Calculation of COVID-19 positivity rate by university produces lower numbers than state

More

The Illinois Department of Public Health reports 7-day rolling averages of the COVID-19 test positivity rate for each region of Illinois. Champaign County, home of the University of Illinois at Urbana-Champaign, is part of region 6 which has experienced 8 days of positivity increases, reaching 12.2% on November 8.

The University of Illinois at Urbana-Champaign’s revised calculation of the campus positivity rate of COVID-19 has potentially lowered the rate substantially by reducing the number of reported positive tests.

The university changed its calculation in August, saying it was applying a more commonly accepted standard that differed from the way the state reports positive tests.

The university had calculated the positivity rate by dividing the total number of positive tests by the total number of tests. The state currently determines the positivity test rate by that calculation.

On Aug. 31, however, the university said it had changed to calculating the positivity rate by dividing the number of unique positive tests by total tests. That method eliminates multiple positive testing for one individual but keeps all of an individual’s negative tests.

For example, if 1,000 total positive tests are divided by 10,000 total tests, a positivity rate of 10% results. But if multiple tests of individuals were eliminated and only 500 unique positive tests remain, then the positivity rate would drop to 5%.

Rebecca Smith, epidemiologist and professor at the university, said removing tests performed on the same individual would be “difficult”, but may be considered in the future.

“I agree that [unique cases divided by unique tests] is a better calculation for decision-making,” she said in an email in November. “We do our best to use that internally. For the dashboard, though, we want to match the methods used by the state and by other dashboards.”

The university required COVID-19 testing for students beginning in August, with most students testing at least twice a week. Testing was later reduced to once a week for faculty, staff and graduate students and increased for some students to three times a week.

According to Robin Kaler, associate chancellor for public affairs, the SHIELD testing team at the university said persistent test positivity of individuals infected previously is a risk in reporting data.

“As we have accounted for [persistent test positivity] in our testing response, we ask everyone on campus to continue testing,” she wrote in an email. “That results in a number of non-infectious positive tests, which we remove from our data before posting to the dashboard… Reporting all positive tests would be uninformative or misleading. Therefore, we only report new positives.”

The Illinois Department of Public Health has said the numerous saliva tests for COVID-19 at the University of Illinois at Urbana-Champaign have dramatically driven down the overall positivity rate for a 21-county region that includes Champaign County. It did not respond to questions about whether it knew that the university did not use the same calculation as the state.

Like many states across the nation, Illinois publishes the rate of total positive tests over total tests, known as “test positivity”.

As a result of the university’s extensive testing, the state has separated out the tens of thousands of saliva tests from the university when calculating the positivity rate for the region. 

For example, taking the saliva tests out of the calculation of the positivity rate has led to the region having above an 8% positivity rate, instead of a 2.9% positivity rate on October 26. The rate grew higher in November, reaching 12.2% excluding saliva tests, and reaching 4.6% including saliva tests on November 8.

The exceeding of the 8% rate led in early November to the reinstatement of tighter restrictions on gatherings and businesses.

Without the university numbers, Champaign County’s positivity rate has exceeded 5%.  Even though it has not reached 8%, Champaign County came under the same restrictions as the region.

At the time of the revised calculation, the university said the method was more in keeping with the standards set out by Johns Hopkins University of Medicine, Johns Hopkins recommends that the number of people who test positive be divided by either unique people, encounters, or tests.

Epidemiologists say that the more accurate way to calculate positivity would be to divide unique positive tests by unique total tests, but that information is not always available.

Johns Hopkins Coronavirus Research Center also notes on its website: “…test positivity is a measure of testing capacity and while it can provide important context about case totals and trends, it is NOT a measure of how prevalent the virus is in communities. Policy decisions, like openings and closings or interstate travel, should not be determined based on test positivity alone.”

Johns Hopkins also notes that there are no federal standards for reporting COVID-19 testing data in the U.S. 

“This makes it impossible to offer a fully apples-to-apples view of testing data at the national level. Without federal standards, states have been left to forge their own paths, and as a result, they report testing data differently,” the site said.

Awais Vaid, epidemiologist and deputy public health administrator for Champaign-Urbana, said in an email, “I personally agree with Johns Hopkins that we [ideally] should calculate unique case positivity rate.”

Epidemiologists have praised the University of Illinois for its extensive testing of students, faculty and staff, saying the intense testing helps to quickly identify infected persons and get them isolated. However, the university has struggled to isolate and quarantine students who have come in contact with infected students because some students avoided county contract tracing staff or violated isolation and quarantine orders.

Multiple requests for comment from the Illinois Department of Public Health were not returned. Requests for comment from the governor’s press office have not yet been returned.

Leave a Reply