Municipalities in Vermont are collecting data on Covid levels in their wastewater. But that data doesn’t appear in state or national data systems, confusing both government officials and contractors supposed to work with the data.
In February, local sewage treatment plants began announcing their participation in the US Centers for Disease Control and Prevention’s sewage monitoring program for Covid-19.
Public health officials and experts were excited to have access to a new type of Covid data as the number of PCR tests dwindled. Some said sewage monitoring, which measures the level of SARS-CoV-2 virus in a local sewage system, has the potential to be even more comprehensive than other data sources.
“This allows for a direct estimate of Covid-19 activity for anyone using a toilet that empties into a sewage treatment plant,” said Timothy Plante, assistant professor at Larner College of Medicine at the University of Vermont. , at VTDigger in April.
Months later, the CDC reports that 10 wastewater treatment plants across the state are included in its National Wastewater Surveillance System, or NWSS.
Yet on the CDC website and in reports from the Vermont Department of Health, the actual data showing viral levels and changes over time are confusing and variable. Over the past week, there was no data for the 10 program participants.
The health department had warned there could be a pause in data collection as the national program transitioned to a new contractor, BioBot, in April. The company would be responsible for testing the sewage samples and publishing the results.
But health department officials told VTDigger via email this week that now is the time for BioBot to start reporting data again, and the state is puzzled by the pause.
“Our health surveillance team also noticed this apparent discrepancy in the data,” health department spokesman Ben Truman said. “We contacted NWSS and Biobot to find out why we don’t yet have trending data for some of these sites, and why some seem to be disconnecting – such as Essex Junction, which had trending data for a few weeks and now don’t.
Several local officials at participating sites told VTDigger they had no idea their results weren’t showing up in state and national data sources. They had been receiving reports directly from BioBot for a month or more.
St. Albans, for example, had reports of BioBot dating back to May 24, according to Brian Willett, the city’s chief wastewater operator.
The ten sites participating in the scheme, according to the health department, are Bennington, Brighton, Essex Junction, Johnson, Morrisville, Newport city, St. Albans, St. Johnsbury, Troy and Jay’s combined sewershed and Winooski.
It’s unclear exactly which sites appear in CDC data, as the agency lists three sites for Chittenden County — presumably including Essex Junction and Winooski — without specifying which is which.
The 10 sites also include St. Johnsbury, which has not started collecting data as it awaits the arrival of the right type of sampler to use for the program, said project manager Jim Brimblecombe.
But of the other nine sites, only five are displayed online. Four of them have data, but not for the most recent week, and one has no data.
In response to questions from VTDigger about Vermont’s involvement, Brian Katzowitz, a CDC spokesperson, said via email that “the majority of these (sites) are submitting data. It may take several weeks for enough data to be collected to calculate the metrics displayed on COVID Data Tracker. However, state, tribal, local and territorial health departments will still be able to track COVID-19 in their communities and make public health decisions,” he wrote.
The agency did not respond to follow-up questions about which sites are listed on the CDC’s website, or why sites that had been submitting test results for more than a month were not showing up online.
Jennings Heussner, BioBot’s government and commercial director, said he was also unsure why the data was missing.
“Honestly, I’m not quite sure why exactly every site wouldn’t be included,” he said.
He said if sites are part of the program and send BioBot sample kits, they should be on the CDC’s website “at some point.”
“At the end of the day, we’re an entrepreneur,” he said. “And we do all the data generation, but we have no say in exactly how CDC uses the data once it’s in their hands.”
Under this program, local wastewater officials are responsible for collecting wastewater samples and sending them to BioBot labs via FedEx.
Most program operators reported that collecting samples did not take long. But Nate Fraser, chief operator of the Springfield sewage treatment plant, said the city passed on the chance to continue with the new contractor because the department found it too difficult to handle with its small staff.
“My staff are very spread out on a normal day,” he said by email. “Collecting samples, labeling, washing processed sample containers and packing twice a week took up time that we felt was more usefully spent in other areas of our day-to-day operations.”
Paradoxically, the only site in Vermont that does not participate in CDC data collection was also the most stable data source in the state. The City of Burlington releases data on the virus level at its three wastewater treatment plants weekly.
Nancy Stetson, senior policy and data analyst for the city, said she’s been reporting data since the fall of 2020 using a contractor who was paid with federal relief money. The city may be open to joining the CDC in the future, but for now, “we just have our own thing going on,” she said.
This slice of data has proven useful for policy makers. The state’s Department of Health and Department of Financial Regulation cited Burlington’s wastewater data in press conferences and reports.
Elaine Wang, the city manager of Winooski, said she noticed that when Burlington reported an increase in sewage data, cases in Winooski tended to increase. Wang began using Burlington’s sewage reports to spur the city council to discuss mask mandates or other actions.
Timothy LaPara, a professor of civil engineering at the University of Minnesota, was one of the first researchers to study the ability to detect SARS-CoV-2 levels in wastewater.
LaPara said sewage levels aren’t a great way to pinpoint the exact number of cases in your community, but they work well as a relative measure of whether the virus is trending up or down.
In fact, he thinks sewage data is “the best way to track this disease.”
“All other test methods have catastrophic flaws,” LaPara said. Most people don’t get a PCR test when they’re asymptomatic, and antigen testing is rarely reported to state agencies.
“There might be issues with it that someone might find somewhere, but it’s trustworthy. It’s the best we have,” he said.
Sign up for our guide to the global coronavirus outbreak and its impact on Vermont, with the latest developments delivered to your inbox.