February 10, 2021
By: Bill Osmulski
The Wisconsin Department of Health Services (DHS) was forced to finally abandon its flawed positivity formula this week because it was producing increasing inaccurate results – exactly what the MacIver Institute predicted would happen back in September.
That formula attempted to calculate Wisconsin’s positivity rate using a “by person” method, DHS incorrectly believed was backed by the CDC.
Every day, DHS compared the number of new COVID-19 cases and the number of people tested for the first time. The CDC’s instructions for the “by person” method are to compare the number of new people with COVID-19 and the number of individual people tested. MacIver warned that DHS’ mistake would lead to an inaccurate and inflated positivity rate.
“As the pool of people who have never been tested before continues to shrink,” according to the September 23rd MacIver Institute report. “Subsequently the positive test rate will continue to artificially rise.”
MacIver recommended DHS use a much simpler method, also recommended by the CDC, that focuses on tests rather than people. The CDC’s instructions for the “by test” method are to compare the number of new positive tests with the total number of new tests processed.
Milwaukee Journal Sentinel reporter Eric Litke “PolitiFacted” the MacIver Report. He claimed it was “both false and ridiculous,” and gave it a “Pants on Fire” rating.
Unfortunately for Litke, the Journal Sentinel, and PolitiFact, DHS adopted MacIver’s recommendations to use the “by test” method immediately after the story was published.
.@DHSWI caves to MacIver recommendations and now posts the correct positivity rate on its website. Wisconsin’s real rate is 9.4%, while the old phony rate is 17%. A MacIver study exposed DHS’ bad math. https://t.co/8qx2wIe0lh #WIright @ericlitke @journalsentinel pic.twitter.com/zFaKqO2bbA
— MacIver Institute (@MacIverWisc) September 30, 2020
For a time, DHS also continued to calculate positivity using the old “by person” method. The contrast between the two results was striking. In October, the “by person” positivity rate was double the “by test” rate. It would eventually quadruple. On February 2nd, the “by person” rate was 20% compared to the “by test” rate of 4.9%.
Four months after MacIver’s initial report, on February 8th, DHS acknowledged the “by person” method was not working.
“As the pandemic continues, calculating 7-day percent positive by person becomes a less useful metric than percent positive by test,” DHS states on its COVID-19 website. “This is because many Wisconsinites have been tested multiple times for COVID-19. In percent positive by person calculations, people are only counted once – either as a positive or negative case. Over time, the proportion of people with positive test results has increased, causing the 7-day percent positive by person to increase as well.”
DHS not only took down the “by person” metric from its website. It also removed all its historical data from the past year.
This decision by DHS to abandon its flawed “by person” formula validates and vindicates the MacIver Institute report, which PolitiFact still labels as a “Pants on fire” lie.