"At a time when unemployment is skyrocketing, we're trying to figure out how to make basic ends meet," said New Castle County Executive Matt Meyer. "So one of the things we're doing is looking everywhere we can possibly look to get more information about this invisible enemy."
One of those places, according to Meyer, ended up being poop.
After New Castle County Public Works Stormwater and Environmental Program Manager Mike Harris teamed up with Massachusetts Institute of Technology's startup Biobot Analytics, officials studied samples of waste matter at a pretreatment center for the sewer system. According to estimations provided following bacterial testing at the plant, the county appeared to have a staggering number of novel coronavirus COVID-19 cases--15,200 in a catchment area that serves about 100,000 individuals in an area of the county.
"I want to emphasize it is an imprecise estimate, but their imprecise estimate is that there are approximately 15 times the number of positive cases as there are confirmed, positive tests, Meyer said during a Zoom call with reporters Thursday. "Again, emphasize this is a new technology, it's unconfirmed, but they have been doing it in various areas throughout the country, and our result is somewhat comparable...to the other areas around the country."
Harris said wastewater helps circumvent privacy issues and incomplete data samples, since everyone poops and the waste produced isn't identifiable, so they can test a large area very quickly for gaining a better handle on the situation as it stands in New Castle County.
"Though preliminary, though an estimate, it is very helpful to us," said Meyer. "First of all, it's very helpful as we try to determine where we are on the curve...Where we are on the slope? Have we flattened quicker, or is the curve decreasing? I've asked [Harris] to speak to Biobot about taking samples from multiple treatment plants across the county, which, potentially, again, could help us decipher where some hotspots are without needing to do any tests. So, for example, when you have a limited testing potential, you can say in this particular area...there's a higher percentage of viral bacteria, so we need to do more intensive testing of asymptomatic people in this particular area."
While the data points received from an unproven method might have a degree of uncertainty built in, Meyer hopes as they pile up, they start to show a clearer picture for those studying the spread of COVID-19.
"Some of the information we're getting, I'm quite certain that as time goes by, we will start to be 'partially inaccurate' more so than 'completely inaccurate.' So I just consider this another data point as part of a larger story and once you get a lot of data points that it helps to draw conclusions over time, including determining how effective and accurate these numbers are. So we're not going to make any grand decisions like asking the governor to open our borders up for, you know, months or weeks or days, based on this information but it's a data point."
The results from the first sample, collected as of April 15, 2020, are included here. A second, third, and fourth data set with additional samples will be furnished over the next few weeks as the county looks to establish a reliable way to track COVID-19 in the community.
The numbers break down to roughly 3% of the county's population north of the C&D Canal, according to officials--an estimation 15 times greater than currently reported confirmed positive tests provided by the Department of Public Health.
There was no cost to the county to enter into the partnership with Biobot, Meyer said, and the only out-of-pocket expenditures incurred came from shipping costs related to sending samples to the MIT lab.