Duke says testing program prevented campus COVID-19 outbreaks
Duke University's aggressive pooled surveillance COVID-19 testing program enabled large-scale testing, successfully reduced transmission, and prevented major outbreaks seen on other college campuses, according to a Morbidity and Mortality Weekly Report (MMWR) study yesterday.
The Duke campaign relied on a combination of strategies, including a 14-day pre-arrival self-quarantine for all enrolled students and a code-of-conduct pledge to wear masks, follow physical distancing guidelines, and undergo regular COVID-19 testing. Daily symptom monitoring—carried out via a custom smartphone app—was accompanied by contact tracing and quarantine. On-campus students were tested twice a week with self-swab kits collected at strategically located sites on campus. Off-campus students were tested once a week.
The surveillance testing used pooled samples to conserve resources, and the researchers reported positive results with testing samples in batches of five and re-testing individual samples within batches that showed a positive result. The batch method allowed the Duke Human Vaccine Institute to process 80,000 samples from Aug 2 to Oct 11, including 68,913 specimens from 10,265 graduate and undergraduate students—excluding 781 student-athletes who participated in a separate surveillance program.
The Duke surveillance approach resulted in a lower average per-capita infection prevalence among students (0.08%) than in the surrounding community (Durham County, 0.1%), and no large campus outbreaks. There were 84 COVID-19 cases among students, with 51% of the cases occurring among asymptomatic people, highlighting the importance of comprehensive versus symptom-based testing.
"By late summer there were still things we didn't fully understand about SARS-CoV-2 transmission, so there was some uncertainty going into the fall semester," said Steve Haase, PhD, of Duke University School of Medicine in a Duke Health news release yesterday. "Over the course of the semester we've learned many things, including that it is possible to limit the spread of the virus and create a safer environment for our students to have that invaluable on-campus learning experience."
Nov 17 MMWR study
Nov 17 Duke Health news release
Smartwatch data may help identify pre-symptomatic COVID-19
A study in Nature Biomedical Engineering today shows that smartwatches and other wearable devices may detect pre-symptomatic COVID-19 infection and allow for early-stage interventions that reduce transmission.
Among infected smartwatch users, 81% showed alterations in their heart rate, number of daily steps, or time asleep. Changes before symptom onset identified 63% of the COVID-19–positive individuals, showing that consumer wearable device data can recognize pre-symptomatic infection.
Stanford University researchers analyzed data from 4,642 smartwatch users, focusing on 32 COVID-19–positive participants with Fitbit data that captured resting heart rate (RHR), steps per day, and sleep duration.
Of individuals with a symptom-onset date or a diagnosis date, 88% (22 out of 25) and 100% (25 out of 25) showed a 7-beats-per-minute median elevation in RHR relative to baseline both in advance of and at the time of symptom onset or diagnosis. Elevated signals were detected several days ahead of symptom onset and diagnosis—4 days and 7 days, respectively—highlighting the potential for early identification of infection and mitigation of community spread.
Steps per day significantly decreased and sleep duration increased after the onset of elevated heart rate signals, but no association was found between the magnitude of the increased heart rate signal and specific symptoms, illness length, or temperature.
The researchers also developed an online detection method using smartwatch data to explore the potential detection of early stage COVID-19 illness in advance of symptoms, finding that they were able to detect 63% of known COVID-19 infections.
"Our findings suggest that activity tracking and health monitoring via consumer wearable devices may be used for the large-scale, real-time detection of respiratory infections, often pre-symptomatically," the authors concluded.
Nov 18 Nat Biomed Eng study