Traditional screening for COVID-19 typically includes survey questions about symptoms and travel history, as well as temperature measurements. Here, we explore whether personal sensor data collected over time may help identify subtle changes indicating an infection, such as in patients with COVID-19. We have developed a smartphone app that collects smartwatch and activity tracker data, as well as self-reported symptoms and diagnostic testing results, from individuals in the United States, and have assessed whether symptom and sensor data can differentiate COVID-19 positive versus negative cases in symptomatic individuals. We enrolled 30,529 participants between 25 March and 7 June 2020, of whom 3,811 reported symptoms. Of these symptomatic individuals, 54 reported testing positive and 279 negative for COVID-19. We found that a combination of symptom and sensor data resulted in an area under the curve (AUC) of 0.80 (interquartile range (IQR): 0.73-0.86) for discriminating between symptomatic individuals who were positive or negative for COVID-19, a performance that is significantly better (P < 0.01) than a model 1 that considers symptoms alone (AUC = 0.71; IQR: 0.63-0.79). Such continuous, passively captured data may be complementary to virus testing, which is generally a one-off or infrequent sampling assay. Owing to the current lack of fast and reliable testing, one of the greatest challenges for preventing transmission of SARS-CoV-2 is the ability to quickly identify, trace and isolate cases before they can further spread the infection to susceptible individuals. As regions across the United States start implementing measures to reopen businesses, schools and other activities, many rely on current screening practices for COVID-19, which typically include a combination of symptom and travel-related survey questions and temperature measurements. However, this method is likely to miss pre-symptomatic or asymptomatic cases, which make up ~40-45% of those infected with SARS-CoV-2, and who can still be infectious 1,2. An elevated temperature (>100 °F (>37.8 °C)) is not as common as frequently believed, being present in only 12% of individuals who tested positive for COVID-19 3 and just 31% of patients hospitalized with COVID-19 (at the time of admission) 4. Smartwatches and activity trackers, which are now worn by one in five Americans 5 , can improve our ability to objectively characterize each individual's unique baseline for resting heart rate 6 , sleep 7 and activity and can therefore be used to identify subtle changes in that user's data that may indicate that they are coming down with a viral illness. Previous research from our group has shown that this method, when aggregated at the population level, can significantly improve real-time predictions for influenza-like illness 8. Consequently, we created a prospective app-based research platform, called DETECT (Digital Engagement and Tracking for Early Control and Treatment), where individuals can share their sensor data, self-reported symptoms, diagnoses and ele...

BackgroundNew models of scalable population-based data collection that integrate digital and mobile health (mHealth) data are necessary.ObjectiveThe aim of this study was to describe a cardiovascular digital and mHealth electronic cohort (e-cohort) embedded in a traditional longitudinal cohort study, the Framingham Heart Study (FHS).MethodsWe invited eligible and consenting FHS Generation 3 and Omni participants to download the electronic Framingham Heart Study (eFHS) app onto their mobile phones and co-deployed a digital blood pressure (BP) cuff. Thereafter, participants were also offered a smartwatch (Apple Watch). Participants are invited to complete surveys through the eFHS app, to perform weekly BP measurements, and to wear the smartwatch daily.ResultsUp to July 2017, we enrolled 790 eFHS participants, representing 76% (790/1044) of potentially eligible FHS participants. eFHS participants were, on average, 53±8 years of age and 57% were women. A total of 85% (675/790) of eFHS participants completed all of the baseline survey and 59% (470/790) completed the 3-month survey. A total of 42% (241/573) and 76% (306/405) of eFHS participants adhered to weekly digital BP and heart rate (HR) uploads, respectively, over 12 weeks.ConclusionsWe have designed an e-cohort focused on identifying novel cardiovascular disease risk factors using a new smartphone app, a digital BP cuff, and a smartwatch. Despite minimal training and support, preliminary findings over a 3-month follow-up period show that uptake is high and adherence to periodic app-based surveys, weekly digital BP assessments, and smartwatch HR measures is acceptable.

Individual smartwatch or fitness band sensor data in the setting of COVID-19 has shown promise to identify symptomatic and pre-symptomatic infection or the need for hospitalization, correlations between peripheral temperature and self-reported fever, and an association between changes in heart-rate-variability and infection. In our study, a total of 38,911 individuals (61% female, 15% over 65) have been enrolled between March 25, 2020 and April 3, 2021, with 1118 reported testing positive and 7032 negative for COVID-19 by nasopharyngeal PCR swab test. We propose an explainable gradient boosting prediction model based on decision trees for the detection of COVID-19 infection that can adapt to the absence of self-reported symptoms and to the available sensor data, and that can explain the importance of each feature and the post-test-behavior for the individuals. We tested it in a cohort of symptomatic individuals who exhibited an AUC of 0.83 [0.81–0.85], or AUC = 0.78 [0.75–0.80] when considering only data before the test date, outperforming state-of-the-art algorithm in these conditions. The analysis of all individuals (including asymptomatic and pre-symptomatic) when self-reported symptoms were excluded provided an AUC of 0.78 [0.76–0.79], or AUC of 0.70 [0.69–0.72] when considering only data before the test date. Extending the use of predictive algorithms for detection of COVID-19 infection based only on passively monitored data from any device, we showed that it is possible to scale up this platform and apply the algorithm in other settings where self-reported symptoms can not be collected.

No abstract

Rationale: A sedentary lifestyle is associated with increased risk for cardiovascular disease (CVD). Smartwatches enable accurate daily activity monitoring for physical activity measurement and intervention. Few studies, however, have examined physical activity measures from smartwatches in relation to traditional risk factors associated with future risk for CVD. Objective: To investigate the association of habitual physical activity measured by smartwatch with predicted CVD risk in adults. Methods and Results: We enrolled consenting Framingham Heart Study (FHS) participants in an ongoing electronic Framingham Heart Study (eFHS) at the time of their FHS research center examination. We provided participants with a smartwatch (Apple Watch Series 0) and instructed them to wear it daily, which measured their habitual physical activity as the average daily step count. We estimated the 10-year predicted risk of CVD using the ACC/AHA 2013 pooled cohort risk equation. We estimated the association between physical activity and predicted risk of CVD using linear mixed effects models adjusting for age, sex, wear time, and familial structure. Our study included 903 eFHS participants (mean age 53{plus minus}9 years, 61% women, 9% non-white) who wore the smartwatch greater than 5 hours per day for more than 30 days. Median daily step count was similar among men (7,202 with interquartile range [IQR] 3,619) and women (7,260 with IQR 3,068, P=0.52). Average 10-year predicted CVD risk was 4.5% (IQR 6.1%) for men and 1.2% (IQR 2.2%) for women (P=1.3x10-26). Every 1,000 steps higher habitual physical activity was associated with 0.18% lower predicted CVD risk (P=3.2x10-4). The association was attenuated but remained significant after further adjustment for body mass index (P=0.01). Conclusions: In this community-based sample of adults, higher daily physical activity measured by a study smartwatch was associated with lower predicted risk of CVD. Future research should examine the longitudinal association of prospectively measured daily activity and incident CVD.