Topological Persistence Guided Knowledge Distillation for Wearable Sensor Data.

Deep learning methods have achieved a lot of success in various applications involving converting wearable sensor data to actionable health insights. A common application areas is activity recognition, where deep-learning methods still suffer from limitations such as sensitivity to signal quality, sensor characteristic variations, and variability between subjects. To mitigate these issues, robust features obtained by topological data analysis (TDA) have been suggested as a potential solution. However, there are two significant obstacles to using topological features in deep learning: (1) large computational load to extract topological features using TDA, and (2) different signal representations obtained from deep learning and TDA which makes fusion difficult. In this paper, to enable integration of the strengths of topological methods in deep-learning for time-series data, we propose to use two teacher networks - one trained on the raw time-series data, and another trained on persistence images generated by TDA methods. These two teachers are jointly used to distill a single student model, which utilizes only the raw time-series data at test-time. This approach addresses both issues. The use of KD with multiple teachers utilizes complementary information, and results in a compact model with strong supervisory features and an integrated richer representation. To assimilate desirable information from different modalities, we design new constraints, including orthogonality imposed on feature correlation maps for improving feature expressiveness and allowing the student to easily learn from the teacher. Also, we apply an annealing strategy in KD for fast saturation and better accommodation from different features, while the knowledge gap between the teachers and student is reduced. Finally, a robust student model is distilled, which can at test-time uses only the time-series data as an input, while implicitly preserving topological features. The experimental results demonstrate the effectiveness of the proposed method on wearable sensor data. The proposed method shows 71.74% in classification accuracy on GENEActiv with WRN16-1 (1D CNNs) student, which outperforms baselines and takes much less processing time (less than 17 sec) than teachers on 6k testing samples.

StandUPTV: Preparation and optimization phases of a mHealth intervention to reduce sedentary screen time in adults.

Recreational sedentary screen time (rSST) is the most prevalent sedentary behavior for adults outside of work, school, and sleep, and is strongly linked to poor health. StandUPTV is a mHealth trial that uses the Multiphase Optimization Strategy (MOST) framework to develop and evaluate the efficacy of three theory-based strategies for reducing rSST among adults. This paper describes the preparation and optimization phases of StandUPTV within the MOST framework. We identified three candidate components based on previous literature: (a) rSST electronic lockout (LOCKOUT), which restricts rSST through electronic means; (b) adaptive prompts (TEXT), which provides adaptive prompts based on rSST behaviors; and (c) earning rSST through increased moderate-vigorous physical activity (MVPA) participation (EARN). We also describe the mHealth iterative design process and the selection of an optimization objective. Finally, we describe the protocol of the optimization randomized controlled trial using a 23 factorial experimental design. We will enroll 240 individuals aged 23-64 y who engage in >3 h/day of rSST. All participants will receive a target to reduce rSST by 50% and be randomized to one of 8 combinations representing all components and component levels: LOCKOUT (yes vs. no), TEXT (yes vs. no), and EARN (yes vs. no). Results will support the selection of the components for the intervention package that meet the optimization objective and are acceptable to participants. The optimized intervention will be tested in a future evaluation randomized trial to examine reductions in rSST on health outcomes among adults.

SleepWell24, a Smartphone Application to Promote Adherence to Positive Airway Pressure Therapy: Feasibility and Acceptability in a Randomized Controlled Trial.

OBJECTIVE: To investigate the feasibility and acceptability of SleepWell24, a multicomponent, evidence-based smartphone application, to improve positive airway pressure therapy (PAP) adherence, among patients with obstructive sleep apnea (OSA) naive to PAP. METHODS: In a single-blind randomized controlled trial, SleepWell24, with a companion activity monitor was compared to usual care plus the activity monitor and its associated app. SleepWell24 provides objective feedback on PAP usage and sleep/physical activity patterns, and chronic disease management. Patients were recruited from two sleep medicine centers and followed over the first 60 days of PAP. Feasibility and acceptability were measured by recruitment/retention rates, app usage, differences in post-trial Treatment Evaluation Questionnaire (TEQ) scores, and patient interviews. Exploratory, intent-to-treat logistic and linear mixed models estimated PAP adherence and clinical outcomes. RESULTS: Of 103 eligible participants, 87 were enrolled (SleepWell24 n = 40, control n = 47; mean 57.6y [SD = 12.3], 44.8% female). Retention was ≥95% across arms. There were no significant differences in TEQ scores. SleepWell24 participants engaged with the app on 62.9% of trial days. PAP use was high across both arms (SleepWell24 vs. Control: mean hours 5.98 vs. 5.86). There were no differences in PAP adherence or clinical outcomes. CONCLUSIONS: SleepWell24 was feasible and acceptable among PAP-naive patients with OSA. CLINICAL TRIAL REGISTRATION: NCT03156283https://www.clinicaltrials.gov/study/NCT03156283.

Effects of Increased Standing and Light-Intensity Physical Activity to Improve Postprandial Glucose in Sedentary Office Workers: Protocol for a Randomized Crossover Trial.

BACKGROUND: Prolonged bouts of sedentary time, independent from the time spent in engaging in physical activity, significantly increases cardiometabolic risk. Nonetheless, the modern workforce spends large, uninterrupted portions of the day seated at a desk. Previous research suggests-via improved cardiometabolic biomarkers-that this risk might be attenuated by simply disrupting sedentary time with brief breaks of standing or moving. However, this evidence is derived from acute, highly controlled laboratory experiments and thus has low external validity. OBJECTIVE: This study aims to investigate if similar or prolonged cardiometabolic changes are observed after a prolonged (2-week) practice of increased brief standing and moving behaviors in real-world office settings. METHODS: This randomized crossover trial, called the WorkWell Study, will compare the efficacy of two 2-week pilot intervention conditions designed to interrupt sitting time in sedentary office workers (N=15) to a control condition. The intervention conditions use a novel smartphone app to deliver real-time prompts to increase standing (STAND) or moving (MOVE) by an additional 6 minutes each hour during work. Our primary aim is to assess intervention-associated improvements to daily postprandial glucose using continuous glucose monitors. Our secondary aim is to determine whether the interventions successfully evoke substantive positional changes and light-intensity physical activity (LPA). Other outcomes include the feasibility and acceptability of the intervention conditions, fasting blood glucose concentration, femoral artery flow-mediated dilation (f-FMD), and systolic and diastolic blood pressure. RESULTS: The trial is ongoing at the time of submission. CONCLUSIONS: This study is a novel, randomized crossover trial designed to extend a laboratory-based controlled study design into the free-living environment. By using digital health technologies to monitor and prompt participants in real time, we will be able to rigorously test the effects of breaking up sedentary behavior over a longer period of time than is seen in traditional laboratory-based studies. Our innovative approach will leverage the strengths of highly controlled laboratory and free-living experiments to achieve maximal internal and external validity. The research team's multidisciplinary expertise allows for a broad range of biological measures to be sampled, providing robust results that will extend knowledge of both the acute and chronic real-life effects of increased standing and LPA in sedentary office workers. The WorkWell Study uses a rigorous transdisciplinary protocol that will contribute to a more comprehensive picture of the beneficial effects of breaking up sitting behavior. TRIAL REGISTRATION: ClinicalTrials.gov NCT04269070; https://clinicaltrials.gov/study/NCT04269070. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/45133.

Determinants of tap water mistrust among Phoenix, Arizona Latinx adults.

The purpose of this investigation was to characterize factors that predict tap water mistrust among Phoenix, Arizona Latinx adults. Participants (n = 492, 28 ± 7 years, 37.4% female) completed water security experience-based scales and an Adapted Survey of Water Issues in Arizona. Binary logistic regression determined odds ratios (OR) with 95% confidence intervals (95% CI) for the odds of perceiving tap water to be unsafe. Of all participants, 51.2% perceived their tap water to be unsafe. The odds of mistrusting tap water were significantly greater for each additional favorable perception of bottled compared to tap water (e.g., tastes/smells better; OR = 1.94, 95% CI = 1.50, 2.50), negative home tap water experience (e.g., hard water mineral deposits and rusty color; OR = 1.32, 95% CI = 1.12, 1.56), use of alternatives to home tap water (OR = 1.25, 95% CI = 1.04, 1.51), and with decreased water quality and acceptability (OR = 1.21, 95% CI = 1.01, 1.45; P < 0.05). The odds of mistrusting tap water were significantly lower for those whose primary source of drinking water is the public supply (municipal) (OR = 0.07, 95% CI = 0.01, 0.63) and with decreased water access (OR = 0.56, 95% CI = 0.48, 0.66; P < 0.05). Latinx mistrust of tap water appears to be associated with organoleptic perceptions and reliance on alternatives to the home drinking water system.

Assessing the Pragmatic Nature of Mobile Health Interventions Promoting Physical Activity: Systematic Review and Meta-analysis.

BACKGROUND: Mobile health (mHealth) apps can promote physical activity; however, the pragmatic nature (ie, how well research translates into real-world settings) of these studies is unknown. The impact of study design choices, for example, intervention duration, on intervention effect sizes is also understudied. OBJECTIVE: This review and meta-analysis aims to describe the pragmatic nature of recent mHealth interventions for promoting physical activity and examine the associations between study effect size and pragmatic study design choices. METHODS: The PubMed, Scopus, Web of Science, and PsycINFO databases were searched until April 2020. Studies were eligible if they incorporated apps as the primary intervention, were conducted in health promotion or preventive care settings, included a device-based physical activity outcome, and used randomized study designs. Studies were assessed using the Reach, Effectiveness, Adoption, Implementation, Maintenance (RE-AIM) and Pragmatic-Explanatory Continuum Indicator Summary-2 (PRECIS-2) frameworks. Study effect sizes were summarized using random effect models, and meta-regression was used to examine treatment effect heterogeneity by study characteristics. RESULTS: Overall, 3555 participants were included across 22 interventions, with sample sizes ranging from 27 to 833 (mean 161.6, SD 193.9, median 93) participants. The study populations' mean age ranged from 10.6 to 61.5 (mean 39.6, SD 6.5) years, and the proportion of males included across all studies was 42.8% (1521/3555). Additionally, intervention lengths varied from 2 weeks to 6 months (mean 60.9, SD 34.9 days). The primary app- or device-based physical activity outcome differed among interventions: most interventions (17/22, 77%) used activity monitors or fitness trackers, whereas the rest (5/22, 23%) used app-based accelerometry measures. Data reporting across the RE-AIM framework was low (5.64/31, 18%) and varied within specific dimensions (Reach=44%; Effectiveness=52%; Adoption=3%; Implementation=10%; Maintenance=12.4%). PRECIS-2 results indicated that most study designs (14/22, 63%) were equally explanatory and pragmatic, with an overall PRECIS-2 score across all interventions of 2.93/5 (SD 0.54). The most pragmatic dimension was flexibility (adherence), with an average score of 3.73 (SD 0.92), whereas follow-up, organization, and flexibility (delivery) appeared more explanatory with means of 2.18 (SD 0.75), 2.36 (SD 1.07), and 2.41 (SD 0.72), respectively. An overall positive treatment effect was observed (Cohen d=0.29, 95% CI 0.13-0.46). Meta-regression analyses revealed that more pragmatic studies (-0.81, 95% CI -1.36 to -0.25) were associated with smaller increases in physical activity. Treatment effect sizes were homogenous across study duration, participants' age and gender, and RE-AIM scores. CONCLUSIONS: App-based mHealth physical activity studies continue to underreport several key study characteristics and have limited pragmatic use and generalizability. In addition, more pragmatic interventions observe smaller treatment effects, whereas study duration appears to be unrelated to the effect size. Future app-based studies should more comprehensively report real-world applicability, and more pragmatic approaches are needed for maximal population health impacts. TRIAL REGISTRATION: PROSPERO CRD42020169102; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=169102.

Does frequency or duration of standing breaks drive changes in glycemic response? A randomized crossover trial.

Intervention strategies to break up sitting have mostly focused on the modality (i.e., comparing different intensities and/or type of activities) and less on how frequency and duration of breaks affect health outcomes. This study compared the efficacy of different strategies to break up sitting time [i.e., high frequency, low duration standing breaks (HFLD) and low frequency, high duration standing breaks (LFHD)] in reducing postprandial glucose. Eleven sedentary and prediabetic adults (mean ± SD age = 46.8 ± 10.6 years; 73% female) participated in a cross-over trial. There were six blocks that represented all potential combinations (ordering) of the study conditions and participants were randomly assigned to a block. Each participant underwent three 7.5-h laboratory visits (1 week apart) where they engaged in either continuous sitting, HFLD, or LFHD condition while performing their usual office-related tasks. Standardized breakfast and lunch meals were provided. Postprandial mean glucose, area under the curve (AUC), and incremental area under the curve (iAUC) were evaluated using mixed models. Compared with LFHD condition, the HFLD standing breaks condition significantly lowered mean glucose by -9.94 (-14.13, -5.74) mg/dL·h after lunch, and by -6.23 (-9.93, -2.52) mg/dL·h, for the total lab visit time. Overall, the results favor frequently interrupting sitting with standing breaks to improve glycemic control in individuals with prediabetes. Further studies are needed with larger sample sizes to confirm the results.

Constrained Adaptive Distillation Based on Topological Persistence for Wearable Sensor Data.

Wearable sensor data analysis with persistence features generated by topological data analysis (TDA) has achieved great successes in various applications, however, it suffers from large computational and time resources for extracting topological features. In this paper, our approach utilizes knowledge distillation (KD) that involves the use of multiple teacher networks trained with the raw time-series and persistence images generated by TDA, respectively. However, direct transfer of knowledge from the teacher models utilizing different characteristics as inputs to the student model results in a knowledge gap and limited performance. To address this problem, we introduce a robust framework that integrates multimodal features from two different teachers and enables a student to learn desirable knowledge effectively. To account for statistical differences in multimodalities, entropy based constrained adaptive weighting mechanism is leveraged to automatically balance the effects of teachers and encourage the student model to adequately adopt the knowledge from two teachers. To assimilate dissimilar structural information generated by different style models for distillation, batch and channel similarities within a mini-batch are used. We demonstrate the effectiveness of the proposed method on wearable sensor data.

Integration of Time-Based Recommendations with Current Pediatric Health Behavior Guidelines: Implications for Obesity Prevention and Treatment in Youth.

PURPOSE OF REVIEW: Youth-onset obesity is associated with negative health outcomes across the lifespan including cardiovascular diseases, type 2 diabetes, obstructive sleep apnea, dyslipidemias, asthma, and several cancers. Pediatric health guidelines have traditionally focused on the quality and quantity of dietary intake, physical activity, and sleep. RECENT FINDINGS: Emerging evidence suggests that the timing (time of day when behavior occurs) and composition (proportion of time spent allocated to behavior) of food intake, movement (i.e., physical activity, sedentary time), and sleep may independently predict health trajectories and disease risks. Several theoretically driven interventions and conceptual frameworks feature behavior timing and composition (e.g., 24 h movement continuum, circadian science and chronobiology, intermittent fasting regimens, structured day hypothesis). These literatures are, however, disparate, with little crosstalk across disciplines. In this review, we examine dietary, sleep, and movement guidelines and recommendations for youths ages 0-18 in the context of theoretical models and empirical findings in support of time-based approaches. The review aims to inform a unifying framework of health behaviors and guide future research on the integration of time-based recommendations into current quantity and quality-based health guidelines for children and adolescents.

Effect of Weight Goals on Sitting and Moving During a Worksite Sedentary Time Reduction Intervention.

Introduction/Purpose: Although many US adults report trying to lose weight, little research has examined weight loss goals as a motivator for reducing workplace sitting and increasing physical activity. This exploratory analysis examined weight goals and the association with changes in workplace sitting, physical activity, and weight. Methods: Employees (N = 605) were drawn from worksites participating in Stand and Move at Work. Worksites (N = 24) were randomized to a multilevel behavioral intervention with (STAND+) or without (MOVE+) sit-stand workstations for 12 months; MOVE+ worksites received sit-stand workstations from 12 to 24 months. At each assessment (baseline and 3, 12, and 24 months), participants were weighed and wore activPAL monitors. Participants self-reported baseline weight goals and were categorized into the "Lose Weight Goal" (LWG) group if they reported trying to lose weight or into the "Other Weight Goal" (OWG) group if they did not. Results: Generalized linear mixed models revealed that within STAND+, LWG and OWG had similar sitting time through 12 months. However, LWG sat significantly more than OWG at 24 months. Within MOVE+, sitting time decreased after introduction of sit-stand workstations for LWG and OWG, although LWG sat more than OWG. Change in physical activity was minimal and weight remained stable in all groups. Conclusions: Patterns of change in workplace sitting were more favorable in OWG relative to LWG, even in the absence of notable weight change. Expectations of weight loss might be detrimental for reductions in workplace sitting. Interventionists may want to emphasize non-weight health benefits of reducing workplace sitting.

A Smartphone Physical Activity App for Patients in Alcohol Treatment: Single-Arm Feasibility Trial.

BACKGROUND: Alcohol use disorder (AUD) is a significant public health concern worldwide. Alcohol consumption is a leading cause of death in the United States and has a significant negative impact on individuals and society. Relapse following treatment is common, and adjunct intervention approaches to improve alcohol outcomes during early recovery continue to be critical. Interventions focused on increasing physical activity (PA) may improve AUD treatment outcomes. Given the ubiquity of smartphones and activity trackers, integrating this technology into a mobile app may be a feasible, acceptable, and scalable approach for increasing PA in individuals with AUD. OBJECTIVE: This study aims to test the Fit&Sober app developed for patients with AUD. The goals of the app were to facilitate self-monitoring of PA engagement and daily mood and alcohol cravings, increase awareness of immediate benefits of PA on mood and cravings, encourage setting and adjusting PA goals, provide resources and increase knowledge for increasing PA, and serve as a resource for alcohol relapse prevention strategies. METHODS: To preliminarily test the Fit&Sober app, we conducted an open pilot trial of patients with AUD in early recovery (N=22; 13/22, 59% women; mean age 43.6, SD 11.6 years). At the time of hospital admission, participants drank 72% of the days in the last 3 months, averaging 9 drinks per drinking day. The extent to which the Fit&Sober app was feasible and acceptable among patients with AUD during early recovery was examined. Changes in alcohol consumption, PA, anxiety, depression, alcohol craving, and quality of life were also examined after 12 weeks of app use. RESULTS: Participants reported high levels of satisfaction with the Fit&Sober app. App metadata suggested that participants were still using the app approximately 2.5 days per week by the end of the intervention. Pre-post analyses revealed small-to-moderate effects on increase in PA, from a mean of 5784 (SD 2511) steps per day at baseline to 7236 (SD 3130) steps per day at 12 weeks (Cohen d=0.35). Moderate-to-large effects were observed for increases in percentage of abstinent days (Cohen d=2.17) and quality of life (Cohen d=0.58) as well as decreases in anxiety (Cohen d=-0.71) and depression symptoms (Cohen d=-0.58). CONCLUSIONS: The Fit&Sober app is an acceptable and feasible approach for increasing PA in patients with AUD during early recovery. A future randomized controlled trial is necessary to determine the efficacy of the Fit&Sober app for long-term maintenance of PA, ancillary mental health, and alcohol outcomes. If the efficacy of the Fit&Sober app could be established, patients with AUD would have a valuable adjunct to traditional alcohol treatment that can be delivered in any setting and at any time, thereby improving the overall health and well-being of this population. TRIAL REGISTRATION: ClinicalTrials.gov NCT02958280; https://www.clinicaltrials.gov/ct2/show/NCT02958280.

Role of Data Augmentation Strategies in Knowledge Distillation for Wearable Sensor Data.

Deep neural networks are parametrized by several thousands or millions of parameters, and have shown tremendous success in many classification problems. However, the large number of parameters makes it difficult to integrate these models into edge devices such as smartphones and wearable devices. To address this problem, knowledge distillation (KD) has been widely employed, that uses a pre-trained high capacity network to train a much smaller network, suitable for edge devices. In this paper, for the first time, we study the applicability and challenges of using KD for time-series data for wearable devices. Successful application of KD requires specific choices of data augmentation methods during training. However, it is not yet known if there exists a coherent strategy for choosing an augmentation approach during KD. In this paper, we report the results of a detailed study that compares and contrasts various common choices and some hybrid data augmentation strategies in KD based human activity analysis. Research in this area is often limited as there are not many comprehensive databases available in the public domain from wearable devices. Our study considers databases from small scale publicly available to one derived from a large scale interventional study into human activity and sedentary behavior. We find that the choice of data augmentation techniques during KD have a variable level of impact on end performance, and find that the optimal network choice as well as data augmentation strategies are specific to a dataset at hand. However, we also conclude with a general set of recommendations that can provide a strong baseline performance across databases.

Qualitative comparative analysis of the implementation fidelity of a workplace sedentary reduction intervention.

BACKGROUND: Stand and Move at Work was a 12-month, multicomponent, peer-led (intervention delivery personnel) worksite intervention to reduce sedentary time. Although successful, the magnitude of reduced sedentary time varied by intervention worksite. The purpose of this study was to use a qualitative comparative analysis approach to examine potential explanatory factors that could distinguish higher from lower performing worksites based on reduced sedentary time. METHODS: We assessed 12-month changes in employee sedentary time objectively using accelerometers at 12 worksites. We ranked worksites based on the magnitude of change in sedentary time and categorized sites as higher vs. lower performing. Guided by the integrated-Promoting Action on Research Implementation in Health Services framework, we created an indicator of intervention fidelity related to adherence to the protocol and competence of intervention delivery personnel (i.e., implementer). We then gathered information from employee interviews and surveys as well as delivery personnel surveys. These data were aggregated, entered into a truth table (i.e., a table containing implementation construct presence or absence), and used to examine differences between higher and lower performing worksites. RESULTS: There were substantive differences in the magnitude of change in sedentary time between higher (-75.2 min/8 h workday, CI95: -93.7, -56.7) and lower (-30.3 min/8 h workday, CI95: -38.3, -22.7) performing worksites. Conditions that were present in all higher performing sites included implementation of indoor/outdoor walking route accessibility, completion of delivery personnel surveys, and worksite culture supporting breaks (i.e., adherence to protocol). A similar pattern was found for implementer willingness to continue role and employees using face-to-face interaction/stair strategies (i.e., delivery personnel competence). However, each of these factors were also present in some of the lower performing sites suggesting we were unable to identify sufficient conditions to predict program success. CONCLUSIONS: Higher intervention adherence and implementer competence is necessary for greater program success. These findings illustrate the need for future research to identify what factors may influence intervention fidelity, and in turn, effectiveness. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT02566317 . Registered 2 October 2015, first participant enrolled 11 January 2016.

Sitting less and moving more for improved metabolic and brain health in type 2 diabetes: 'OPTIMISE your health' trial protocol.

BACKGROUND: Clinical practice guidelines recommend that adults with type 2 diabetes (T2D) sit less and move more throughout the day. The 18-month OPTIMISE Your Health Clinical Trial was developed to support desk-based workers with T2D achieve these recommendations. The two-arm protocol consists of an intervention and control arms. The intervention arm receives 6 months health coaching, a sit-stand desktop workstation and an activity tracker, followed by 6 months of text message support, then 6 months maintenance. The control arm receives a delayed modified intervention after 12 months of usual care. This paper describes the methods of a randomised controlled trial (RCT) evaluating the effectiveness and cost-effectiveness of the intervention, compared to a delayed intervention control. METHODS: This is a two-arm RCT being conducted in Melbourne, Australia. Desk-based workers (≥0.8 full-time equivalent) aged 35-65 years, ambulatory, and with T2D and managed glycaemic control (6.5-10.0% HbA1c), are randomised to the multicomponent intervention (target n = 125) or delayed-intervention control (target n = 125) conditions. All intervention participants receive 6 months of tailored health coaching assisting them to "sit less" and "move more" at work and throughout the day, supported by a sit-stand desktop workstation and an activity tracker (Fitbit). Participants receive text message-based extended care for a further 6-months (6-12 months) followed by 6-months of non-contact (12-18 months: maintenance). Delayed intervention occurs at 12-18 months for the control arm. Assessments are undertaken at baseline, 3, 6, 12, 15 and 18-months. Primary outcomes are activPAL-measured sitting time (h/16 h day), glycosylated haemoglobin (HbA1c; %, mmol/mol) and, cognitive function measures (visual learning and new memory; Paired Associates Learning Total Errors [adjusted]). Secondary, exploratory, and process outcomes will also be collected throughout the trial. DISCUSSION: The OPTIMISE Your Health trial will provide unique insights into the benefits of an intervention aimed at sitting less and moving more in desk-bound office workers with T2D, with outcomes relevant to glycaemic control, and to cardiometabolic and brain health. Findings will contribute new insights to add to the evidence base on initiating and maintaining behaviour change with clinical populations and inform practice in diabetes management. TRIAL REGISTRATION: ANZCTRN12618001159246 .

Accuracy and Precision of Energy Expenditure, Heart Rate, and Steps Measured by Combined-Sensing Fitbits Against Reference Measures: Systematic Review and Meta-analysis.

BACKGROUND: Although it is widely recognized that physical activity is an important determinant of health, assessing this complex behavior is a considerable challenge. OBJECTIVE: The purpose of this systematic review and meta-analysis is to examine, quantify, and report the current state of evidence for the validity of energy expenditure, heart rate, and steps measured by recent combined-sensing Fitbits. METHODS: We conducted a systematic review and Bland-Altman meta-analysis of validation studies of combined-sensing Fitbits against reference measures of energy expenditure, heart rate, and steps. RESULTS: A total of 52 studies were included in the systematic review. Among the 52 studies, 41 (79%) were included in the meta-analysis, representing 203 individual comparisons between Fitbit devices and a criterion measure (ie, n=117, 57.6% for heart rate; n=49, 24.1% for energy expenditure; and n=37, 18.2% for steps). Overall, most authors of the included studies concluded that recent Fitbit models underestimate heart rate, energy expenditure, and steps compared with criterion measures. These independent conclusions aligned with the results of the pooled meta-analyses showing an average underestimation of -2.99 beats per minute (k comparison=74), -2.77 kcal per minute (k comparison=29), and -3.11 steps per minute (k comparison=19), respectively, of the Fitbit compared with the criterion measure (results obtained after removing the high risk of bias studies; population limit of agreements for heart rate, energy expenditure, and steps: -23.99 to 18.01, -12.75 to 7.41, and -13.07 to 6.86, respectively). CONCLUSIONS: Fitbit devices are likely to underestimate heart rate, energy expenditure, and steps. The estimation of these measurements varied by the quality of the study, age of the participants, type of activities, and the model of Fitbit. The qualitative conclusions of most studies aligned with the results of the meta-analysis. Although the expected level of accuracy might vary from one context to another, this underestimation can be acceptable, on average, for steps and heart rate. However, the measurement of energy expenditure may be inaccurate for some research purposes.

Cost and cost-effectiveness of the 'Stand and Move at Work' multicomponent intervention to reduce workplace sedentary time and cardiometabolic risk.

OBJECTIVE: Few studies have reported the cost and cost-effectiveness of workplace interventions to reduce sedentary time. The purpose of this study was to complete an economic evaluation of a multilevel intervention to reduce sitting time and increase light-intensity physical activity (LPA) among employees. METHODS: We conducted a retrospective within-trial cost and cost-effectiveness analysis (CEA) to compare a 12-month multilevel intervention with (STAND+) and without (MOVE+) a sit-stand workstation, across 24 worksites (N=630 employee participants) enrolled in a cluster randomized clinical trial. We estimated the intervention costs using activity-based costing strategy. The intervention costs were further expressed as per person and per worksite. CEA was conducted using an incremental cost-effectiveness ratio (ICER) metric, expressed as costs for additional unit of sitting time (minute/day), LPA (minutes/day), cardiometabolic risk score, and quality-adjusted life years (QALY) increased/decreased at 12 months. We assessed the cost analysis and CEA from the organizational (ie, employer) perspective with a one-year time horizon. RESULTS: Total intervention costs were $134 and $72 per person, and $3939 and $1650 per worksite for the STAND+ (N worksites = 12; N employees = 354) and MOVE+ (N worksites = 12; N employees = 276) interventions, respectively. The ICER was $1 (95% CI $0.8-1.4) for each additional minute reduction of workplace sitting time (standardized to 8-hour workday); and $4656 per QALY gained at 12 months. There was a modest and non-significant change of loss of work productivity improvement (-0.03 hours, 95% CI -4.16-4.09 hours), which was associated with a $0.34 return for every $1 invested. CONCLUSIONS: The multi-level intervention with sit-stand workstations has the potential to be widely implemented to reduce workplace sitting time. Future research into work productivity outcomes in terms of cost-benefits for employers is warranted.

Early Care and Education Center Environmental Factors Associated with Product- and Process-Based Locomotor Outcomes in Preschool-Age Children.

Environmental characteristics of early care and education centers (ECECs) are an important context for preschool-aged children's development, but few studies have examined their relationship with children's locomotor skills. We examined the association between characteristics of the ECEC environment with quantitatively (i.e., product-based) and qualitatively (i.e., process-based) measured locomotor skills, using the Progressive Aerobic Cardiovascular Endurance Run (PACER) and the locomotor portion of the Children's Activity and Movement in Preschool Study (CHAMPS) motor skills protocol (CMSP), respectively. ECEC characteristics included outdoor and indoor play environment quality, outdoor and indoor play equipment, screen-time environment quality, and policy environment quality. Mean (SD) scores for the PACER (n = 142) and CSMP (n = 91) were 3.7 ± 2.3 laps and 19.0 ± 5.5 criteria, respectively, which were moderately correlated with each other (Pearson r = 0.5; p < 0.001). Linear regression models revelated that a better policy environment score was associated with fewer PACER laps. Better outdoor play and screen-time environment quality scores and more outdoor play equipment were positively associated with higher CMSP scores. ECEC environments that reflect best practice guidelines may be opportunities for locomotor skills development in preschool-aged children. ClinicalTrials.gov Identifier: NCT03261492 (8/25/17).

Prenatal yoga and excessive gestational weight gain: A review of evidence and potential mechanisms.

PURPOSE: To review the evidence of the potential mechanisms (behavioral, psychological/emotional, and physical factors) of prenatal yoga for preventing excessive gestational weight gain (GWG) in pregnant women to guide future research. MAIN BODY: Prenatal yoga is a common form of physical activity during pregnancy and includes a combination of physical postures, breath control and meditation. This review theorizes how combining physical activity (i.e., prenatal yoga postures) with the add-ons brought by prenatal yoga (e.g., breath control, meditation), might provide a more comprehensive and effective strategy to prevent excessive GWG than physical activity alone. This article a) summarizes the literature on potential mechanisms of prenatal yoga to prevent excessive GWG specifically focusing on behavioral (diet, physical activity, and sleep), psychological/emotional (self-awareness, emotion regulation, stress, mood, mindfulness) and physical factors (pregnancy discomforts), b) highlights limitations of current studies, and c) provides suggestions for future research. The findings demonstrate there is insufficient evidence that prenatal yoga improves behavioral, psychological/emotional and physical factors in pregnant women and more research is needed. Though these factors have been more strongly linked to improved weight outcomes in non-pregnant populations, further testing in pregnant women is necessary to draw definitive conclusions for the efficacy of prenatal yoga to prevent excessive GWG. CONCLUSION: Effective strategies are needed to prevent excessive GWG to encourage optimal maternal and child health outcomes. More research is warranted to evaluate the impact of prenatal yoga on weight outcomes during pregnancy and design studies to test the proposed mechanisms discussed in this review.

Topological Knowledge Distillation for Wearable Sensor Data.

Converting wearable sensor data to actionable health insights has witnessed large interest in recent years. Deep learning methods have been utilized in and have achieved a lot of successes in various applications involving wearables fields. However, wearable sensor data has unique issues related to sensitivity and variability between subjects, and dependency on sampling-rate for analysis. To mitigate these issues, a different type of analysis using topological data analysis has shown promise as well. Topological data analysis (TDA) captures robust features, such as persistence images (PI), in complex data through the persistent homology algorithm, which holds the promise of boosting machine learning performance. However, because of the computational load required by TDA methods for large-scale data, integration and implementation has lagged behind. Further, many applications involving wearables require models to be compact enough to allow deployment on edge-devices. In this context, knowledge distillation (KD) has been widely applied to generate a small model (student model), using a pre-trained high-capacity network (teacher model). In this paper, we propose a new KD strategy using two teacher models - one that uses the raw time-series and another that uses persistence images from the time-series. These two teachers then train a student using KD. In essence, the student learns from heterogeneous teachers providing different knowledge. To consider different properties in features from teachers, we apply an annealing strategy and adaptive temperature in KD. Finally, a robust student model is distilled, which utilizes the time series data only. We find that incorporation of persistence features via second teacher leads to significantly improved performance. This approach provides a unique way of fusing deep-learning with topological features to develop effective models.