scholarly journals High-Resolution Digital Phenotypes from Consumer Wearables Enhance Prediction of Cardiometabolic Risk Markers

2021 ◽  
Author(s):  
Weizhuang Zhou ◽  
Yu En Chan ◽  
Chuan Sheng Foo ◽  
Jingxian Zhang ◽  
Jing Xian Teo ◽  
...  
Keyword(s):  
Heart Rate ◽  
High Resolution ◽  
Disease Risk ◽  
Cardiometabolic Disease ◽  
Risk Markers ◽  
Free Living ◽  
Heart Rate Dynamics ◽  
Health And Disease ◽  
Free Living Conditions ◽  
Cardiometabolic Disease Risk

Background: Consumer-grade wearable devices enable detailed recordings of heart rate and step counts in free-living conditions. Recent studies have shown that summary statistics from these wearable recordings have potential uses for longitudinal monitoring of health and disease states. However, the relationship between higher resolution physiological dynamics from wearables and known markers of health and disease remains largely uncharacterized. Objective: We aimed to (i) derive high resolution digital phenotypes from observational wearable recordings, (ii) characterize their ability to predict modifiable markers of cardiometabolic disease, and (iii) study their connections with genetic predispositions for cardiometabolic disease and with lifestyle factors. Methods: We introduce a principled framework to extract interpretable high resolution phenotypes from wearable data recorded in free-living conditions. The proposed framework standardizes handling of data irregularities, encodes contextual information about underlying physiological state at any given time, and generates a set of 66 minimally redundant features across active, sedentary and sleep states. We applied our approach on a multimodal dataset, from the SingHEART study (NCT02791152), that comprises of heart rate and step count time series from wearables, clinical screening profiles, whole genome sequences and lifestyle survey responses from 692 healthy volunteers. We employed machine learning to model non-linear relationships between the high resolution phenotypes and clinical risk markers for blood pressure, lipid and weight abnormalities. For each risk type, we performed model comparisons based on Brier Skill Scores (BSS) to assess predictive value of the high resolution features over and beyond typical baselines. We then examined associations between the wearable-derived features, polygenic risk for cardiometabolic disease, and lifestyle habits and health perceptions. Results: Compared to typical summary statistic measures like resting heart rate, we find that the high-resolution features collectively have greater predictive value for modifiable clinical markers associated with cardiometabolic disease risk (average improvement in Brier Skill Score=52.3%, P<.001). Further, we show that heart rate dynamics from different activity states contain distinct information about type of cardiometabolic risk, with dynamics in sedentary states being most predictive of lipid abnormalities and patterns in active states being most predictive of blood pressure abnormalities (P<.001). Finally, our results reveal that subtle heart rate dynamics in wearable recordings serve as physiological correlates of genetic predisposition for cardiometabolic disease, lifestyle habits and health perceptions. Conclusions: High resolution digital phenotypes recorded by consumer wearables in free-living states have the potential to enhance prediction of cardiometabolic disease risk, and could enable more proactive and personalized health management. Clinical Trial Registration ID #NCT02791152. Keywords: Wearable device, heart rate, cardiometabolic disease, risk prediction, digital phenotypes, polygenic risk scores, time series analysis, machine learning, free-living

scholarly journals Effect of an office-based intervention on visceral adipose tissue: the WorkACTIVE-P randomized controlled trial

2021 ◽  
Vol 46 (2) ◽  
pp. 117-125
Author(s):  
James L. Dorling ◽  
Christoph Höchsmann ◽  
Catrine Tudor-Locke ◽  
Robbie Beyl ◽  
Corby K. Martin
Keyword(s):  
Adipose Tissue ◽  
Food Intake ◽  
Visceral Adipose Tissue ◽  
Disease Risk ◽  
Office Workers ◽  
Cardiometabolic Disease ◽  
Risk Markers ◽  
Randomized Controlled ◽  
Visceral Adipose ◽  
Cardiometabolic Disease Risk

Office-based activity reduces sedentariness, yet no randomized controlled trials (RCTs) have assessed how such activity influences visceral adipose tissue (VAT). This study examined the effect of an office-based, multicomponent activity intervention on VAT. The WorkACTIVE-P RCT enrolled sedentary office workers (body mass index: 31.4 (standard deviation (SD) 4.4) kg/m2) to an intervention (n = 20) or control (n = 20) group. For 3 months, the intervention group received an office-based pedal desk, further to an intervention promoting its use and increased walking. The control group maintained habitual activity. At baseline and follow-up, VAT, cardiometabolic disease risk markers, physical activity, and food intake were measured. Steps/day were not altered relative to control (P ≥ 0.51), but the pedal desk was utilized for 127 (SD 61) min/day. The intervention reduced VAT relative to control (−0.15 kg; 95% confidence interval (CI) = −0.29 to −0.01; P = 0.04). Moreover, the intervention decreased fasting glucose compared with control (−0.29 mmol/L; 95% CI = −0.51 to −0.06; P = 0.01), but no differences in other cardiometabolic disease markers or food intake were revealed (P ≥ 0.11). A multicomponent intervention decreased VAT in office workers who were overweight or obese. Though longer-term studies are needed, office-based, multicomponent activity regimens may lower cardiometabolic disease risk. Trial registered at ClinicalTrials.gov (NCT02561611). Novelty: In WorkACTIVE-P, a multicomponent activity intervention decreased visceral adipose tissue relative to control in office workers. The intervention also reduced glucose compared with control, though other metabolic risk markers and food intake were not altered. Such multicomponent interventions could help reduce cardiometabolic disease risk, but longer studies are needed.

scholarly journals Distinct phenotypic characteristics of normal-weight adults at risk of developing cardiovascular and metabolic diseases

2020 ◽  
Vol 112 (4) ◽  
pp. 967-978
Author(s):  
Abishek Stanley ◽  
John Schuna ◽  
Shengping Yang ◽  
Samantha Kennedy ◽  
Moonseong Heo ◽  
...  
Keyword(s):  
Health Risk ◽  
Body Shape ◽  
Metabolic Diseases ◽  
Disease Risk ◽  
White Men ◽  
Normal Weight ◽  
Cardiometabolic Disease ◽  
Men And Women ◽  
Distinct Phenotype ◽  
Cardiometabolic Disease Risk

ABSTRACT Background The normal-weight BMI range (18.5–24.9 kg/m2) includes adults with body shape and cardiometabolic disease risk features of excess adiposity, although a distinct phenotype developed on a large and diverse sample is lacking. Objective To identify demographic, behavioral, body composition, and health-risk biomarker characteristics of people in the normal-weight BMI range who are at increased risk of developing cardiovascular and metabolic diseases based on body shape. Methods Six nationally representative waist circumference index (WCI, weight/height0.5) prediction formulas, with BMI and age as covariates, were developed using data from 17,359 non-Hispanic (NH) white, NH black, and Mexican-American NHANES 1999–2006 participants. These equations were then used to predict WCI in 5594 NHANES participants whose BMI was within the normal weight range. Men and women in each race/Hispanic-origin group were then separated into high, medium, and low tertiles based on the difference (residual) between measured and predicted WCI. Characteristics were compared across tertiles; P values for significance were adjusted for multiple comparisons. Results Men and women in the high WCI residual tertile, relative to their BMI and age-equivalent counterparts in the low tertile, had significantly lower activity levels; higher percent trunk and total body fat (e.g. NH white men, X ± SE, 25.3 ± 0.2% compared with 20.4 ± 0.2%); lower percent appendicular lean mass (skeletal muscle) and bone mineral content; and higher plasma insulin and triglycerides, higher homeostatic model assessment of insulin resistance (e.g. NH white men, 1.45 ± 0.07 compared with 1.08 ± 0.06), and lower plasma HDL cholesterol. Percent leg fat was also significantly higher in men but lower in women. Similar patterns of variable statistical significance were present within sex and race/ethnic groups. Conclusions Cardiometabolic disease risk related to body shape in people who are normal weight according to BMI is characterized by a distinct phenotype that includes potentially modifiable behavioral health risk factors.

scholarly journals Critical evaluation of the factorial and heart-rate recording methods for the determination of energy expenditure of free-living elderly people

1997 ◽  
Vol 78 (5) ◽  
pp. 709-722 ◽  
Author(s):  
Beatrice Morio ◽  
Patrick Ritz ◽  
Elisabeth Verdier ◽  
Christophe Montaurier ◽  
Bernard Beaufrere ◽  
...  
Keyword(s):  
Heart Rate ◽  
Energy Expenditure ◽  
Elderly People ◽  
Living Conditions ◽  
Energy Costs ◽  
Free Living ◽  
Factorial Method ◽  
Recording Method ◽  
Mean Differences ◽  
Free Living Conditions

The aim of the present study was to validate against the doubly-labelled water (DLW) technique the factorial method and the heart rate (HR) recording method for determining daily energy expenditure (DEE) of elderly people in free-living conditions. The two methods were first calibrated and validated in twelve healthy subjects (six males and six females; 70·1 (sd 2·7) years) from opencircuit whole-body indirect calorimetry measurements during three consecutive days and during 1 d respectively. Mean energy costs of the various usual activities were determined for each subject using the factorial method, and individual relationships were set up between HR and energy expenditure for the HR recording method. In free-living conditions, DEE was determined over the same period of time by the DLW, the factorial and the HR recording methods during 17, 14 and 4 d respectively. Mean free-living DEE values for men estimated using the DLW, the factorial and the HR recording methods were 12·8 (sd 3·1), 12·7 (sd 2·2) and 13·5 (sd 2·7) MJ/d respectively. Mean free-living DEE values for women were 9·6 (sd 0·8), 8·8 (sd 1·2) and 10·2 (sd 1·5) MJ/d respectively. No significant differences were found between the three methods for either sex, using the Bland & Altman (1986) test. Mean differences in DEE of men were -0·9 (sd 11·8) % between the factorial and DLW methods, and +4·7 (sd 16·1) % between the HR recording and DLW methods. Similarly, in women, mean differences were -7·7 (sd 12·7) % between the factorial and DLW methods, and +5·9 (sd 8·8) % between the HR recording and DLW methods. It was concluded that the factorial and the HR recording methods are satisfactory alternatives to the DLW method when considering the mean DEE of a group of subjects. Furthermore, mean energy costs of activities calculated in the present study using the factorial method were shown to be suitable for determining free-living DEE of elderly people when the reference value (i.e. sleeping metabolic rate) is accurately measured.

Sign in / Sign up

Export Citation Format

Share Document