‘It’s getting confusing.’ Miami-Dade and Florida can’t agree on metrics for COVID-19
A long-running difference in the way Miami-Dade County and Florida’s health department calculate the local rate of positive test results for COVID-19 — a key measure that shows whether infections are rising or falling in the area — came to a head this week, confusing residents and policy makers on the county commission.
On Friday, the rate of positive test results in Miami-Dade on the county’s New Normal dashboard reflected a 14-day average of 27.05%. The state health department’s two-week average was much lower, at 20.5%.
While both numbers are far above the 3% that public health officials say would indicate the virus is being suppressed, the continuing discrepancy between the two figures — which are calculated by looking at the percentage of positive results out of all the tests processed in a day — is causing consternation in the county as leaders try to base decisions about closures on the number.
“We need to get to the bottom of the issue of positivity reporting,” Jennifer Moon, the county’s budget director and deputy mayor overseeing the reports, wrote in an email Wednesday to Yesenia Villalta, administrator of the Florida health department’s Miami office. “It is critical that since this data is being used to make determinations regarding Emergency Orders, business closures and school re-openings, that we understand what is being reported and what causes these differences.”
On Wednesday, during a meeting of the commission’s health committee, Commissioner Jose “Pepe” Diaz asked for an explanation of the statistics being reported by the county and state to gauge the disease’s spread.
“It’s getting confusing,” Diaz said.
Moon told Diaz the county’s numbers reflect all tests reported in a day while the state health department matches its count to when a test was actually administered.
“Their data is not accurate until a week or 10 days after the test is taken,” she said.
Different data, different results
But the issue is more complicated.
Florida actually reports at least three different positivity rates for COVID-19 each day: a statewide rate based on all tests taken the prior day, including retests; a statewide rate for first-time positive tests only, excluding retests; and a cumulative positivity rate for everyone tested since March. All those rates are available daily on Florida’s COVID-19 report.
Moon said there may even be a fourth positivity calculation being provided by the state to Gov. Ron DeSantis, though she didn’t know what it said or how it was calculated. But everyone should be looking at the same number, she said: “We should all be on the same page. We want to work with them.”
The email exchange between Moon and Villalta after Wednesday’s meeting showed that the two officials were far apart. Both tried to explain how they calculate the rates but they didn’t seem to reach agreement, with Villalta at one point saying that “the state and county reports provide different data and should not be compared.”
The number of tests being counted to reach the figures was also unclear, according to the email exchange. Moon closed the conversation by asking that the health department provide her with the data it uses to create its charts. She and Villalta are scheduled to meet Monday to go over the differences in reporting. Villalta was not available for comment on Friday, an agency spokesperson said.
What can we learn from the data?
Florida has tested more people for COVID-19 in the past two weeks than at any other time during the pandemic, an average of 84,359 people a day, representing an increase of nearly 60% over the prior two-week period.
But even with such a dramatic increase in testing, public health experts say Florida’s high positivity rate strongly suggests that the state may not be casting a wide enough net to know how much the virus is spreading.
The 11,466 new cases confirmed by the state on Friday represent a positivity rate of 11.85% for new cases. The other positivity rates calculated offer a slightly different number: the overall percentage of positive results for persons tested each day, including retests, was 14.34% on Friday, and a cumulative positivity rate for everyone tested since March was 11%.
Again, all of those are far above the level that epidemiologists say would give them hope the virus is being suppressed and tell them if there’s enough testing.
Brittany Kmush, an infectious disease epidemiologist and assistant professor of public health at Syracuse University in New York, said the positivity rate can be calculated in different ways depending on what someone wants to measure. The cumulative rate of posivity can provide insight into the extent of the pandemic, she said.
But the calculation Kmush follows most closely is the daily positivity rate for new cases, which tracks only first-time positive results and excludes all retests.
“It would answer the question: Is the pandemic increasing in your area, or decreasing in your area,” Kmush said. “That’s the one I’ve been kind of focusing most on — that kind of instantaneous positivity rate.”
If the positivity rate is too high, that may indicate that the state is only testing the sickest patients who seek medical attention, and is not casting a wide enough net to know how much of the virus is spreading within a community, said Jennifer Nuzzo, a public health epidemiologist with the Johns Hopkins Center for Health Security.
A low rate of positivity could mean that a state has enough testing capacity for the size of its outbreak and is testing enough of its population to make informed decisions about reopening, she said.
“The key thing is we’re trying to figure out with the question of positivity is are we catching all the infections that are out there,” Nuzzo said.
Nuzzo added that communities should strive for a positivity rate of 5% or lower for at least 14 days before reopening, as recommended by the World Health Organization. But she also cautioned that positivity rate is not infallible.
“It’s not a perfect metric,” she said. “You can gain a low positivity if you’re sure to test all the wrong people.”
This story was originally published July 17, 2020 at 7:14 PM.
