Health Care

What’s in Vermont’s 1st Covid ‘surveillance report’ — and what’s not

Note: This story is more than a week old. Given how quickly the Covid-19 pandemic is evolving, we recommend that you read our latest coverage here.

Mark Levine at podium
Dr. Mark Levine, commissioner of the Vermont Department of Health, speaks at a Covid-19 press conference in 2020. File photo by Mike Dougherty/VTDigger

On Tuesday, Vermont Health Commissioner Mark Levine announced the end of the state’s Covid-19 case dashboard, the central repository for the latest information on cases, hospitalizations and deaths.

The last dashboard update is scheduled for May 18. Instead, the state has launched a weekly surveillance report that will “look at this disease in a way that’s similar to how we look at other infectious diseases, such as the flu,” Levine said.

The Vermont Department of Health issued the first edition of this new report on Wednesday, including data from May 1 to May 7. 

The update contains some of the same information as the dashboard, such as weekly case totals. It also contains new information on topics such as syndromic surveillance — a measure of emergency department visits for Covid-like illness — and wastewater monitoring

It skips other data points that are published in the dashboard, including daily case counts, PCR tests conducted and the number of patients in Vermont hospitals and intensive care units with Covid.

Levine said the changes were designed to “focus on the larger picture of Covid-19’s impact on our health as a population and health care systems.”

“The widespread use of at-home tests has already made case counts less meaningful, in addition to data such as percent positivity,” he said. “Nor do the number of cases reflect the amount of severe disease in our communities, which is our main concern when it comes to Covid-19.”

Experts have said that the rise of at-home antigen testing makes it more difficult to draw conclusions from Covid case data, which is mostly based on PCR tests. 

But Anne Sosin, a health equity researcher at Dartmouth College, called the decision to take away the dashboard data “premature” and a step backward.

“If we are going to rely on individuals to make public health decisions, then we need to equip them with more, not less, meaningful data,” she said.

The latest report includes data, such as syndromic surveillance, geared toward people with formal epidemiologic training, Sosin said. “Most people don’t have the skills” to interpret that kind of information.

“Many who are looking closely at the data are those in our communities who are at highest risk. It's those people who are most concerned, and so they need real tools and information to be able to protect themselves,” she said.

Vermonters have to look at several different sources to get a full picture of what’s happening right now, she said. Along with the dashboard and the new weekly report, the state publishes a separate page of data on “community settings,” and the U.S. Centers for Disease Control and Prevention publishes its own data on community Covid levels.

The Vermont Department of Financial Regulation has also ceased publishing its weekly modeling updates, spokesperson Stephanie Brackin told VTDigger. The last update was on May 3. The department recently lost former commissioner Michael Pieciak, who resigned to run for state treasurer, and two data analysts who worked on the updates.

Although Sosin agrees Covid case data is likely underreported now, she said it would be better to report that data along with a clear explanation of what’s not captured.

“Just because data has limitations doesn't mean that it's totally useless,” she said. “We can observe trends in the cases that we're seeing and then we can make our best guesses about what that means and doesn't mean.”

Timothy Plante, an assistant professor at the University of Vermont’s Larner College of Medicine, said it may be useful to have a report that looks at “higher-level trends” rather than potential daily blips in the data.

However, he cautioned that the data in the new report may be 10 or more days behind by the time it’s published, and there’s a potential to “miss clear trends.”

“It was clear weeks ago that BA.2 was coming and we were at the start of the next wave, and we could see that on the daily dashboard,” Plante said. “And my own personal behaviors changed based upon what I was seeing in the day-to-day changes on the dashboard.”

He said the surveillance report may have some “growing pains” in the first edition, suggesting that this may not be the final version of what the department publishes going forward.

Here’s what the latest surveillance update contains, and what the latest dashboard update shows for comparison.

Community levels

The report begins with an overall picture of Vermont’s Covid community levels: High, for the week of May 1 to May 7.

That’s based on metrics similar to the ones the CDC uses to create its county-by-county community levels: The daily case average, recent hospital admissions and the percentage of hospital beds occupied by Covid patients. 

But the department’s conclusions diverge from the CDC recommendations. The CDC recommends that all people in high-level counties take action to prevent Covid transmission, such as wearing a mask. 

Vermont, in contrast, says that “each person can decide if they want to take precautions,” including masking, based on their personal risk level.

Syndromic surveillance

The report contains a chart showing the relative level of people coming to Vermont emergency departments with Covid-like illnesses.

It shows that about 5% to 7% of emergency visits are due to Covid-like illness, or people with symptoms that match the symptoms of Covid. That’s far higher than the rate of about 3% in mid-March.

Plante said that tracking emergency visits involving Covid symptoms “helps us get a sort of an indirect insight into” the level of viral activity in the community.

But he said it can sometimes be difficult to parse out Covid symptoms from other, similar respiratory diseases, such as the flu, upper respiratory illnesses or even allergies. “Covid is not the only circulating respiratory virus,” he said.

Proportion of circulating variants 

The report contains a chart detailing the proportion of different Covid variants in the state, including the subvariant BA.2, which made up 92% of samples as of April 10.

Plante said the data might be used to infer where Vermont is in the “next wave” as new variants circulate worldwide. 

“When we see identified variants popping up in South Africa or in European countries, and once we start detecting those in our own community, we might be able to predict the likelihood of a surge, and maybe even the length of a surge, based upon our knowledge of how these other variants have circulated in other communities,” he said.

At this point, the data is about a month behind. The CDC has more recent data, but it is broken down only to the New England region, not state by state. The New England data shows an increasing number of samples with the BA.2.12.1 subvariant.

Wastewater monitoring

The report has some details on how sewage treatment plants around the state are reporting the level of virus in their community.

Eleven communities in Vermont test their wastewater for SARS-CoV-2, the virus that causes Covid. Burlington reports data for three different treatment plants and has shown an elevated level of the virus in recent weeks.

The 10 other locations did not release data in this latest report because of a recent change in their sampling contractor, according to the report. When they do start to release data, it would show the percentage change in viral levels over a 15-day period.

Plante said wastewater data could be one of the “key measures moving forward,” particularly for detecting an increase in transmission at early stages. 

In contrast to the shrinking number of people getting tested for Covid, “you are capturing direct viral samples from a huge swath of people: everyone who uses the toilet that drains to that sewage area,” he said.

On the other hand, the measurements are not very precise, he said. “Getting multiple measures from multiple different areas and multiple different wastewater treatment plants will help with that,” he said.

Reported and confirmed outbreaks

The report shows the number of active outbreaks at different types of congregate settings and breaks the data down by county.

As of May 10, there were 22 active outbreaks at long-term care facilities, by far the largest category, the report shows. A different section of the health department website contains a list of the exact facilities and their number of cases.

The outgoing dashboard does not include any outbreak data, but it does contain the number of cases and the case rate by county. That information is not included in the surveillance report.

Vaccination rates 

The report has a breakdown of the percentage of Vermonters 5 and older who are up to date on their vaccines, including all recommended boosters, by race and age. (A second booster is recommended only for people 50 and older, or immunocompromised people.)

In total, 59% of eligible Vermonters are up-to-date on their vaccines, the report shows. A separate section of the health department website shows that 88% of eligible Vermonters have received at least one dose of the vaccine.

Identified cases 

At the end of the report is a chart showing that the state had 2,340 cases of Covid in the week of May 1. It also breaks down that number by age group, and compares people of color versus non-Hispanic white Vermonters.

The outgoing dashboard has more detailed data, including the exact number of cases for different racial groups, along with the week-by-week rate of Covid per 100,000 people for different age groups.

The dashboard also has the number of cases, hospitalizations and deaths by vaccination status, and the number of deaths in different age and racial groups. That information is not included in the new report.

What the dashboard shows now

According to the health department dashboard, the state reported 261 Covid cases, 59 hospitalizations and no additional deaths on Wednesday.

VTDigger uses the underlying data to compare metrics over time and calculate the seven-day case average. As of Wednesday, the average was 299 cases per day, down from 335 the day before.

The dashboard includes the percentage positivity rate of tests, but only including PCR tests, not antigen tests that were reported to the health department. On Wednesday, the seven-day average of the positivity rate was 14.4%, compared to 14.5% the day before.

As of Wednesday, 59 people were hospitalized with Covid, including six in intensive care, according to the dashboard. The underlying data shows that’s down from 75 hospitalizations and 10 people in intensive care on Tuesday.

The dashboard has a chart that shows the number of deaths over time. No additional deaths were reported on Wednesday. In total, 648 people have died over the course of the pandemic, including eight in May.

The Vermont Center of Geographic Information will continue to host the same underlying data after the dashboard ends, Levine said Tuesday.

Sign up for our guide to the global coronavirus outbreak and its impact on Vermont, with latest developments delivered to your inbox.

 

Did you know VTDigger is a nonprofit?

Our journalism is made possible by member donations from readers like you. If you value what we do, please contribute during our annual fund drive and send 10 meals to the Vermont Foodbank when you do.

Erin Petenko

Send us your thoughts

VTDigger is now accepting letters to the editor. For information about our guidelines, and access to the letter form, please click here.

 

Recent Stories

Thanks for reporting an error with the story, "What’s in Vermont’s 1st Covid ‘surveillance report’ — and w..."
  • Hidden
  • Hidden
  • This field is for validation purposes and should be left unchanged.