Fitbit Charge Research Articles

SAT291 Remote Monitoring Of Biometric, Proteomic, And Patient-reported Outcomes In Adrenal Insufficiency

Abstract Disclosure: A. Ben-Shlomo: None. B. Chazarin: None. G. Gresham: None. M. Choudhary: None. K. Barnhill: None. K. Haptonstall: None. J. Van Eyk: None. Background: Damaged hypothalamic-pituitary adrenal axis may result in adrenal insufficiency (AI) and require hydrocortisone replacement. Currently, there are no biometric or circulating biomarkers to assess hydrocortisone replacement adequacy. Dosing is based on symptomatology, often resulting in over- or undertreatment. Methods: We used remote monitoring precision health solutions to identify AI molecular and biometric biomarkers over 2 weeks. Using remote sampling devices (Mitra Neoteryx), 10-µL dried blood samples were collected weekly for protein analysis by AI subjects before and 5 hours after daily hydrocortisone doses and age- and sex matched adrenal sufficient (AS) subjects at approximately the same times. Continuous physical activity were measured with a wearable biosensor (Fitbit Charge HR) along with remote systolic (SBP) and diastolic (DBP) blood pressure and heart rate (HR) (Omron Evolv). Self-reported AddiQoL questionnaires were completed twice daily. For biometric endpoints and questionnaires, data are summarized as mean ± SD using two-sided t-tests, where a p value <0.05 is considered statistically significant. For proteomic analysis, median value was calculated, and statistical testing identified differentially expressed proteins (DEPs; FDR<5%). Results: Fifteen AI (6 primary AI [PAI] and 9 secondary AI [SAI] and 15 AS subjects completed the study. AI group mean age was 46.3±15.7 and AS was 41.7±13.2. In both groups, 80% were female. Adherence to home monitoring was 100% in AI subjects and >90% in AS (13/15 assessable for biometrics (activity and BP) and 14/15 for AddiQoL). SBP was lower in PAI (106.2±7.9/68.3±7.6 [n=5]) vs SAI (119.1±15.3/71.1±8.9 [n=8]; p=0.04). Nighttime BP (22:00-05:00) for both PAI (102.3±7.8/68.5±7.9) and SAI (101.2±8.2/63.8±9) were significantly lower than AS (115.5±15.6; p=0.04 and p=0.02, respectively). AI heart rate was lower vs AS at night (61.4±11.6 [n=13] vs 72.5±11.1 [n=11]; p=0.01), especially in PAI (57.7±5 [n=5]; p=0.001). Average daily activity was higher in AS than AI as measured by steps (9010 vs 7787; p=0.02), miles (3.94 vs 3.36; p=0.03) and stair flights (14.19 vs 9.85; p=0.01). On AddiQoL, AI subjects scored lower than AS subjects (126.8±18.1 vs 143.3±14.9, respectively; p<0.0001). Across all dried blood samples, 444 proteins were quantified. Comparing AI vs AS, 78 were identified before morning hydrocortisone dose vs 88 before afternoon dose. Forty-nine DEPs were commonly identified and showed the same regulation trends. Nine DEPs before and 18 DEPs after morning and afternoon dosing, respectively, were identified in AI. None of these DEPs were common, indicating potential blood proteome signature changes in the morning and in the afternoon. Conclusion: Remote monitoring of biometrics and blood proteome in AI is feasible and informative and may identify biomarkers to optimize hydrocortisone replacement therapy. Presentation: Saturday, June 17, 2023

Interchangeability of Research and Commercial Wearable Device Data for Assessing Associations With Cardiometabolic Risk Markers

Introduction: While there is evidence on agreement, it is unknown whether commercial wearables can be used as surrogates for research-grade devices when investigating links with markers of cardiometabolic risk. Therefore, the aim of this study was to investigate whether data from a commercial wearable device could be used to assess associations between behavior and cardiometabolic risk markers, compared with physical activity from a research-grade monitor. Methods: Forty-five adults concurrently wore a wrist-worn Fitbit Charge 2 and a waist-worn ActiGraph wGT3X-BT during waking hours over 7 consecutive days. Log-linear regression models were fitted, and predictive fit via a one-out cross-validation was performed for each device between behavioral (steps, and light and moderate-to-vigorous physical activity) and cardiometabolic variables (body mass index, weight, body fat percentage, systolic and diastolic blood pressure, glycated haemoglobin, grip strength, estimated maximal oxygen uptake, and waist circumference). Results: Overall, step count was the most consistent predictor of cardiometabolic risk factors, with negative associations across both Fitbit and ActiGraph devices for body mass index (−0.017 vs. −0.020, p < .01), weight (−0.014 vs. −0.017, p < .05), body fat percentage (−0.021 vs. −0.022, p < .01), and waist circumference (−0.013 vs. −0.015, p < .01). Neither device was found to provide a consistently better prediction across all included cardiometabolic risk markers. Conclusions: Step count data from a commercial-grade wearable device showed similar associations and predictive relationships with cardiometabolic risk markers compared with a research-grade wearable device, providing preliminary support for their use in health research.

The impact of mindfulness training on posttraumatic stress disorder symptoms, subjective sleep quality, and objective sleep outcomes in police officers.

Sleep disturbances cooccur with posttraumatic stress disorder (PTSD) and are often correlated with PTSD severity. Previous research has shown that sleep problems mediate the relationship between PTSD and negative physical and mental health outcomes but has relied on self-reported sleep quality. We tested the effects of mindfulness training-previously shown to improve sleep quality and reduce PTSD symptoms-on subjective and objective sleep metrics and relationships with reduced PTSD symptoms. Following baseline data collection in 114 law enforcement officers, we randomly assigned participants to either an 8-week mindfulness training group or a waitlist control group. We repeated assessments immediately posttraining and at 3-month follow-up. Self-reported PTSD symptoms and subjective sleep quality were measured at each visit with the PTSD checklist and Pittsburgh Sleep Quality Index (PSQI), respectively. Participants also wore a Fitbit Charge 2 continuously over the course of a 4- to 6-day work week following each visit, from which we extracted two distinct objective sleep metrics: total minutes of sleep and sleep efficiency. At baseline, PTSD symptoms were correlated with PSQI scores but not objective Fitbit metrics. Relative to waitlist, mindfulness training led to improved subjective sleep quality and reduced PTSD symptoms. Reduced PTSD symptoms mediated the improvement in subjective sleep quality following mindfulness training. Neither objective sleep metric demonstrated improvements following mindfulness training, nor did these metrics mediate reduced PTSD symptoms. This study provides evidence linking improved subjective sleep quality, but not objective sleep markers, to reductions in PTSD symptoms following mindfulness training. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Feasibility and preliminary efficacy of a novel web-based physical activity intervention in adults with overweight/obesity: A pilot randomized controlled trial

BackgroundOnly 14% of adults with obesity attain federal guidelines for physical activity (PA), but few interventions address obesity-specific barriers to PA. We designed the web-based Physical Activity for The Heart (PATH) intervention to address this gap. PurposeTest the feasibility and preliminary efficacy of PATH for promoting PA and reducing cardiovascular disease (CVD) risk in adults with overweight/obesity. MethodsIn a 12-week pilot RCT, participants were randomized to PATH (n = 41) or wait-list control (n = 41) groups. Treatment group received access to PATH and met twice/month with a remote coach. The control group received a self-help PA guide and newsletters on general health. Moderate-to-vigorous PA (MVPA) was assessed via Actigraph-GT3X, steps via Fitbit Charge 2™, weight via smart scale, blood pressure (BP) via Omron BP device, and lipids/HbAIC via dry blood spot. Linear mixed modeling examined between- and within-group differences in PA and CVD risk. ResultsThe sample (N = 82) was on average 55.9 ± 8.2 years old; mean BMI 35.5 ± 6.2 kg/m2; 57.3% white and 80.5% female. Recruitment lasted 6-months, and 12-week retention was 96.3%. Treatment group accessed PATH ≥twice/week (92.1%), spent ≥10 min/visit (89.5%) and thought the site was culturally appropriate (79%). At 12 wks, the PATH group had greater mean changes in weekly MVPA (+58.9 vs. +0.9 min, p = .024) and daily steps (+1246.4 vs. -64.2 steps, p = .002) compared to the control group. Also, the PATH group improved in weight, BMI, body fat, waist circumference, and BP (p < .05). ConclusionThe PATH intervention is feasible/acceptable and demonstrated preliminary efficacy for promoting PA among adults with overweight/obesity.

Ecological Momentary Assessments and Passive Sensing in the Prediction of Short-Term Suicidal Ideation in Young Adults

Advancements in technology, including mobile-based ecological momentary assessments (EMAs) and passive sensing, have immense potential to identify short-term suicide risk. However, the extent to which EMA and passive data, particularly in combination, have utility in detecting short-term risk in everyday life remains poorly understood. To examine whether and what combinations of self-reported EMA and sensor-based assessments identify next-day suicidal ideation. In this intensive longitudinal prognostic study, participants completed EMAs 4 times daily and wore a sensor wristband (Fitbit Charge 3) for 8 weeks. Multilevel machine learning methods, including penalized generalized estimating equations and classification and regression trees (CARTs) with repeated 5-fold cross-validation, were used to optimize prediction of next-day suicidal ideation based on time-varying features from EMAs (affective, cognitive, behavioral risk factors) and sensor data (sleep, activity, heart rate). Young adult patients who visited an emergency department with recent suicidal ideation and/or suicide attempt were recruited. Identified via electronic health record screening, eligible individuals were contacted remotely to complete enrollment procedures. Participants (aged 18 to 25 years) completed 14 708 EMA observations (64.4% adherence) and wore a sensor wristband approximately half the time (55.6% adherence). Data were collected between June 2020 and July 2021. Statistical analysis was performed from January to March 2023. The outcome was presence of next-day suicidal ideation. Among 102 enrolled participants, 83 (81.4%) were female; 6 (5.9%) were Asian, 5 (4.9%) were Black or African American, 9 (8.8%) were more than 1 race, and 76 (74.5%) were White; mean (SD) age was 20.9 (2.1) years. The best-performing model incorporated features from EMAs and showed good predictive accuracy (mean [SE] cross-validated area under the receiver operating characteristic curve [AUC], 0.84 [0.02]), whereas the model that incorporated features from sensor data alone showed poor prediction (mean [SE] cross-validated AUC, 0.56 [0.02]). Sensor-based features did not improve prediction when combined with EMAs. Suicidal ideation-related features were the strongest predictors of next-day ideation. When suicidal ideation features were excluded, an alternative EMA model had acceptable predictive accuracy (mean [SE] cross-validated AUC, 0.76 [0.02]). Both EMA models included features at different timescales reflecting within-day, end-of-day, and time-varying cumulative effects. In this prognostic study, self-reported risk factors showed utility in identifying near-term suicidal thoughts. Best-performing models required self-reported information, derived from EMAs, whereas sensor-based data had negligible predictive accuracy. These results may have implications for developing decision algorithms identifying near-term suicidal thoughts to guide risk monitoring and intervention delivery in everyday life.

Knee Extension Exercise Effects on Physiological and Psychophysical Performance: Normal Weight Versus Obese Office Workers.

Knee extension exercise is useful and practical for obese and overweight people as this form of exercise is effective in minimizing body weight loading on joints and improving the body's physiological function. This study aimed to compare the physiological and psychophysical parameters of office workers while computer-based working in an active workstation equipped with an active footrest (AFR) prototype with a mechanism for performing knee extension exercises in a sitting position, and also to compare the physiological and psychophysical parameters among normal-weight and obese office workers. In this quasi-experimental study, the physiological parameters of heart rate (HR) and energy expenditure (EE) (measured with the Fitbit Charge HR smartwatch) were measured in two cross-over random sessions for 32 office workers (16 normal-weight and 16 obese) aged 28 to 50 years (M = 42.72, standard deviation [SD] = 4.37) while performing office tasks in sitting and active workstations (equipped with AFR). Perceived physical exertion, comfort, fatigue, and liking were also measured by rating the participants. Short-term activity of the participants with AFR performing computer tasks significantly improved physiological and psychophysical parameters compared with the participants in sitting workstations. However, there was no significant difference in the effect of AFR on physiological and psychophysical parameters between normal-weight and obese participants. Given the significant increase in EE and HR resulting from exercise with AFR compared with the conventional workstation, the use of AFR can help office workers achieve the minimum standard of physical activity at their workplace.

A Digital Coach (E-Supporter 1.0) to Support Physical Activity and a Healthy Diet in People With Type 2 Diabetes: Acceptability and Limited Efficacy Testing.

A healthy lifestyle, including regular physical activity and a healthy diet, is increasingly part of type 2 diabetes (T2D) management. As many people with T2D have difficulty living and maintaining a healthy lifestyle, there is a need for effective interventions. eHealth interventions that incorporate behavior change theories and tailoring are considered effective tools for supporting a healthy lifestyle. The E-Supporter 1.0 digital coach contains eHealth content for app-based eHealth interventions and offers tailored coaching regarding physical activity and a healthy diet for people with T2D. This study aimed to assess the acceptability of E-Supporter 1.0 and explore its limited efficacy on physical activity, dietary behavior, the phase of behavior change, and self-efficacy levels. Over a span of 9 weeks, 20 individuals with T2D received daily motivational messages and weekly feedback derived from behavioral change theories and determinants through E-Supporter 1.0. The acceptability of the intervention was assessed using telephone-conducted, semistructured interviews. The interview transcripts were coded using inductive thematic analysis. The limited efficacy of E-Supporter 1.0 was explored using the Fitbit Charge 2 to monitor step count to assess physical activity and questionnaires to assess dietary behavior (using the Dutch Healthy Diet index), phase of behavior change (using the single-question Self-Assessment Scale Stages of Change), and self-efficacy levels (using the Exercise Self-Efficacy Scale). In total, 5 main themes emerged from the interviews: perceptions regarding remote coaching, perceptions regarding the content, intervention intensity and duration, perceived effectiveness, and overall appreciation. The participants were predominantly positive about E-Supporter 1.0. Overall, they experienced E-Supporter 1.0 as a useful and easy-to-use intervention to support a better lifestyle. Participants expressed a preference for combining E-Supporter with face-to-face guidance from a health care professional. Many participants found the intensity and duration of the intervention to be acceptable, despite the coaching period appearing relatively short to facilitate long-term behavior maintenance. As expected, the degree of tailoring concerning the individual and external factors that influence a healthy lifestyle was perceived as limited. The limited efficacy testing showed a significant improvement in the daily step count (z=-2.040; P=.04) and self-efficacy levels (z=-1.997; P=.046) between baseline and postintervention. Diet was improved through better adherence to Dutch dietary guidelines. No significant improvement was found in the phase of behavior change (P=.17), as most participants were already in the maintenance phase at baseline. On the basis of this explorative feasibility study, we expect E-Supporter 1.0 to be an acceptable and potentially useful intervention to promote physical activity and a healthy diet in people with T2D. Additional work needs to be done to further tailor the E-Supporter content and evaluate its effects more extensively on lifestyle behaviors.

How was remote technology accepted among cardiac rehabilitees in outpatient self-rehabilitation context?

The purpose of the article The purpose of this study was to investigate cardiac rehabilitees’ experiences with wrist-worn activity monitor (Fitbit Charge HR) and internet-based coaching software (Movendos m-coach) in outpatient rehabilitation to promote physical activity and a healthy lifestyle. Materials and methods A 12-month cardiac rehabilitation intervention consisted of three 5-day inpatient periods in a rehabilitation centre, between which participants (n = 32; age mean 62.3 years old (SD 6.3), men n = 24 (75%)) used an activity monitor and internet-based coaching software at home. Experiences with the use of technologies were measured at the end of the 12-month intervention with a questionnaire that was designed using the framework of the Technology Acceptance Model (Likert scale 1–7; 1 = strongly agree and 7 = strongly disagree). Data analysis was based on indicators describing the average experiences and Spearman’s correlation coefficient. Results Both technologies had positive acceptance (internet application mean: 3.2, 95% CI: 2.6–3.7; activity monitor mean 2.1, 95% CI: 1.7–2.5). Sociodemographic factors, physical activity, and familiarity with using technology in daily life were not correlated with technology acceptance of either technology. Conclusions Cardiac rehabilitees had positive acceptance on wrist worn activity monitor and internet-based coaching software, which suggests the use of these tools in outpatient cardiac rehabilitation. Trial Registration ISRCTN Registry, url: https://doi.org/10.1186/ISRCTN61225589, reg. number: ISRCTN61225589

The Association between Neighborhood Walkability and Physical Activity in a Behavioral Weight Loss Trial Testing the Addition of Remotely Delivered Feedback Messages to Self-Monitoring

Adding feedback messages (FB) to self-monitoring (SM) may reinforce behavior change. However, socio-environmental conditions (e.g., limited access to parks or low walkability) may limit one’s ability to respond to FB focused on physical activity (PA). In this analysis, we hypothesized that high neighborhood walkability will positively modify the treatment effect of FB on PA, and residents of high walkability neighborhoods will achieve higher PA levels at 12 months than those in low walkability neighborhoods. The study is a secondary analysis of a 12-month behavioral weight-loss trial. Adults with overweight/obesity were randomized to SM + FB (n = 251) or SM alone (n = 251). SM + FB group received smartphone pop-up messages thrice/week tailored to their PA SM data. The assessment included neighborhood walkability via Walk Score (low [<50] vs. high [≥50]), moderate to vigorous PA (MVPA) and step count via Fitbit Charge 2™, and weight via smart scale. We report adjusted linear regression coefficients (b) with standard errors (SE). The analysis included participants who were primarily white, female, and with obesity. In adjusted models, neighborhood walkability did not moderate the effect of treatment assignment on log-transformed (ln) MVPA or steps count over 12 months. The SM + FB group had greater lnMVPA than the SM group, but lnMVPA and steps were similar between walkability groups. There were no significant interactions for group and time or group, time, and walkability. These findings suggest that adding FB to SM had a small but significant positive impact on PA over 12 months, but neighborhood walkability did not moderate the treatment effect of FB on PA.

251-OR: The Impact of Identifying Sleep with a Fitbit on Calculated Rates of Nocturnal Hypoglycemia—The Hypo-Metrics Study

Consensus defines nocturnal hypoglycemia as occurring between 00:00 and 06:00 hrs. However, this may introduce bias, as for many, these hours only partially match real sleeping hours. Therefore, we compared rates of hypoglycemia while asleep (defined by Fitbit Charge 4) to rates of conventionally-defined nocturnal hypoglycemia. Hypo-METRICS study participants wore a Fitbit charge 4 and blinded FreeStyle Libre glucose monitor for up to 70 days. Every hypoglycemia episode (≤70mg/dL or ≤54mg/dL for ≥15 min) was classified as ‘asleep’ or ‘awake’ based on Fitbit sleep intervals and ‘nocturnal’ (00:00-06:00 hrs) or ‘day-time’. We compared weekly sensor-detected hypoglycemia (SDH) rates whilst asleep to nocturnal SDH rates using a paired sample Wilcoxon test. We included 542 participants (269 with type 1 diabetes, median (IQR) age 56 (44-66) years, HbA1c 7.3%(6.7-7.9), with 69 (65-70) nights with sleep data/participant). Sleep duration was 6.1 (5.2-6.8) hours. Median weekly rates of SDH while asleep were higher than rates of nocturnal SDH both ≤70mg/dL (1.3 (0.6-2.5) vs 0.9 (0.4-1.8), p&amp;lt;0.001) and ≤54mg/dL (0.4 (0.2-0.9) vs 0.3 (0.1-0.7), p&amp;lt;0.001). Stratifying by type of diabetes did not change the direction or magnitude of the differences in SDH rates. Using clock time to estimate nocturnal hypoglycemia underestimates hypoglycemia whilst asleep by up to a third. Using activity trackers such as Fitbit to identify sleep offers a novel way to estimate hypoglycemia whilst asleep to better reflect lived experience of nocturnal hypoglycemia. Disclosure G.Martine-edith: Other Relationship; Novo Nordisk A/S. R.J.Mccrimmon: Advisory Panel; Sanofi, Speaker's Bureau; Novo Nordisk A/S. E.Renard: Consultant; Abbott Diabetes, Dexcom, Inc., AstraZeneca, Boehringer-Ingelheim, Eli Lilly and Company, Insulet Corporation, MannKind Corporation, Novo Nordisk, Sanofi, Roche Diabetes Care. S.R.Heller: Advisory Panel; Zealand Pharma A/S, Zucara Therapeutics, Other Relationship; Eli Lilly and Company, Research Support; Dexcom, Inc., Speaker's Bureau; Novo Nordisk, Medtronic. M.Evans: Advisory Panel; Zucara Therapeutics, Pila Pharma, Dexcom, Inc., Other Relationship; Novo Nordisk, AstraZeneca, Abbott Diabetes, Speaker's Bureau; Eli Lilly and Company. J.K.Mader: Advisory Panel; Novo Nordisk A/S, Abbott Diabetes, Roche Diabetes Care, Eli Lilly and Company, Sanofi, Medtronic, Becton, Dickinson and Company, Pharmasense, embecta, Research Support; A. Menarini Diagnostics, Abbott Diabetes, Roche Diabetes Care, Dexcom, Inc., Profusa, Inc., Speaker's Bureau; Novo Nordisk A/S, A. Menarini Diagnostics, Abbott Diabetes, Roche Diabetes Care, Eli Lilly and Company, Sanofi, Boehringer Ingelheim Inc., Becton, Dickinson and Company, Ypsomed AG, Viatris Inc., Servier Laboratories, Medtrust, Stock/Shareholder; Decide Clinical Software GmbH. S.A.Amiel: Advisory Panel; Medtronic, Other Relationship; Sanofi, Novo Nordisk. P.Choudhary: Advisory Panel; Medtronic, Novo Nordisk, Dexcom, Inc., MannKind Corporation, Insulet Corporation, Research Support; Abbott Diabetes, Speaker's Bureau; Sanofi, Lilly. Hypo-resolve consortium: n/a. P.Divilly: None. N.Zaremba: None. U.Soeholm: Employee; Novo Nordisk A/S. A.Kingsnorth: None. Z.Mahmoudi: Employee; Novo Nordisk. M.Gomes: Employee; Novo Nordisk A/S. B.E.De galan: Research Support; Novo Nordisk. U.Pedersen-bjergaard: Advisory Panel; Novo Nordisk A/S, Sanofi, Vertex Pharmaceuticals Incorporated. Funding Innovative Medicines Initiative 2 Joint Undertaking (777460)

Performance evaluation of Fitbit Charge 3 and actigraphy vs. polysomnography: Sensitivity, specificity, and reliability across participants and nights

Goal and aimsCompare the accuracy and reliability of sleep/wake classification between the Fitbit Charge 3 and the Micro Motionlogger actigraph when applying either the Cole-Kripke or Sadeh scoring algorithms. Accuracy was established relative to simultaneous Polysomnography recording.Focus technology: Fitbit Charge 3 and actigraphy.Reference technology: Polysomnography. SampleTwenty-one university students (10 females). DesignSimultaneous Fitbit Charge 3, actigraphy, and polysomnography were recorded over 3 nights at the participants’ homes. Core analyticsTotal sleep time, wake after sleep onset, sensitivity, specificity, positive predictive value, and negative predictive value. Additional analytics and exploratory analysesVariability of specificity and negative predictive value across subjects and across nights. Core outcomesFitbit Charge 3 and actigraphy using the Cole-Kripke or Sadeh algorithms exhibited similar sensitivity in classifying sleep segments relative to polysomnography (sensitivity of 0.95, 0.96, and 0.95, respectively). Fitbit Charge 3 was significantly more accurate in classifying wake segments (specificity of 0.69, 0.33, and 0.29, respectively). Fitbit Charge 3 also exhibited significantly higher positive predictive value than actigraphy (0.99 vs. 0.97 and 0.97, respectively) and a negative predictive value that was significantly higher only relative to the Sadeh algorithm (0.41 vs. 0.25, respectively). Important additional outcomesFitbit Charge 3 exhibited significantly lower standard deviation in specificity values across subjects and negative predictive value across nights. Core conclusionThis study demonstrates that Fitbit Charge 3 is more accurate and reliable in identifying wake segments than the examined FDA-approved Micro Motionlogger actigraphy device. The results also highlight the need to create devices that record and save raw multi-sensor data, which are necessary for developing open-source sleep or wake classification algorithms.

The Early Detection of Neurodegenerative diseases initiative: an international and multidisciplinary effort for transforming the early detection of dementia‐causing diseases

AbstractBackgroundAround 50 million people have dementia worldwide, with nearly 10 million new cases every year. Diagnosis is complex and often relies on expensive and invasive measures, with most patients accessing medical support when they already experience symptoms.MethodThe Early Detection of Neurodegenerative diseases (EDoN) initiative, spearheaded by Alzheimer’s Research UK, brings together over 60 experts from 49 universities, research projects, patient cohorts and technology providers to create machine learning models to detect the earliest stages of dementia‐causing diseases. EDoN has reviewed behavioural and physiological modalities with the strongest association with pre‐clinical disease.ResultOver 140 modalities were identified from the review and were shortlisted to create the version 1 digital toolkit. This first version includes Mezurio and Longevity smartphone apps, a Fitbit charge 4 activity tracker and Dreem 3 sleep headband. This Toolkit was further refined through patient and public involvement studies and collects 26 measures related to 7 aspects of behaviour and physiology (cognition, neural activity, physical activity, heart rate, fine motor movement, sleep, language and speech). The Toolkit is now being used to collect digital data in four international cohorts (Boston University Alzheimer’s Disease Research Center ‐ BU ADRC; The predictors of COgnitive DECline in attenders of memory clinic using digital devices ‐ CODEC‐2; Western Australia Memory Study ‐ WAMS; Healthy Brain Aging ‐ HBA), alongside prospective and retrospective clinical data, to inform the development of machine learning models.ConclusionEDoN will build models with digital markers, validating them against other biomarkers to predict dementia subtypes and individualised disease trajectories. Based on the outputs of the initial models, EDoN will go through a series of iterations of cohort engagement, modality and tool refinement, and data collection. Workstreams are underway to inform data security, privacy, ethics and open policy research, as well as considering the integration of the final EDoN Toolkit into healthcare systems globally. EDoN aims to deliver a cost‐effective, low burden and population‐wide method for early detection of dementia‐causing diseases that will benefit the public, patients, carers, researchers and clinicians, as well as the broader healthcare system and the delivery of new therapies.

A feasibility study to evaluate a purposeful walk intervention with a distance goal using a commercially available activity monitor in elderly people post total hip replacement surgery.

Total hip replacement (THR) is performed in an increasing number of individuals around the world and while improvements in pain reduction and long-term enhancement of muscle strength are well documented, the improvement in daily activity does not follow the same trend. This study aimed to determine the feasibility of a 5-week intervention where a personalised outdoor walking distance is monitored using a commercial activity monitor (Fitbit Charge 4). Data was collected on gait and activities of daily living using patient reported outcome measures. Following the completion of the intervention period, participants took part in a semi-structured interview to voice their opinion on the use of the activity monitor, their experiences, and any challenges in order to assess the feasibility of the intervention. All quantitative data were presented descriptively, using appropriate summary statistics. Interviews were analysed using thematic analysis. Five participants who had undergone total hip replacement surgery within the postoperative period of 3 to 6months were recruited from the local community. The findings suggest that the intervention was feasible and that it encouraged all participants to increase their daily activity. Therefore, it can be concluded that a follow-up effectiveness trial is warranted.

0096 Short sleep duration and poor sleep quality associated with elevated serum NfL and mild cognitive impairment in older Veterans

Abstract Introduction Ten percent of adults aged ≥ 65 and 50% of adults ≥ 85 years exhibit cognitive impairment. In addition, 50% of older adults experience poor sleep quality. Veterans are more vulnerable to sleep disturbances and potential cognitive impairment as long-term consequences of on-duty responsibilities, co-morbidities, and conditions such as post-traumatic stress disorder. Previous studies demonstrated a relationship between sleep and cognitive impairment through impaired glymphatic function and neurodegeneration. Biomarkers such as neurofilament light (NfL) are elevated in neurodegeneration and dementia. Here, we sought to investigate NfL levels in cognitive-intact older Veterans exhibiting poor sleep quality. Methods A cross-sectional study with eighteen Veterans aged 65-86 without a clinical diagnosis of dementia. Self-reported sleep quality was assessed with the Pittsburgh Sleep Quality Index (PSQI), while time in bed (TIB), and total sleep time (TST) were retrieved from Fitbit Charge. Cognitive function was evaluated with the St. Louis University Mental Status Exam (VA-SLUMS). We used an automated multiplex immunoassay system to identify dementia biomarkers using an enzyme-linked lectin assay. T-tests, one-way or two-way ANOVAs with Tukey’s post-hoc (with or without repeated measures) were used for the analysis. Results Participants were classified as good sleepers (PSQI&amp;gt;5) and poor sleepers (PSQI&amp;lt; 5). Poor sleepers spent less TIB (Fitbit 5.3h vs. 7.7h, p =0.001) and slept fewer hours (Fitbit 4.7h vs. 7.2h, p =0.001) when compared to good sleepers. Additionally, poor sleepers exhibited mild cognitive impairment (VA SLUMS 26.9, p, =0.01). Lastly, poor sleepers exhibited higher serum NfL (mean 66.7, p=0.04). Conclusion Poor sleep quality was associated with short sleep duration, mild cognitive impairment, and higher serum levels of NfL in older Veterans without a dementia diagnosis. Results may contribute to the early identification of dementia and sleep-centered treatment guidance. Further studies with more extensive and diverse samples are warranted before clinical recommendations. Support (if any) Carleara Weiss, Ph.D., MS, RN. 1K99AG079117-01 (PI), National Institute on Aging.

0748 Using daily wearable cardio fitness and sleep data to predict obesity in early adolescence

Abstract Introduction Obesity, and its co-morbidities, is a serious health risk currently facing adolescents in the US, with increased risk for adult obesity and cardiovascular health problems. Sleep is crucial for energy restorative processes and insufficient sleep influences food intake and appetite regulation. Adolescents are at risk for poor/insufficient sleep, which could, therefore, contribute to obesity risk. Given multifactorial risk factors, there is a need to identify the most salient predictors of adolescent obesity. Predicting adolescent obesity using objective Fitbit data has not previously been investigated and is the aim of the current study. Methods We used data collected with the Fitbit Charge 2 device from 2491 participants (Year 2, Mean-age=11.95 years, SD=0.65, 49.5% female) in the Adolescent Brain Cognitive Development (ABCD) Study®. We trained machine learning models (Gradient Boosted Trees and Regularized Logistic Regression) to identify the most important measures that predict obesity (&amp;gt;95 BMI percentile). In addition to sleep (timing, efficiency, duration, and regularity across the week) and cardio fitness (step counts, resting heart rate) Fitbit measures, averaged across seven days, we included socio-demographic characteristics. We ran a grid search embedded in a 2x5 fold cross-validation and we evaluated the performance based on the area under the curve-AUC and accuracy-ACC metrics. Results 15% percent of the sample were classified as having obesity. Both classification models performed acceptably high, with the LogReg having better performance (AUC-LogReg=0.76, ACC=85%) than the GBT model (AUC-GBT=0.75, ACC=0.69). Shorter sleep duration, more day-to-day variability in sleep measures, low number of steps with low variability over the week, and higher resting heart rate, with high day-to-day variability, were important predictors. Sociodemographic factors (Hispanic/Latinx ethnicity, African American race, low access to food) were also important predictors. Conclusion Results show that there are socio-economic disparities in obesity risk and that shorter sleep duration and more variable sleep patterns, as detected with a wearable, are important predictors of obesity risk. Results highlight the clinical utility of wearable measures and the importance of continuous monitoring of sleep, physical activity, and cardio fitness in adolescents. Support (if any) NIH U01DA041022

Technical characterization of the single-lead electrocardiogram signal from four different smartwatches and its clinical implications

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): Department of Cardiology, University Hospital Basel, University of Basel, Switzerland Institute for Medical Engineering and Medical Informatics, School of Life Sciences, University of Applied Sciences and Arts Northwestern Switzerland, Muttenz, Switzerland Background Smartwatches with the capability of single-lead electrocardiograms (ECGs) acquisition have been established as valuable tool for the detection of atrial fibrillation (AF). For the currently available smartwatches, however, the technical specification of the acquired ECG signal is unclear. Purpose The aim of the current study was to characterize quantitatively the raw signal of the single-lead ECG of four currently available smartwatches (Apple watch series 6, Withings ScanWatch, Samsung Galaxy Watch 4, Fitbit Charge 5) and assess their clinical applicability based on their technical features. Method A bench test was performed applying a synthetic, well-characterized signal to the electrodes of the watches (Figure: (A) Apple watch series 6, (B) Withings ScanWatch, (C) Samsung Galaxy Watch 4, (D) Fitbit Charge 5). The frequency response to this input signal was acquired and Bode plots were calculated (Figure, second column: Bode diagrams show the frequency response (upper) and the phase angles (lower)). To compare the ECG characteristics of the four watches, the synthetical response to a pre-recorded surface ECG dataset was calculated based on the frequency characteristics determined above (Figure, third column: The synthetical raw signal is shown in grey, the response of the watch in blue). Results The frequency responses of all four watches shows a bandpass filter, with an additional 50/60 Hz narrow-band filter (Figure). While the recommended high-pass cut-off at 0.5 Hz was fulfilled by all devices (1), a low-pass cut-off was observed for all devices below 70Hz (instead of 150Hz) resulting in a significantly damped signal. In detail, a damping of -6dB, corresponding to a signal amplitude reduction of 50%, was observed above 70 Hz and below 0.4 Hz for all watches. A QRS complex amplitude reduction was observed for the Apple, Withings and Samsung watch. A pronounced noise filtering is visible for the Samsung and Fitbit Smartwatch. A difference for the clinically relevant characteristic fiducial points, such as the onset of the P-wave or QRS complex, cannot be observed in these examples. However, when using the tangent method to detect the end of the T-wave to calculate the QTc value, a difference between the measures can be observed for the Withings Watch. Conclusion The technical characteristics in terms of filtering characteristics of the single-lead ECG from four commercially available smartwatches differ and must be taken into consideration when attempting to use smartwatches beyond the detection of AF.

Accuracy of real time heart rate monitoring using smartwatch in a stroke care unit

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): ISCIII Smartwatch with heart rate (HR) tracker measures may allow for identification of people with high resting HR who have a higher risk of atrial fibrillation (AF), a potent independent risk factor for embolic strokes. In this research, we propose a new protocol for collecting real time data of patients with cryptogenic stroke including HR at rest along with monitoring electrocardiogram (ECG). First, we assess the concordance of HR measurement between telemetry (TLM) and two smartwatches in acute stroke patients, as well as the determinants of measurement error. Method We included patients over 50 years of age hospitalized in our Stroke Unit who were monitored by TLM and used the Garmin Vivosmart HR® (S1) and Fitbit Charge 5® (S2) devices. Mean HR was collected at 5, 10, 15, 20, and 25 minutes. We applied the Bland-Altman method in combination with Lin's Concordance Correlation Coefficient (CCC) to assess agreement between measurements. A coefficient CCC &amp;gt; 0.9 indicates acceptable agreement. Results The experimental setup was performed using 49 stroke patients (59% women, mean age 78 ± 9.86 years, and 69% with a history of AF). Poor agreement was observed between HR measurements by SI or S2 and TLM (CCC = 0.48 and 0.57 respectively), which was conditioned by the presence of AF during the measurement (CCC = -0.17 and -0.06). There was only good concordance in the absence of AF (CCC = 0.97 and 0.92). A higher percentage of error between measurements was significantly associated with a HR &amp;gt; 90 bpm, a history of AF, or presenting AF during the measurement (p&amp;lt;0.001). Conclusions Real-time monitoring of HR at rest in each smartwatch was not accurate in acute stroke patients who had presented AF during the intake, or when HR by TLM was greater than 90. This suggests certain limitations when considering the application of processed data from this wearable technology.

Accuracy of a Smartwatch to Assess Heart Rate Monitoring and Atrial Fibrillation in Stroke Patients.

(1) Background: Consumer smartwatches may be a helpful tool to screen for atrial fibrillation (AF). However, validation studies on older stroke patients remain scarce. The aim of this pilot study from RCT NCT05565781 was to validate the resting heart rate (HR) measurement and the irregular rhythm notification (IRN) feature in stroke patients in sinus rhythm (SR) and AF. (2) Methods: Resting clinical HR measurements (every 5 min) were assessed using continuous bedside ECG monitoring (CEM) and the Fitbit Charge 5 (FC5). IRNs were gathered after at least 4 h of CEM. Lin's concordance correlation coefficient (CCC), Bland-Altman analysis, and mean absolute percentage error (MAPE) were used for agreement and accuracy assessment. (3) Results: In all, 526 individual pairs of measurements were obtained from 70 stroke patients-age 79.4 years (SD ± 10.2), 63% females, BMI 26.3 (IQ 22.2-30.5), and NIHSS score 8 (IQR 1.5-20). The agreement between the FC5 and CEM was good (CCC 0.791) when evaluating paired HR measurements in SR. Meanwhile, the FC5 provided weak agreement (CCC 0.211) and low accuracy (MAPE 16.48%) when compared to CEM recordings in AF. Regarding the accuracy of the IRN feature, analysis found a low sensitivity (34%) and high specificity (100%) for detecting AF. (4) Conclusion: The FC5 was accurate at assessing the HR during SR, but the accuracy during AF was poor. In contrast, the IRN feature was acceptable for guiding decisions regarding AF screening in stroke patients.

Gut Microbiota Composition and Functionality Are Associated With REM Sleep Duration and Continuous Glucose Levels.

Sleep disruption is associated with worse glucose metabolic control and altered gut microbiota in animal models. We aimed to evaluate the possible links among rapid eye movement (REM) sleep duration, continuous glucose levels, and gut microbiota composition. This observational, prospective, real-life, cross-sectional case-control study included 118 (60 with obesity), middle-aged (39.1-54.8 years) healthy volunteers recruited at a tertiary hospital. Glucose variability and REM sleep duration were assessed by 10-day continuous glucose monitoring (CGM) (Dexcom G6) and wrist actigraphy (Fitbit Charge 3), respectively. The coefficient of variation (CV), interquartile range (IQR), and SD of glucose variability was assessed and the percentage of time in range (% TIR), at 126-139 mg/dL (TIR2), and 140-199 mg/dL (TIR3) were calculated. Shotgun metagenomics sequencing was applied to study gut microbiota taxonomy and functionality. Increased glycemic variability (SD, CV, and IQR) was observed among subjects with obesity in parallel to increased % TIR2 and % TIR3. REM sleep duration was independently associated with % TIR3 (β = -.339; P < .001) and glucose variability (SD, β = -.350; P < .001). Microbial taxa from the Christensenellaceae family (Firmicutes phylum) were positively associated with REM sleep and negatively with CGM levels, while bacteria from Enterobacteriacea family and bacterial functions involved in iron metabolism showed opposite associations. Decreased REM sleep duration was independently associated with a worse glucose profile. The associations of species from Christensenellaceae and Enterobacteriaceae families with REM sleep duration and continuous glucose values suggest an integrated picture of metabolic health.

0788 Sleep Profile of Adolescents Following Posterior Spinal Fusion

Abstract Introduction Major musculoskeletal operations can have a prolonged impact on patients’ quality of life during the postoperative period. Sleep patterns can be affected by general anesthesia, complex and prolonged operations, and post-surgical care. Sleep disturbance can be considered a determinant of poor postsurgical outcomes. In this study, we examined the sleep profile of adolescents who underwent Posterior Spine Fusion (PSF) utilizing the FitBit Charge 4TM wearable device. Methods Patients were approached at the time of their evaluation in the orthopedic office once they had been scheduled for PSF. Patients received a Fitbit Charge 4TM at the time of enrollment. Total sleep time (TST), the duration of REM and Deep Sleep phases, and sleep efficiency were obtained from patients’ FitBit4 data and were calculated by averaging data recorded over a 2-4-day period during the week prior to surgery, during the hospital stay, one-two-three weeks, and 3 months after patients were discharged home. Patients were given a PROMIS sleep questionnaire delivered to their cellular phone using Qualtrics the week before the hospitalization, during hospitalization, and 3 months after surgery. Pain scores were obtained every 4-6 hours during the patient’s hospital stay using the self-reporting numeric scale. Results 16 patients (10 female (62%)) completed the 3 months study period. Mean age was 16±1.9 years. The average length of hospitalization was 3 days. The average TST, REM, and DEEP sleep at baseline was 387±111, 74±28, 75±25 minutes respectively. TST, REM, and DEEP sleep decreased during the acute post-operative period and gradually increased during the follow-up period going to baseline values by the 3rd post-surgery week in 72% of the cases. Perception of sleep showed mild disturbance at baseline and worsened during the hospital stay based on PROMIS scores with improvement or return to baseline by the 3rd month post-operatively in 86% of patients. Conclusion Sleep phases of adolescents are affected after a major surgery. Similar to patients of other major surgeries, these changes are short-lived and returned to baseline by the 3rd post-operative week. There was a broad range of changes among subjects during the study period. Support (if any)