Health Advice



A community's fecal matter could give early warning of COVID-19 outbreaks, Yale study finds

Emily Brindley, The Hartford Courant on

Published in Health & Fitness

By studying sewage at a New Haven wastewater treatment facility, a team of Yale researchers has determined that genetic code embedded in feces could be used as an early warning sign of COVID-19 outbreaks.

The team, led by Jordan Peccia of the Yale School of Engineering and Applied Science, tested daily samples of sludge for bits of coronavirus code, known as RNA. They then found that they could use just their own data to recreate the curve of COVID-19 cases in the New Haven area.

"Except we see it seven days earlier," Peccia said.

The study, which was posted online Friday but has not yet been peer-reviewed, has implications for Connecticut's coronavirus response. For the first two months of the pandemic, the state struggled to increase testing capacity. Even now, after a major testing ramp-up, clinical testing focuses on those who have symptoms.

That means patients often aren't tested until they begin showing symptoms, or may not be tested at all if they remain asymptomatic. In the meantime, they could be silently spreading the virus.

But sewage keeps a record of all cases.


"Before you're symptomatic and after you're infected, you can certainly shed that virus and be infectious," Peccia said. "As soon as you start shedding it, whether you feel it or not, we're gonna see it in the wastewater."

For their study, Peccia and his team collected daily sewage samples from the East Shore Water Pollution Abatement Facility in New Haven, from mid-March until May 1. They tested each sample's concentration of coronavirus RNA and then compared those daily concentrations to actual data on COVID-19 cases and hospitalizations in the towns served by the water treatment facility.

They found that the concentration of coronavirus RNA increased and decreased several days before corresponding fluctuations in actual COVID-19 cases and hospitalizations, as reported by the local hospital and the state.

According to the study, the sludge samples predicted hospitalization fluctuations three days before they occurred, and testing data fluctuations seven days before they occurred.


swipe to next page