The COVID-19 pandemic has led more people to start using wearable technology to track vital signs, physical activity, and sleep. The significant features of these devices include their capability to collect continuous, noninvasive data. We developed a COVID-19 risk stratification model using the Biostrap wearable device which utilizes a baseline-adjusted continuous scale and other escalation points-based on our recent case report, to enhance the National Early Warning Score (NEWS2). Preliminary research has found that our adjusted Early Warning Score (Biostrap-EWS) might be highly specific in identifying early-stage respiratory infections. We present the case of Biostrap CEO Sameer Sontakey, a 35-year-old man, whom the app notified as having a high likelihood of respiratory illness after which the diagnosis SARS-CoV-2 was confirmed with a nasal swab. Our Biostrap-EWS algorithm appears to detect respiratory infections in a real-world environment via passively collected biometric data.
With the Biostrap wrist-worn device (Biostrap USA LLC, Duarte, CA, USA), a commercially available, clinically validated wearable device that uses photoplethysmography to automatically record physiological data such as resting heart rate, respiratory rate, oxygen saturation (SpO2), and arterial stiffness (AS), we collected biometric data from 933 subjects. We present two cases of patients who have tested positive for the presence of severe acute respiratory syndrome (SARS-CoV-2), a 24-year-old man experiencing major symptoms and another a 49-year-old man with only intermittent fatigue, and show the marked changes in biometric measurements around dates of symptom onset and positive test. We observed a pattern of sustained respiratory rate elevation in both patients, punctuated by a sharp spike in heart rate and decreased AS. The latter contradicted our expectation that during the onset of symptoms of COVID-19, an increase in AS might occur.
