With Coronary Artery Disease: An Integrative Review

BACKGROUND

Physical activity is necessary for improved health outcomes among older adults with coronary artery disease (CAD), and device-based assessment of physical activity is preferred for accurate measurement. Many previous studies have applied accelerometry to examine physical activity in this population, but no reviews have systematically examined the application of various accelerometers to measure physical activity in older adults with CAD.

OBJECTIVE

This integrative review aimed to examine accelerometry application to measure physical activity in older adults with CAD and provide guidance for accelerometer selection and settings.

METHODS

Six databases-CINAHL, PubMed, PsycINFO, Scopus, EMBASE, and Google Scholar-were searched for information sources. Authors of selected studies applied accelerometers to measure physical activity and included adults 60 years or older with CAD.

RESULTS

Among 12 studies reviewed, 5 were randomized controlled trials, and most used an age cutoff of 65 years for older adults. The most frequently used accelerometer was the RT3, and the most common device placement was the waist/hip. Data collection duration was typically 3 consecutive days. However, many study authors did not report epoch length, sampling frequency, number of valid hours of data required per day, total number of valid days of data needed, or criteria for nonwear time.

CONCLUSIONS

On the basis of data synthesis and previous study results, triaxial research-grade accelerometers, waist/hip placement, and a 5- to 7-day monitoring period are recommended for measuring physical activity in older adults with CAD. However, the study purpose, device and participant characteristics, and physical activity outcomes of interest should be considered during device selection.

Day-to-day variability in accelerometer-measured physical activity in mid-aged Australian adults

PURPOSE

The aim was to use accelerometer data to describe day-to-day variability in physical activity in a single week, according to sociodemographic variables, in mid-aged Australian adults.

METHODS

Data were from participants in the How Areas in Brisbane Influence HealTh and AcTivity (HABITAT) study who took part in a 2014 sub-study (N = 612; Mean age 60.6 [SD 6.9; range 48-73]). Participants wore a triaxial accelerometer (ActiGraph wGT3X-BT) on their non-dominant wrist for seven days, and data were expressed as acceleration in gravitational equivalent units (1 mg = 0.001 g). These were, used to estimate daily acceleration (during waking hours) and daily time spent in moderate-vigorous physical activity (MVPA, defined as ≥ 100mg). Coefficient of variation (calculated as [standard deviation/mean of acceleration and MVPA across the seven measurement days] * 100%) was used to describe day-to-day variability.

RESULTS

Average values for both acceleration (24.1-24.8 mg/day) and MVPA (75.9-79.7 mins/day) were consistent across days of the week, suggesting little day-to-day variability (at the group level). However, over seven days, average individual day-to-day variability in acceleration was 18.8% (SD 9.3%; range 3.4-87.7%) and in MVPA was 35.4% (SD 15.6%; range 7.3-124.6%), indicating considerable day-to-day variability in some participants. While blue collar workers had the highest average acceleration (28.6 mg/day) and MVPA (102.5 mins/day), their day-to-day variability was low (18.3% for acceleration and 31.9% for MVPA). In contrast, variability in acceleration was highest in men, those in professional occupations and those with high income; and variability in MVPA was higher in men than in women.

CONCLUSION

Results show group-level estimates of average acceleration and MVPA in a single week conceal considerable day-to-day variation in how mid-age Australians accumulate their acceleration and MVPA on a daily basis. Overall, there was no clear relationship between overall volume of activity and variability. Future studies with larger sample sizes and longitudinal data are needed to build on the findings from this study and increase the generalisability of these findings to other population groups.

Arterial stiffness and the reallocation of time between device-measured 24-hour movement behaviours: A compositional data analysis

BACKGROUND AND AIMS

Arterial stiffness predicts cardiovascular morbidity and mortality. We aimed to quantify the differences in arterial stiffness associated with reallocating time between 24-h movement behaviours.

METHODS

This observational cross-sectional study included Luxembourg residents aged 25-79y who each provided ≥4 valid days of triaxial accelerometry (n = 1001). Covariable adjusted compositional isotemporal substitution models were used to examine if theoretical reallocations of time between device-measured sedentariness, the sleep period, light physical activity (PA), and moderate-to-vigorous PA (MVPA) were associated with the percentage difference in carotid-femoral pulse wave velocity (cfPWV). We further investigated if replacing sedentary time accumulated in prolonged (≥30 min) with non-prolonged (<30 min) bouts was associated with arterial stiffness. The results are presented as 30 min time exchanges (β (95% confidence interval)).

RESULTS

Beneficial associations with lower cfPWV were observed when reallocating time to MVPA from the sleep period (-1.38 (-2.63 to -0.12) %), sedentary time (-1.70 (-2.76 to -0.62) %), and light PA (-2.51 (-4.55 to -0.43) %), respectively. Larger associations in the opposite direction were observed when reallocating MVPA to the same behaviours (for example, replacing MVPA with sedentary time: 2.50 (0.85-4.18) %). Replacing prolonged with non-prolonged sedentary time was not associated with cfPWV (-0.27 (-0.86 to 0.32) %). In short sleepers, reallocating sedentary time to the sleep period was favourable (-1.96 (-3.74 to -0.15) %).

CONCLUSIONS

Increasing or at least maintaining MVPA appears to be important for arterial health in adults. Extending sleep in habitually short sleepers, specifically by redistributing sedentary time, may also be important.

Substituting device-measured sedentary time with alternative 24-hour movement behaviours: compositional associations with adiposity and cardiometabolic risk in the ORISCAV-LUX 2 study

BACKGROUND

There is a considerable burden of sedentary time in European adults. We aimed to quantify the differences in adiposity and cardiometabolic health associated with theoretically exchanging sedentary time for alternative 24 h movement behaviours.

METHODS

This observational cross-sectional study included Luxembourg residents aged 18-79 years who each provided  ≥ 4 valid days of triaxial accelerometry (n = 1046). Covariable adjusted compositional isotemporal substitution models were used to examine if statistically replacing device-measured sedentary time with more time in the sleep period, light physical activity (PA), or moderate-to-vigorous PA (MVPA) was associated with adiposity and cardiometabolic health markers. We further investigated the cardiometabolic properties of replacing sedentary time which was accumulated in prolonged (≥ 30 min) with non-prolonged (< 30 min) bouts.

RESULTS

Replacing sedentary time with MVPA was favourably associated with adiposity, high-density lipoprotein cholesterol, fasting glucose, insulin, and clustered cardiometabolic risk. Substituting sedentary time with light PA was associated with lower total body fat, fasting insulin, and was the only time-exchange to predict lower triglycerides and a lower apolipoprotein B/A1 ratio. Exchanging sedentary time with more time in the sleep period was associated with lower fasting insulin, and with lower adiposity in short sleepers. There was no significant evidence that replacing prolonged with non-prolonged sedentary time was related to outcomes.

CONCLUSIONS

Artificial time-use substitutions indicate that replacing sedentary time with MVPA is beneficially associated with the widest range of cardiometabolic risk factors. Light PA confers some additional and unique metabolic benefit. Extending sleep, by substituting sedentary time with more time in the sleep period, may lower obesity risk in short sleepers.

Device-measured physical activity and sedentary time in a national sample of Luxembourg residents: the ORISCAV-LUX 2 study

BACKGROUND

Existing information about population physical activity (PA) levels and sedentary time in Luxembourg are based on self-reported data.

METHODS

This observational study included Luxembourg residents aged 18-79y who each provided ≥4 valid days of triaxial accelerometry in 2016-18 (n=1122). Compliance with the current international PA guideline (≥150 min moderate-to-vigorous PA (MVPA) per week, irrespective of bout length) was quantified and variability in average 24h acceleration (indicative of PA volume), awake-time PA levels, sedentary time and accumulation pattern were analysed by linear regression. Data were weighted to be nationally representative.

RESULTS

Participants spent 51% of daily time sedentary (mean (95% confidence interval (CI)): 12.1 (12.0 to 12.2) h/day), 11% in light PA (2.7 (2.6 to 2.8) h/day), 6% in MVPA (1.5 (1.4 to 1.5) h/day), and remaining time asleep (7.7 (7.6 to 7.7) h/day). Adherence to the PA guideline was high (98.1%). Average 24h acceleration and light PA were higher in women than men, but men achieved higher average accelerations across the most active periods of the day. Women performed less sedentary time and shorter sedentary bouts. Older participants (aged ≥55y) registered a lower average 24h acceleration and engaged in less MVPA, more sedentary time and longer sedentary bouts. Average 24h acceleration was higher in participants of lower educational attainment, who also performed less sedentary time, shorter bouts, and fewer bouts of prolonged sedentariness. Average 24h acceleration and levels of PA were higher in participants with standing and manual occupations than a sedentary work type, but manual workers registered lower average accelerations across the most active periods of the day. Standing and manual workers accumulated less sedentary time and fewer bouts of prolonged sedentariness than sedentary workers. Active commuting to work was associated with higher average 24h acceleration and MVPA, both of which were lower in participants of poorer self-rated health and higher weight status. Obesity was associated with less light PA, more sedentary time and longer sedentary bouts.

CONCLUSIONS

Adherence to recommended PA is high in Luxembourg, but half of daily time is spent sedentary. Specific population subgroups will benefit from targeted efforts to replace sedentary time with PA.

Application of Accelerometer to Monitor Students’ Exercise Load in 50 m Round Trip

With the further advancement of microelectronics innovation and sensors, sensors can be broadly implanted in cell phone gadgets, compact gadgets, and so forth. The utilization of speed increase sensors for human running checking has expansive application possibilities. From one perspective, the everyday development of the human body is firmly connected with the physical and emotional wellness of the person. Observing the day-to-day developments of the human body is of incredible importance in planning a logical running activity plan and working on actual wellbeing. On the other hand, it is also of practical value to monitor human abnormal movements. This kind of abnormal movement caused by accidental falls can bring certain harm to the human body. Real-time monitoring of the fall can provide timely assistance to the person and reduce the risk brought by the fall. This article analyzes and summarizes the research theories and common research methods in the field of 50 m round-trip movement monitoring based on the acceleration sensor. According to the process of 50 m round-trip movement pattern recognition, the data collection, preprocessing, feature extraction, and selection of 50 m round-trip movement are evaluated. The classification and recognition of each module were analyzed. This article proposes a human body motion recognition mechanism based on acceleration sensors by looking at the three trademark upsides, the wavefront edge, wavefront limit, and time stretch between the pinnacle and valley of the speed increase sensor vertical information waveform, and joining the rule of choice tree order to accomplish the activities of hunching down, taking off, and running. To get an accurate recognizable proof and recognize ways of behaving, a human fall identification calculation is proposed. This calculation removes human movement attributes throughout the fall and focuses on four sorts of falls: forward fall, reverse fall, left fall, and right fall by utilizing the connection of the three tomahawks of the speed increase sensor. The trial results show that the normal right acknowledgment pace of the human body's 50 m full-circle running way of behaviour is more than 90%, which has specific useful application esteem.

Quantitative assessment of sitting time in ambulant adults with Muscular Dystrophy

BACKGROUND

Current investigations into physical behaviour in Muscular Dystrophy (MD) have focussed largely on physical activity (PA). Negative health behaviours such as sedentary behaviour (Physical Behaviour) and sitting time (Posture Classification) are widely recognised to negatively influence health, but by contrast are poorly reported, yet could be easier behaviours to modify.

METHODS

14 ambulant men with MD and 12 healthy controls (CTRL) subjects completed 7-days of free-living with wrist-worn accelerometry, assessing physical behaviour (SB or PA) and Posture Classification (Sitting or Standing), presented at absolute (minutes) or relative (% Waking Hours). Participant body composition (Fat Mass and Fat Free Mass) were assessed by Bioelectrical Impedance, while functional status was assessed by 10 m walk test and a functional scale (Swinyard Scale).

RESULTS

Absolute Sedentary Behaviour (2.2 Hours, p = 0.025) and Sitting Time (1.9 Hours, p = 0.030 was greater in adults with MD compared to CTRL and Absolute Physical Activity (3.4 Hours, p < 0.001) and Standing Time (3.2 Hours, p < 0.001) was lower in adults with MD compared to CTRL. Absolute hours of SB was associated with Fat Mass (Kg) (R = 0.643, p < 0.05) in ambulatory adults with MD.

DISCUSSION

This study has demonstrated increased Sedentary Behaviour (2.2 hours) and Sitting time (1.9 Hours) in adults with MD compared to healthy controls. Extended waking hours in sitting and SB raises concerns with regards to progression of potential cardio-metabolic diseases and co-morbidities in MD.

Is there a bidirectional association between sedentary behaviour and cognitive decline in older adults? Findings from the Irish Longitudinal Study on Ageing

Research on whether sedentary behaviour (SB) is related to cognitive decline in older individuals is conflicting, potentially caused by methodological differences in previous studies. To inform public health policies, we analysed both the forward and reverse association across four-years between subjective TV time and objectively-measured SB and four cognitive outcome measures in older adults. The Irish Longitudinal Study on Ageing (TILDA) quantified time spent watching TV using a questionnaire and objective physical activity patterns with a GENEActiv accelerometer. Mixed model analysis examined whether these two measures of SB related to changes in cognitive function (immediate and delayed recall, MMSE, and animal naming task) during a four-year follow-up period. Furthermore, the reverse association between changes in cognition over the preceding four years and SB was investigated. We included 1,276 participants (67 ± 9 years). Longitudinally, every hour of objective SB per day was associated with a -0.01 (95%CI = -0.03;-0.00) lower MMSE score per year. Reversely, a worse decline in immediate and delayed recall over the preceding waves was related to slightly more objective SB (B = -0.24 (95%CI = -0.41;-0.07)) and TV time (B = -0.25 (95%CI = -0.48;-0.03)) at the end of those four years. To conclude, in healthy older individuals, higher levels of objective SB are related to cognitive decline across a four-year follow-up, although the magnitude and clinical relevance are questionable. As preceding cognitive decline is associated with more SB across follow-up, this suggests that a bidirectional association is plausible.

Assessment of Physical Activity in Adults using Wrist Accelerometers

The health benefits of physical activity have been widely recognized, yet traditional measures of physical activity including questionnaires and category-based assessments of volume and intensity provide only broad estimates of daily activities. Accelerometers have advanced epidemiologic research on physical activity by providing objective and continuous measurement of physical activity in free-living conditions. Wrist-worn accelerometers have become especially popular due to low participant burden. However, the validity and reliability of wrist-worn devices for adults have yet to be summarized. Moreover, accelerometer data provide rich information on how physical activity is accumulated throughout the day, but only a small portion of these rich data have been utilized by researchers. Lastly, new methodological developments that aim to overcome some of the limitations of accelerometers are emerging. The purpose of this review is to provide an overview of accelerometry research, with a special focus on wrist-worn accelerometers. We describe briefly how accelerometers work, summarize the validity and reliability of wrist-worn accelerometers, discuss the benefits of accelerometers including measuring light-intensity physical activity, and discuss pattern metrics of daily physical activity recently introduced in the literature. A summary of large-scale cohort studies and randomized trials that implemented wrist-worn accelerometry is provided. We conclude the review by discussing new developments and future directions of research using accelerometers, with a focus on wrist-worn accelerometers.

Relationship between disease activity level and physical activity in rheumatoid arthritis using a triaxial accelerometer and self-reported questionnaire

Objective

This study evaluated the relationship between rheumatoid arthritis (RA) disease activity level and physical activity (PA) by using an accelerometer and self-reported questionnaire.

Results

The cross-sectional study was part of a cohort study designed to determine disease activity is associated with PA in RA patients. We classified patients with a Disease Activity Score 28-erythrocyte sedimentation rate (DAS28-ESR) of less than and higher than 3.2 into the low-disease-activity (LDA) group and moderate/high-disease-activity (MHDA) group, respectively. We measured the wear time, time of vigorous-intensity PA, moderate-intensity PA, light-intensity PA, and sedentary behavior per day using a triaxial accelerometer. 34 patients were included in the study. The accelerometer-measured moderate-to-vigorous PA (MVPA) was 17.2 min/day and 10.6 min/day in the LDA group and MHDA group (p < 0.05), respectively. There was no significant association between RA disease activity level and accelerometer-measured PA with adjustment for age and Functional Assessment of Chronic Illness Therapy-Fatigue score. There was no correlation between accelerometer-measured MVPA and self-reported MVPA in the MHDA group, but these factors were correlated in the LDA group (rs = 0.57, p < 0.05). In conclusion, no significant association was noted between RA disease activity level and accelerometer-measured PA.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13104-021-05666-w.

Posture and Physical Activity Detection: Impact of Number of Sensors and Feature Type

Studies using wearable sensors to measure posture, physical activity (PA), and sedentary behavior typically use a single sensor worn on the ankle, thigh, wrist, or hip. Although the use of single sensors may be convenient, using multiple sensors is becoming more practical as sensors miniaturize.

PURPOSE

We evaluated the effect of single-site versus multisite motion sensing at seven body locations (both ankles, wrists, hips, and dominant thigh) on the detection of physical behavior recognition using a machine learning algorithm. We also explored the effect of using orientation versus orientation-invariant features on performance.

METHODS

Performance (F1 score) of PA and posture recognition was evaluated using leave-one-subject-out cross-validation on a 42-participant data set containing 22 physical activities with three postures (lying, sitting, and upright).

RESULTS

Posture and PA recognition models using two sensors had higher F1 scores (posture, 0.89 ± 0.06; PA, 0.53 ± 0.08) than did models using a single sensor (posture, 0.78 ± 0.11; PA, 0.43 ± 0.03). Models using two nonwrist sensors for posture recognition (F1 score, 0.93 ± 0.03) outperformed two-sensor models including one or two wrist sensors (F1 score, 0.85 ± 0.06). However, two-sensor models for PA recognition with at least one wrist sensor (F1 score, 0.60 ± 0.05) outperformed other two-sensor models (F1 score, 0.47 ± 0.02). Both posture and PA recognition F1 scores improved with more sensors (up to seven; 0.99 for posture and 0.70 for PA), but with diminishing performance returns. Models performed best when including orientation-based features.

CONCLUSIONS

Researchers measuring posture should consider multisite sensing using at least two nonwrist sensors, and researchers measuring PA should consider multisite sensing using at least one wrist sensor and one nonwrist sensor. Including orientation-based features improved both posture and PA recognition.

Objective and subjective measurement of sedentary behavior in human adults: A toolkit

OBJECTIVES

Objectives: Human biologists are increasingly interested in measuring and comparing physical activities in different societies. Sedentary behavior, which refers to time spent sitting or lying down while awake, is a large component of daily 24 hours movement patterns in humans and has been linked to poor health outcomes such as risk of all-cause and cardiovascular mortality, independently of physical activity. As such, it is important for researchers, with the aim of measuring human movement patterns, to most effectively use resources available to them to capture sedentary behavior.

METHODS

This toolkit outlines objective (device-based) and subjective (self-report) methods for measuring sedentary behavior in free-living contexts, the benefits and drawbacks to each, as well as novel options for combined use to maximize scientific rigor. Throughout this toolkit, emphasis is placed on considerations for the use of these methods in various field conditions and in varying cultural contexts.

RESULTS

Objective measures such as inclinometers are the gold-standard for measuring total sedentary time but they typically cannot capture contextual information or determine which specific behaviors are taking place. Subjective measures such as questionnaires and 24 hours-recall methods can provide measurements of time spent in specific sedentary behaviors but are subject to measurement error and response bias.

CONCLUSIONS

We recommend that researchers use the method(s) that suit the research question; inclinometers are recommended for the measurement of total sedentary time, while self-report methods are recommended for measuring time spent in particular contexts of sedentary behavior.

Estimating Sedentary Time from a Hip- and Wrist-Worn Accelerometer

PURPOSE

This study aimed to determine the validity of existing methods to estimate sedentary behavior (SB) under free-living conditions using ActiGraph GT3X+ accelerometers (AG).

METHODS

Forty-eight young (18-25 yr) adults wore an AG on the right hip and nondominant wrist and were video recorded during four 1-h sessions in free-living settings (home, community, school, and exercise). Direct observation videos were coded for postural orientation, activity type (e.g., walking), and METs derived from the Compendium of Physical Activities, which served as the criterion measure of SB (sitting or lying posture, <1.5 METs). Thirteen methods using cut points from vertical counts per minute (CPM), counts per 15-s (CP15s), and vector magnitude (VM) counts (e.g., CPM1853VM), raw acceleration and arm angle (sedentary sphere), Euclidean norm minus one (ENMO) corrected for gravity (mg) thresholds, uni- or triaxial sojourn hybrid machine learning models (Soj1x and Soj3x), random forest (RF), and decision tree (TR) models were used to estimate SB minutes from AG data. Method bias, mean absolute percent error, and their 95% confidence intervals were estimated using repeated-measures linear mixed models.

RESULTS

On average, participants spent 34.1 min per session in SB. CPM100, CPM150, Soj1x, and Soj3x were the only methods to accurately estimate SB from the hip. Sedentary sphere and ENMO44.8 overestimated SB by 3.9 and 6.1 min, respectively, whereas the remaining wrist methods underestimated SB (range, 9.5-2.5 min). In general, mean absolute percent error was lower using hip methods compared with wrist methods.

CONCLUSION

Accurate group-level estimates of SB from a hip-worn AG can be achieved using either simpler count-based approaches (CPM100 and CPM150) or machine learning models (Soj1x and Soj3x). Wrist methods did not provide accurate or precise estimates of SB. The development of large open-source free-living calibration data sets may lead to improvements in SB estimates.

Development of raw acceleration cut-points for wrist and hip accelerometers to assess sedentary behaviour and physical activity in 5-7-year-old children

This study validated sedentary behaviour (SB), moderate-to-vigorous physical activity (MVPA) and vigorous physical activity (VPA) accelerometer cut-points in 5-7-year-old children. Participants (n = 49, 55% girls) wore an ActiGraph GT9X accelerometer, recording data at 100 Hz downloaded in 1 s epochs, on both wrists and the right hip during a standardised protocol and recess. Cut-points were generated using ROC analysis with direct observation as a criterion. Subsequently, cut-points were optimised using Confidence intervals equivalency analysis and then cross-validated in a cross-validation group. SB cut-points were 36 mg (Sensitivity (Sn) = 79.8%, Specificity (Sp) = 56.8%) for non-dominant wrist, 39 mg (Sn = 75.4%, Sp = 70.2%) for dominant wrist and 20 mg (Sn = 78%, Sp = 50.1%) for hip. MVPA cut-points were 189 mg (Sn = 82.6%, Sp = 78%) for non-dominant wrist, 181 mg (Sn = 79.1%, Sp = 76%) for dominant wrist and 95 mg (Sn = 79.3%, Sp = 75.6%) for hip. VPA cut-points were 536 mg (Sn = 75.1%, Sp = 68.7%) for non-dominant wrist, 534 mg (Sn = 67.6%, Sp = 95.6%) for dominant wrist and 325 mg (Sn = 78.2%, Sp = 96.1%) for hip. All placements demonstrated adequate levels of accuracy for SB and PA assessment.

Sedentary Behavior and Chronic Disease: Mechanisms and Future Directions

Background: Recent updates to physical activity guidelines highlight the importance of reducing sedentary time. However, at present, only general recommendations are possible (ie, “Sit less, move more”). There remains a need to investigate the strength, temporality, specificity, and dose–response nature of sedentary behavior associations with chronic disease, along with potential underlying mechanisms. Methods: Stemming from a recent research workshop organized by the Sedentary Behavior Council themed “Sedentary behaviour mechanisms—biological and behavioural pathways linking sitting to adverse health outcomes,” this paper (1) discusses existing challenges and scientific discussions within this advancing area of science, (2) highlights and discusses emerging areas of interest, and (3) points to potential future directions. Results: A brief knowledge update is provided, reflecting upon current and evolving thinking/discussions, and the rapid accumulation of new evidence linking sedentary behavior to chronic disease. Research “action points” are made at the end of each section—spanning from measurement systems and analytic methods, genetic epidemiology, causal mediation, and experimental studies to biological and behavioral determinants and mechanisms. Conclusion: A better understanding of whether and how sedentary behavior is causally related to chronic disease will allow for more meaningful conclusions in the future and assist in refining clinical and public health policies/recommendations.

Association among Weather Conditions, Ambient Air Temperature, and Sedentary Time in Chinese Adults

This study is aimed to quantify the association among weather conditions, ambient air temperature, and sedentary time in Chinese adults. The participants were 3,270 Chinese users of a wrist-worn activity tracker. Their daily activity data were collected using an algorithm based on raw data to determine the sedentary time. The data of ambient air temperature and weather were collected from the meteorological data released by China Central Meteorological Observatory. Two-level linear regression analyses showed that weather conditions had a significant influence on sedentary time in Chinese adults after adjustments for some covariates were made. When the weather condition changed from rainy days to sunny and cloudy days, sedentary time might decrease by about 6.89 and 5.60 min, respectively. In conclusion, weather conditions were independently associated with sedentary time in Chinese adults. The daily sedentary time was shorter on sunny and cloudy days than on rainy days.

Towards a Portable Model to Discriminate Activity Clusters from Accelerometer Data

Few methods for classifying physical activity from accelerometer data have been tested using an independent dataset for cross-validation, and even fewer using multiple independent datasets. The aim of this study was to evaluate whether unsupervised machine learning was a viable approach for the development of a reusable clustering model that was generalisable to independent datasets. We used two labelled adult laboratory datasets to generate a k-means clustering model. To assess its generalised application, we applied the stored clustering model to three independent labelled datasets: two laboratory and one free-living. Based on the development labelled data, the ten clusters were collapsed into four activity categories: sedentary, standing/mixed/slow ambulatory, brisk ambulatory, and running. The percentages of each activity type contained in these categories were 89%, 83%, 78%, and 96%, respectively. In the laboratory independent datasets, the consistency of activity types within the clusters dropped, but remained above 70% for the sedentary clusters, and 85% for the running and ambulatory clusters. Acceleration features were similar within each cluster across samples. The clusters created reflected activity types known to be associated with health and were reasonably robust when applied to diverse independent datasets. This suggests that an unsupervised approach is potentially useful for analysing free-living accelerometer data.

Agreement Between GT3X Accelerometer and ActivPAL Inclinometer for Estimating and Detecting Changes in Different Contexts of Sedentary Time Among Adolescents

Background: This investigation aimed to analyze the agreement between the GT3X accelerometer and the ActivPAL inclinometer for estimating and detecting changes in sedentary behavior of different contexts among adolescents. Methods: Secondary data from an intervention using standing desks in the classroom conducted within 2 sixth-grade classes (intervention [n = 22] and control [n = 27]) were used. The intervention took place over 16 weeks, with activity assessments (ActivPAL and GT3X) being performed 7 days before and in the last week of the intervention. Baseline information from both groups was considered for cross-sectional analysis (209 valid days), while data from 20 participants (intervention group) were used for longitudinal analysis. Results: The authors observed that GT3X overestimated sedentary time at school (16.8%), after school (13.5%), and during weekends (7.3%) compared with ActivPAL (P < .05). Outside the school (after school [r = −.188] and on weekends [r = −.260]), there was a trend to higher overestimation among adolescents with less sedentary behavior. Longitudinally, the GT3X was unable to detect changes resulting from an intervention in school hours (ActivPAL = −34.7 min·9 h−1 vs GT3X = +6.7 min·9 h−1; P < .05). Conclusions: The authors conclude that GT3X (cut-point of <100 counts·min−1) overestimated sedentary time of free-living activities and did not detect changes resulting from a classroom standing desk intervention in adolescents.

Wrist-worn Accelerometry for Runners: Objective Quantification of Training Load

Purpose

This study aimed to apply open-source analysis code to raw habitual physical activity data from wrist-worn monitors to: 1) objectively, unobtrusively, and accurately discriminate between “running” and “nonrunning” days; and 2) develop and compare simple accelerometer-derived metrics of external training load with existing self-report measures.

Methods

Seven-day wrist-worn accelerometer (GENEActiv; Activinsights Ltd, Kimbolton, UK) data obtained from 35 experienced runners (age, 41.9 ± 11.4 yr; height, 1.72 ± 0.08 m; mass, 68.5 ± 9.7 kg; body mass index, 23.2 ± 2.2 kg·m⁻²; 19 [54%] women) every other week over 9 to 18 wk were date-matched with self-reported training log data. Receiver operating characteristic analyses were applied to accelerometer metrics (“Average Acceleration,” “Most Active-30mins,” “Mins≥400 mg”) to discriminate between “running” and “nonrunning” days and cross-validated (leave one out cross-validation). Variance explained in training log criterion metrics (miles, duration, training load) by accelerometer metrics (Mins≥400 mg, “workload (WL) 400-4000 mg”) was examined using linear regression with leave one out cross-validation.

Results

Most Active-30mins and Mins≥400 mg had >94% accuracy for correctly classifying “running” and “nonrunning” days, with validation indicating robustness. Variance explained in miles, duration, and training load by Mins≥400 mg (67%–76%) and WL400–4000 mg (55%–69%) was high, with validation indicating robustness.

Conclusions

Wrist-worn accelerometer metrics can be used to objectively, unobtrusively, and accurately identify running training days in runners, reducing the need for training logs or user input in future prospective research or commercial activity tracking. The high percentage of variance explained in existing self-reported measures of training load by simple, accelerometer-derived metrics of external training load supports the future use of accelerometry for prospective, preventative, and prescriptive monitoring purposes in runners.

Validating the Sedentary Sphere method in children: Does wrist or accelerometer brand matter?

This study aimed to validate the Sedentary Sphere posture classification method from wrist-worn accelerometers in children. Twenty-seven 9-10-year-old children wore ActiGraph GT9X (AG) and GENEActiv (GA) accelerometers on both wrists, and activPAL on the thigh while completing prescribed activities: five sedentary activities, standing with a phone, walking (criterion for all 7: observation) and 10-min free-living play (criterion: activPAL). In an independent sample, 21 children wore AG and GA accelerometers on the non-dominant wrist and activPAL for two days of free-living. Per cent accuracy, pairwise 95% equivalence tests (±10% equivalence zone) and intra-class correlation coefficients (ICC) analyses were completed. Accuracy was similar, for prescribed activities irrespective of brand (non-dominant wrist: 77-78%; dominant wrist: 79%). Posture estimates were equivalent between wrists within brand (±6%, ICC > 0.81, lower 95% CI ≥ 0.75), between brands worn on the same wrist (±5%, ICC ≥ 0.84, lower 95% CI ≥ 0.80) and between brands worn on opposing wrists (±6%, ICC ≥ 0.78, lower 95% CI ≥ 0.72). Agreement with activPAL during free-living was 77%, but sedentary time was underestimated by 7% (GA) and 10% (AG). The Sedentary Sphere can be used to classify posture from wrist-worn AG and GA accelerometers for group-level estimates in children, but future work is needed to improve the algorithm for better individual-level results.

Objective measurement of tummy time in infants (0-6 months): A validation study

The 2017 Australian and Canadian 24-hour movement guidelines recommend infants receive 30 minutes of tummy time daily. Currently, there are no validated objective measurement tools or devices to assess tummy time. The purpose of this study was to: 1) test the practicality of using devices on infants as an objective measure of tummy time, and 2) test the accuracy of developed algorithms and cut-points for predicting prone posture. Thirty-two healthy infants aged 4 to 25 weeks completed a protocol of 12 positions. Infants were placed in each position for 3 minutes while wearing a MonBaby (chest), GENEActiv (right hip) and two ActiGraphs (right hip and ankle). Direct observation was the criterion measure. The accuracy of the algorithms or cut-points to predict prone on floor, non-prone and prone supported positions were analyzed. Parents also completed a practicality questionnaire. Algorithms and cut-points to classify posture using devices from MonBaby, GENEActiv and ActiGraph (hip and ankle) were 79%, 95%, 90% and 88% accurate at defining tummy time and 100%, 98%, 100% and 96% accurate at defining non-prone positions, respectively. GENEActiv had the smallest mean difference and limits of agreement (-8.4s, limits of agreement [LoA]: -78.2 to 61.3s) for the prone on floor positions and ActiGraph Hip had the smallest mean difference and LoA for the non-prone positions (-0.2s, LoA: -1.2 to 0.9s). The majority of parents agreed all devices were practical and feasible to use with MonBaby being the preferred device. The evaluated algorithms and cut-points for GENEActiv and ActiGraph (hip) are of acceptable accuracy to objectively measure tummy time (time spent prone on floor). Accurate measurement of infant positioning practices will be important in the observation of 24-hour movement guidelines in the early years.

The backwards comparability of wrist worn GENEActiv and waist worn ActiGraph accelerometer estimates of sedentary time in children

OBJECTIVES

To examine the backward comparability of a range of wrist-worn accelerometer estimates of sedentary time (ST) with ActiGraph 100countmin^-1 waist ST estimates.

DESIGN

Cross-sectional, secondary data analysis METHODS: One hundred and eight 10-11-year-old children (65 girls) wore an ActiGraph GT3X+ accelerometer (AG) on their waist and a GENEActiv accelerometer (GA) on their non-dominant wrist for seven days. GA ST data were classified using a range of thresholds from 23 to 56mg ST estimates were compared to AG ST 100countmin^-1 data. Agreement between the AG and GA thresholds was examined using Cronbach's alpha, intraclass correlation coefficients (ICC), limits of agreement (LOA), Kappa values, percent agreement, mean absolute percent error (MAPE) and equivalency analysis.

RESULTS

Mean AG total ST was 492.4min over the measurement period. Kappa values ranged from 0.31 to 0.39. Percent agreement ranged from 68 to 69.9%. Cronbach's alpha values ranged from 0.88 to 0.93. ICCs ranged from 0.59 to 0.86. LOA were wide for all comparisons. Only the 34mg threshold produced estimates that were equivalent at the group level to the AG ST 100countmin^-1 data though sensitivity and specificity values of ∼64% and ∼74% respectively were observed.

CONCLUSIONS

Wrist-based estimates of ST generated using the 34mg threshold are comparable with those derived from the AG waist mounted 100countmin^-1 threshold at the group level. The 34mg threshold could be applied to allow group-level comparisons of ST with evidence generated using the ActiGraph 100countmin^-1 method though it is important to consider the observed sensitivity and specificity results when interpreting findings.

Segmenting accelerometer data from daily life with unsupervised machine learning

Purpose

Accelerometers are increasingly used to obtain valuable descriptors of physical activity for health research. The cut-points approach to segment accelerometer data is widely used in physical activity research but requires resource expensive calibration studies and does not make it easy to explore the information that can be gained for a variety of raw data metrics. To address these limitations, we present a data-driven approach for segmenting and clustering the accelerometer data using unsupervised machine learning.

Methods

The data used came from five hundred fourteen-year-old participants from the Millennium cohort study who wore an accelerometer (GENEActiv) on their wrist on one weekday and one weekend day. A Hidden Semi-Markov Model (HSMM), configured to identify a maximum of ten behavioral states from five second averaged acceleration with and without addition of x, y, and z-angles, was used for segmenting and clustering of the data. A cut-points approach was used as comparison.

Results

Time spent in behavioral states with or without angle metrics constituted eight and five principal components to reach 95% explained variance, respectively; in comparison four components were identified with the cut-points approach. In the HSMM with acceleration and angle as input, the distributions for acceleration in the states showed similar groupings as the cut-points categories, while more variety was seen in the distribution of angles.

Conclusion

Our unsupervised classification approach learns a construct of human behavior based on the data it observes, without the need for resource expensive calibration studies, has the ability to combine multiple data metrics, and offers a higher dimensional description of physical behavior. States are interpretable from the distributions of observations and by their duration.

Establishing Raw Acceleration Thresholds to Classify Sedentary and Stationary Behaviour in Children

This study aimed to: (1) compare acceleration output between ActiGraph (AG) hip and wrist monitors and GENEActiv (GA) wrist monitors; (2) identify raw acceleration sedentary and stationary thresholds for the two brands and placements; and (3) validate the thresholds during a free-living period. Twenty-seven from 9- to 10-year-old children wore AG accelerometers on the right hip, dominant- and non-dominant wrists, GA accelerometers on both wrists, and an activPAL on the thigh, while completing seven sedentary and light-intensity physical activities, followed by 10 minutes of school recess. In a subsequent study, 21 children wore AG and GA wrist monitors and activPAL for two days of free-living. The main effects of activity and brand and a significant activity × brand × placement interaction were observed (all p < 0.0001). Output from the AG hip was lower than the AG wrist monitors (both p < 0.0001). Receiver operating characteristic (ROC) curves established AG sedentary thresholds of 32.6 mg for the hip, 55.6 mg and 48.1 mg for dominant and non-dominant wrists respectively. GA wrist thresholds were 56.5 mg (dominant) and 51.6 mg (non-dominant). Similar thresholds were observed for stationary behaviours. The AG non-dominant threshold came closest to achieving equivalency with activPAL during free-living.

Validity of objective methods for measuring sedentary behaviour in older adults: a systematic review

Background

The evidence showing the ill health effects of prolonged sedentary behaviour (SB) is growing. Most studies of SB in older adults have relied on self-report measures of SB. However, SB is difficult for older adults to recall and objective measures that combine accelerometry with inclinometry are now available for more accurately assessing SB. The aim of this systematic review was to assess the validity and reliability of these accelerometers for the assessment of SB in older adults.

Methods

EMBASE, PubMed and EBSCOhost databases were searched for articles published up to December 13, 2017. Articles were eligible if they: a) described reliability, calibration or validation studies of SB measurement in healthy, community-dwelling individuals, b) were published in English, Portuguese or Spanish, and c) were published or in press as journal articles in peer-reviewed journals.

Results

The review identified 15 studies in 17 papers. Of the included studies, 11 assessed the ActiGraph accelerometer. Of these, three examined reliability only, seven (in eight papers) examined validity only and one (in two papers) examined both. The strongest evidence from the studies reviewed is from studies that assessed the validity of the ActiGraph. These studies indicate that analysis of the data using 60-s epochs and a vertical magnitude cut-point < 200 cpm or using 30- or 60-s epochs with a machine learning algorithm provides the most valid estimates of SB. Non-wear algorithms of 90+ consecutive zeros is also suggested for the ActiGraph.

Conclusions

Few studies have examined the reliability and validity of accelerometers for measuring SB in older adults. Studies to date suggest that the criteria researchers use for classifying an epoch as sedentary instead of as non-wear time (e.g., the non-wear algorithm used) may need to be different for older adults than for younger adults. The required number of hours and days of wear for valid estimates of SB in older adults was not clear from studies to date. More older-adult-specific validation studies of accelerometers are needed, to inform future guidelines on the appropriate criteria to use for analysis of data from different accelerometer brands.

Trial registration

PROSPERO ID# CRD42017080754 registered December 12, 2017.

Electronic supplementary material

The online version of this article (10.1186/s12966-018-0749-2) contains supplementary material, which is available to authorized users.

Measuring Sedentary Behavior by Means of Muscular Activity and Accelerometry

Sedentary Behavior (SB) is among the most frequent human behaviors and is associated with a plethora of serious chronic lifestyle diseases as well as premature death. Office workers in particular are at an increased risk due to their extensive amounts of occupational SB. However, we still lack an objective method to measure SB consistent with its definition. We have therefore developed a new measurement system based on muscular activity and accelerometry. The primary aim of the present study was to calibrate the new-developed 8-CH-EMG+ for measuring occupational SB against an indirect calorimeter during typical desk-based office work activities. In total, 25 volunteers performed nine office tasks at three typical workplaces. Minute-by-minute posture and activity classification was performed using subsequent decision trees developed with artificial intelligence data processing techniques. The 8-CH-EMG+ successfully identified all sitting episodes (AUC = 1.0). Furthermore, depending on the number of electromyography channels included, the device has a sensitivity of 83–98% and 74–98% to detect SB and active sitting (AUC = 0.85–0.91). The 8-CH-EMG+ advances the field of objective SB measurements by combining accelerometry with muscular activity. Future field studies should consider the use of EMG sensors to record SB in line with its definition.

Physical activity characterization: does one site fit all?

BACKGROUND

It is evident that a growing number of studies advocate a wrist-worn accelerometer for the assessment of patterns of physical activity a priori, yet the veracity of this site rather than any other body-mounted location for its accuracy in classifying activity is hitherto unexplored.

OBJECTIVE

The objective of this review was to identify the relative accuracy with which physical activities can be classified according to accelerometer site and analytical technique.

METHODS

A search of electronic databases was conducted using Web of Science, PubMed and Google Scholar. This review included studies written in the English language, published between database inception and December 2017, which characterized physical activities using a single accelerometer and reported the accuracy of the technique.

RESULTS

A total of 118 articles were initially retrieved. After duplicates were removed and the remaining articles screened, 32 full-text articles were reviewed, resulting in the inclusion of 19 articles that met the eligibility criteria.

CONCLUSION

There is no 'one site fits all' approach to the selection of accelerometer site location or analytical technique. Research design and focus should always inform the most suitable location of attachment, and should be driven by the type of activity being characterized.

Association between Air Quality and Sedentary Time in 3270 Chinese Adults: Application of a Novel Technology for Posture Determination

This study investigated the association between ambient air quality and sedentary time in Chinese adults. The participants were 3270 Chinese users (2021 men and 1249 women) of wrist-worn activity trackers. The data of participants’ daily activities were collected from July 2015 to October 2015. A novel algorithm based on raw accelerometer data was employed to determine sedentary time. Personal data, including sex, age, weight and height, were self-reported by the participants. Data of air quality, ambient temperature and weather were collected from the data released by the China National Environmental Monitoring Centre and the China Central Meteorological Observatory and matched in accordance with the Global Positioning System and time information. Multilevel regression analyses were conducted to investigate the association between air quality and sedentary time and adjusted for gender, age, region, body mass index, weather, temperature, weekday/weekend and monitored wake time per day. Better air quality index levels and lower concentrations of fine particulate matter were significantly associated with approximately 20 and 45 min reduction in sedentary time, respectively. Poor air quality appears to be an independent factor associated with prolonged sedentary time in Chinese adults.

Measuring activity in patients with sarcoidosis - a pilot trial of two wrist-worn accelerometer devices

Introduction: Increasing physical activity is associated with health benefits. Reduced physical activity has been noted in sarcoidosis, particularly where fatigue co-exists. Monitoring physical activity is possible with wrist-worn devices. This study compared two available devices to determine patient preference and compare wear-time, with a secondary outcome of comparing device outputs with fatigue scores. Methods: Patients with sarcoidosis wore two wrist-worn activity monitors (GENEActiv actiwatch and Actigraph GT3X-bt) separately for seven days each. Participants were randomly allocated to receive either device first. Participants completed the Fatigue Assessment Scale (FAS) questionnaire immediately before wearing the first device. All participants completed a questionnaire of their perception regarding each device after the wear period. Data from the devices was analysed for total wear time, time spent in moderate or vigorous activity (MVPA) and for time spent in sedentary behaviours. Results: Twelve patients with sarcoidosis were included. The GENEActiv device was preferred by ten (83.3%) participants. Wear time was greater with the GENEActiv device (1354 minutes/day vs 1079 minutes/day). Time spent in MVPA was slightly higher when recorded by the GENEActiv compared with the Actigraph. Moderately strong correlation was seen between FAS scores and sedentary time (r=-0.554), light activity (r=-0.585) and moderate activity (r=0.506). Discussion: A clear preference was demonstrated for the GENEActiv. This was reflected in higher wear time and suggests the device can be comfortably worn 24 hours per day. Data from this small cohort also suggests there is correlation between fatigue and activity scores in patients with sarcoidosis. (Sarcoidosis Vasc Diffuse Lung Dis 2018; 35: 62-68).

Using open source accelerometer analysis to assess physical activity and sedentary behaviour in overweight and obese adults

Background

Physical activity and sedentary behaviour are difficult to assess in overweight and obese adults. However, the use of open-source, raw accelerometer data analysis could overcome this. This study compared raw accelerometer and questionnaire-assessed moderate-to-vigorous physical activity (MVPA), walking and sedentary behaviour in normal, overweight and obese adults, and determined the effect of using different methods to categorise overweight and obesity, namely body mass index (BMI), bioelectrical impedance analysis (BIA) and waist-to-hip ratio (WHR).

Methods

One hundred twenty adults, aged 24–60 years, wore a raw, tri-axial accelerometer (Actigraph GT3X+), for 3 days and completed a physical activity questionnaire (IPAQ-S). We used open-source accelerometer analyses to estimate MVPA, walking and sedentary behaviour from a single raw accelerometer signal. Accelerometer and questionnaire-assessed measures were compared in normal, overweight and obese adults categorised using BMI, BIA and WHR.

Results

Relationships between accelerometer and questionnaire-assessed MVPA (Rs = 0.30 to 0.48) and walking (Rs = 0.43 to 0.58) were stronger in normal and overweight groups whilst sedentary behaviour were modest (Rs = 0.22 to 0.38) in normal, overweight and obese groups. The use of WHR resulted in stronger agreement between the questionnaire and accelerometer than BMI and BIA. Finally, accelerometer data showed stronger associations with BMI, BIA and WHR (Rs = 0.40 to 0.77) than questionnaire data (Rs = 0.24 to 0.37).

Conclusions

Open-source, raw accelerometer data analysis can be used to estimate MVPA, walking and sedentary behaviour from a single acceleration signal in normal, overweight and obese adults. Our data supports the use of WHR to categorise overweight and obese adults. This evidence helps researchers obtain more accurate measures of physical activity and sedentary behaviour in overweight and obese populations.

Breaking up sedentary time with seated upper body activity can regulate metabolic health in obese high-risk adults: A randomized crossover trial

AIMS

To investigate the impact of performing short bouts of seated upper body activity on postprandial blood glucose and insulin levels during prolonged sitting.

METHODS

Participants undertook two 7.5-hour experimental conditions in randomized order: (1) prolonged sitting only and (2) sitting, interspersed with 5 minutes of seated arm ergometry every 30 minutes. Blood samples were obtained while fasting and throughout the postprandial period after ingestion of two standardized meals. The incremental area under the curve (iAUC) was calculated for glucose and insulin throughout each experimental condition. A paired samples t-test was used to assess the difference in iAUC data between conditions for glucose (primary outcome) and insulin (secondary outcome).

RESULTS

Thirteen obese adults (7 women, 6 men; mean ± standard deviation [s.d.] age: 66 ± 6 years; body mass index 33.8 ± 3.8 kg/m² ) completed this investigation. Compared with the prolonged sitting-only condition, the implementation of seated arm ergometry every 30 minutes significantly reduced mean blood glucose iAUC (from 7.4 mmol/L/h [95% confidence interval {CI} 5.2, 9.5] to 3.1 mmol/L/h [95% CI 1.3, 5.0]; P = .001). Significant reductions in mean insulin iAUC (from 696 mU/L/h [95% CI 359, 1032] to 554 mU/L/h [95% CI 298, 811]; P = .047) were also observed.

CONCLUSION

Performing short bouts of arm ergometry during prolonged sitting attenuated postprandial glycaemia despite maintaining a seated posture. This may have clinical significance for those with weight-bearing difficulty who may struggle with postural change.

A cluster randomised controlled trial to evaluate the effectiveness and cost-effectiveness of the GoActive intervention to increase physical activity among adolescents aged 13-14 years

INTRODUCTION

Adolescent physical activity promotion is rarely effective, despite adolescence being critical for preventing physical activity decline. Low adolescent physical activity is likely to last into adulthood, increasing health risks. The Get Others Active (GoActive) intervention is evidence-based and was developed iteratively with adolescents and teachers. This intervention aims to increase physical activity through increased peer support, self-efficacy, group cohesion, self-esteem and friendship quality, and is implemented using a tiered-leadership system. We previously established feasibility in one school and conducted a pilot randomised controlled trial (RCT) in three schools.

METHODS AND ANALYSIS

We will conduct a school-based cluster RCT (CRCT) in 16 secondary schools targeting all year 9 students (n=2400). In eight schools, GoActive will run for two terms: weekly facilitation support from a council-funded intervention facilitator will be offered in term 1, with more distant support in term 2. Tutor groups choose two weekly activities, encouraged by older adolescent mentors and weekly peer leaders. Students gain points for trying new activities; points are entered into a between-class competition. Outcomes will be assessed at baseline, interim (week 6), postintervention (week 14-16) and 10-month follow-up (main outcome). The primary outcome will be change from baseline in daily accelerometer-assessed moderate-to-vigorous physical activity. Secondary outcomes include accelerometer-assessed activity intensities on weekdays/weekends; self-reported physical activity and psychosocial outcomes; cost-effectiveness and cost-utility analyses; mixed-methods process evaluation integrating information from focus groups and participation logs/questionnaires.

ETHICS AND DISSEMINATION

Ethical approval for the conduct of the study was gained from the University of Cambridge Psychology Research Ethics Committee. Given the lack of rigorously evaluated interventions, and the inclusion of objective measurement of physical activity, long-term follow-up and testing of causal pathways, the results of a CRCT of the effectiveness and cost-effectiveness of GoActive are expected to add substantially to the limited evidence on adolescent physical activity promotion. Workshops will be held with key stakeholders including students, parents, teachers, school governors and government representatives to discuss plans for wider dissemination of the intervention.

TRIAL REGISTRATION NUMBER

ISRCTN31583496.

Raw Accelerometer Data Analysis with GGIR R-package: Does Accelerometer Brand Matter?

PURPOSE

This study aimed to determine the agreement between outputs from contemporaneous measures of acceleration from wrist-worn GENEActiv and ActiGraph accelerometers when processed using the GGIR open source package.

METHODS

Thirty-four participants wore a GENEActiv and an ActiGraph GT3X+ on their nondominant wrist continuously for 2 d to ensure the capture of one 24-h day and one nocturnal sleep. GENEActiv.bin files and ActiGraph .csv files were analyzed with R-package GGIR version 1.2-0. Key outcome variables were as follows: wear time, average magnitude of dynamic wrist acceleration (Euclidean norm minus one [ENMO]), percentile distribution of accelerations, time spent across acceleration levels in a 40-mg resolution, time in moderate-to-vigorous physical activity (MVPA: total, 10-min bouts), and duration of nocturnal sleep.

RESULTS

There was a high agreement between accelerometer brands for all derived outcomes (wear time, MVPA, and sleep; intraclass correlation coefficient [ICC] > 0.96), ENMO (ICC = 0.99), time spent across acceleration levels (ICC > 0.93), and accelerations ≥50th percentile of the distribution (ICC > 0.82). ENMO (mean ± SD, GENEActiv = 29.9 ± 20.7 mg, ActiGraph = 27.8 ± 21.4 mg) and accelerations between the 5th and the 75th percentile of the distribution measured by the GENEActiv were significantly higher than those measured by the ActiGraph. Correspondingly, the number of minutes recorded between 0 and 40 mg was significantly greater for the ActiGraph (745 min cf. 734 min), and the number of minutes recorded between 40 and 80 mg was significantly greater for the GENEActiv (110 min cf. 105 min).

CONCLUSION

Derived outcomes (wear time, MVPA, and sleep) were similar between brands. Brands compared well for acceleration magnitudes >50-80 mg but not lower magnitudes indicative of sedentary time. Caution is advised when comparing the magnitude of ENMO between brands, but there was a high consistency between brands for the ranking of individuals for activity and sleep outcomes.

Usefulness of motion sensors to estimate energy expenditure in children and adults: a narrative review of studies using DLW

It is well documented that meeting moderate-to-vigorous physical activity guidelines of 150 min per week is protective against chronic disease, and this is likely explained by higher energy expenditure (EE). In opposition, sedentary behavior (low EE) seems to impair health outcomes. There are gold standard methods to measure EE such as the doubly labeled water (DLW) or calorimetry. These methods are highly expensive and rely on complex techniques. Motion sensors present a good alternative to estimate EE and have been validated against these reference methods. This review summarizes findings from previous reviews and the most recently published studies on the validity of different motion sensors to estimate physical activity energy expenditure (PAEE) and total energy expenditure (TEE) against DLW, and whether adding other indicators may improve these estimations in children and adults. Regardless of the recognized validity of motion sensors to estimate PAEE and TEE at the group level, individual bias is very high even when combining biometric or physiological indicators. In children, accelerometers explained 13% of DLW's PAEE variance and 31% of TEE variance. In adults, DLW's explained variance was higher, 29 and 44% for PAEE and TEE, respectively. There is no ideal device, but identifying postures seems to be relevant for both children and adults' PAEE estimates. The variance associated with the number of methodological choices that these devices require invite investigators to work with the raw data in order to standardize all these procedures and potentiate the accelerometer signal-derived information. Models that consider biometric covariates seem only to improve TEE estimations, but adding heart rate enhances PAEE estimations in both children and adults.

Comparison of linear and non-linear models for predicting energy expenditure from raw accelerometer data

This study had three purposes, all related to evaluating energy expenditure (EE) prediction accuracy from body-worn accelerometers: (1) compare linear regression to linear mixed models, (2) compare linear models to artificial neural network models, and (3) compare accuracy of accelerometers placed on the hip, thigh, and wrists. Forty individuals performed 13 activities in a 90 min semi-structured, laboratory-based protocol. Participants wore accelerometers on the right hip, right thigh, and both wrists and a portable metabolic analyzer (EE criterion). Four EE prediction models were developed for each accelerometer: linear regression, linear mixed, and two ANN models. EE prediction accuracy was assessed using correlations, root mean square error (RMSE), and bias and was compared across models and accelerometers using repeated-measures analysis of variance. For all accelerometer placements, there were no significant differences for correlations or RMSE between linear regression and linear mixed models (correlations: r  =  0.71-0.88, RMSE: 1.11-1.61 METs; p  >  0.05). For the thigh-worn accelerometer, there were no differences in correlations or RMSE between linear and ANN models (ANN-correlations: r  =  0.89, RMSE: 1.07-1.08 METs. Linear models-correlations: r  =  0.88, RMSE: 1.10-1.11 METs; p  >  0.05). Conversely, one ANN had higher correlations and lower RMSE than both linear models for the hip (ANN-correlation: r  =  0.88, RMSE: 1.12 METs. Linear models-correlations: r  =  0.86, RMSE: 1.18-1.19 METs; p  <  0.05), and both ANNs had higher correlations and lower RMSE than both linear models for the wrist-worn accelerometers (ANN-correlations: r  =  0.82-0.84, RMSE: 1.26-1.32 METs. Linear models-correlations: r  =  0.71-0.73, RMSE: 1.55-1.61 METs; p  <  0.01). For studies using wrist-worn accelerometers, machine learning models offer a significant improvement in EE prediction accuracy over linear models. Conversely, linear models showed similar EE prediction accuracy to machine learning models for hip- and thigh-worn accelerometers and may be viable alternative modeling techniques for EE prediction for hip- or thigh-worn accelerometers.

Assessing Daily Physical Activity in Older Adults: Unraveling the Complexity of Monitors, Measures, and Methods

At the 67th Gerontological Society of America Annual Meeting, a preconference workshop was convened to discuss the challenges of accurately assessing physical activity in older populations. The advent of wearable technology (eg, accelerometers) to monitor physical activity has created unprecedented opportunities to observe, quantify, and define physical activity in the real-world setting. These devices enable researchers to better understand the associations of physical activity with aging, and subsequent health outcomes. However, a consensus on proper methodological use of these devices in older populations has not been established. To date, much of the validation research regarding device type, placement, and data interpretation has been performed in younger, healthier populations, and translation of these methods to older populations remains problematic. A better understanding of these devices, their measurement properties, and the data generated is imperative to furthering our understanding of daily physical activity, its effects on the aging process, and vice versa. The purpose of this article is to provide an overview of the highlights of the preconference workshop, including properties of the different types of accelerometers, the methodological challenges of employing accelerometers in older study populations, a brief summary of ongoing aging-related research projects that utilize different types of accelerometers, and recommendations for future research directions.

Validation and Comparison of Accelerometers Worn on the Hip, Thigh, and Wrists for Measuring Physical Activity and Sedentary Behavior

Background

Recent evidence suggests that physical activity (PA) and sedentary behavior (SB) exert independent effects on health. Therefore, measurement methods that can accurately assess both constructs are needed.

Objective

To compare the accuracy of accelerometers placed on the hip, thigh, and wrists, coupled with machine learning models, for measurement of PA intensity category (SB, light-intensity PA [LPA], and moderate- to vigorous-intensity PA [MVPA]) and breaks in SB.

Methods

Forty young adults (21 female; age 22.0 ± 4.2 years) participated in a 90-minute semi-structured protocol, performing 13 activities (three sedentary, 10 non-sedentary) for 3–10 minutes each. Participants chose activity order, duration, and intensity. Direct observation (DO) was used as a criterion measure of PA intensity category, and transitions from SB to a non-sedentary activity were breaks in SB. Participants wore four accelerometers (right hip, right thigh, and both wrists), and a machine learning model was created for each accelerometer to predict PA intensity category. Sensitivity and specificity for PA intensity category classification were calculated and compared across accelerometers using repeated measures analysis of variance, and the number of breaks in SB was compared using repeated measures analysis of variance.

Results

Sensitivity and specificity values for the thigh-worn accelerometer were higher than for wrist- or hip-worn accelerometers, > 99% for all PA intensity categories. Sensitivity and specificity for the hip-worn accelerometer were 87–95% and 93–97%. The left wrist-worn accelerometer had sensitivities and specificities of > 97% for SB and LPA and 91–95% for MVPA, whereas the right wrist-worn accelerometer had sensitivities and specificities of 93–99% for SB and LPA but 67–84% for MVPA. The thigh-worn accelerometer had high accuracy for breaks in SB; all other accelerometers overestimated breaks in SB.

Conclusion

Coupled with machine learning modeling, the thigh-worn accelerometer should be considered when objectively assessing PA and SB.

The Wild Wild West: A Framework to Integrate mHealth Software Applications and Wearables to Support Physical Activity Assessment, Counseling and Interventions for Cardiovascular Disease Risk Reduction

Physical activity (PA) interventions constitute a critical component of cardiovascular disease (CVD) risk reduction programs. Objective mobile health (mHealth) software applications (apps) and wearable activity monitors (WAMs) can advance both assessment and integration of PA counseling in clinical settings and support community-based PA interventions. The use of mHealth technology for CVD risk reduction is promising, but integration into routine clinical care and population health management has proven challenging. The increasing diversity of available technologies and the lack of a comprehensive guiding framework are key barriers for standardizing data collection and integration. This paper reviews the validity, utility and feasibility of implementing mHealth technology in clinical settings and proposes an organizational framework to support PA assessment, counseling and referrals to community resources for CVD risk reduction interventions. This integration framework can be adapted to different clinical population needs. It should also be refined as technologies and regulations advance under an evolving health care system landscape in the United States and globally.