Coronavirus

Floridians flattened the COVID curve. Then, amid upbeat talk, the numbers began to rise

READ MORE


A Numbers Game

The Miami Herald analyzed both public and non-public data in order to illustrate the spread of the novel coronavirus in Florida, examine reopening benchmarks and look at what comes next.


In mid-April, Gov. Ron DeSantis started talking about reopening Florida, saying the worst of the novel coronavirus pandemic was behind the state. Reported new COVID-19 cases had steadily declined that month.

“People were literally just frightened to the dickens,” DeSantis said on April 20, criticizing the news media and public health experts for in his view exaggerating the crisis.

At the time, the numbers had been trending downward. But by the time reopening began, the state no longer appeared to be meeting federal reopening guidelines for declines in new cases or positivity rates, although the numbers had not returned to peak level, a Miami Herald data analysis shows. (The Herald was able to perform an independent analysis of positivity because it exclusively obtained non-public state data. Read the methodology for our analysis.)

Alberto Moscoso, a spokesman for the Florida Department of Health, did not challenge the Herald’s finding but defended the governor’s approach without providing the state’s own analysis. “The Governor’s first priority in reopening Florida continues to be maintaining the health and safety of all residents,” he said.

Now, the state is moving into the second phase of reopening as both positivity — the percentage of individuals testing positive out of total people tested — and new cases continue to rise.

The Herald analyzed how COVID-19 entered the state largely unnoticed at first, how cases spiked, how Floridians flattened the curve — and how that encouraging trend began to show signs of reversing.

On March 1, the state reported its first confirmed cases of COVID-19. But an examination of the state’s public data shows investigators at the Department of Health have traced COVID-19 cases as far back as early January, when people who later tested positive first began to feel symptoms.

The Herald used symptom onset dates — which can be loosely correlated with the date someone got sick — to show how COVID-19 spread through Florida in early 2020. (On the main page of its public dashboard, DOH plots cases by the date it received positive test results, which can be weeks after someone fell ill.)

Unable to see the graphic below? Click here to reload the page.

Number of people falling ill by day
= Seven-day average

In January, DOH data show a smattering of people were already getting sick, though the symptoms weren’t originally recognized as COVID-19 because of a lack of testing.

“The first few generations are really hard to see, particularly with a disease that presents like a flu,” said Eric Toner, a senior scholar at the Johns Hopkins Center for Health Security. “I think it’s safe to assume, as we have all along, that the major travel hubs or travel destinations were probably seeded in January.”

As the calendar turned to February, the virus was spreading throughout Florida. Throughout the course of the month, nearly 200 people started showing symptoms that would eventually be identified through testing as COVID-19.

According to a Herald analysis of DOH public data, one out of four people getting sick had not traveled and had no known contact with another person who was ill — meaning the disease was spreading in the state uncontrolled.

While testing was so rare that confirmation wasn’t possible in February, the state was aware of hundreds of potential cases, according to internal memos obtained by the Herald. (1) That information was not shared with the public or even local governments and hospitals, which were begging for guidance. (2)

Floridians continued on with their lives, largely unaware of the risks. Many municipalities allowed spring break and other parties to go forward. (3)

On March 1, when Florida publicly confirmed its first cases, DOH data show more than 70 people fell ill. (Those are only the patients who eventually tested positive for COVID-19.)

With no vaccine or therapy for the new disease, epidemiologists knew the virus would spread widely. The question became how fast that would happen.

If everyone got COVID-19 at once, a graph of cases by day would show a steep upward trajectory. A steep curve would mean rapid spread, which could overwhelm hospitals, a disaster that struck Italy in March. Slowing the spread — through social distancing or emergency closures — to avoid a precipitous spike became the best-case scenario.

DeSantis declared a state of emergency on March 9, allowing him to call out the National Guard, purchase medical supplies more swiftly and establish field hospitals, but he did not shut down the state. (4)

By then, new cases were exploding across Florida. Still, DeSantis denied that COVID-19 had taken a foothold, saying there was no “community spread” of the virus in Florida.

The governor was contradicted the next day by Dr. Anthony Fauci, widely considered the federal government’s top expert on infectious disease. (5)

As the state continued to delay extreme action, some local governments implemented emergency measures. Miami-Dade County closed all nonessential businesses on March 19, then issued a safer-at-home order a week later, on March 26. (6,7)

Floridians seem to have acted even faster on their own, possibly reacting to federal guidelines that suggested social distancing and working from home. By mid-March, residents had already started staying home, the Tampa Bay Times found by analyzing cellphone mobility data. (8)

Experts say it takes up to two weeks to see the impact a policy has on “the curve.” That’s because it can take that long for a person to start feeling sick after exposure to the virus. People falling ill within the first week after a policy change likely contracted the virus before the change went into effect.

Under pressure for not acting sooner, DeSantis issued a statewide safer-at-home order on April 1. The effect of that policy would not be reflected in the data for up to two weeks, experts say. (9)

Around that same time, new cases peaked and began leveling off, data show.

Two to three weeks after Floridians started staying home and local governments started closing businesses, the rate of new infections in Florida had slowed. The curve of new cases began to level off, the data show. Social distancing and business closures were working.

By mid-April, new cases across the state were down significantly. Hospitals had not been overrun. The spike in flu-like illnesses evened out.

Although the disease claimed hundreds of Floridians per week, with the effects especially dire in nursing homes, the count was significantly less than in New York and some other northeastern states. The curve was flattening — but so was the economy.

Emboldened by President Donald Trump, a close ally, and encouraged by business leaders, DeSantis began to talk about reopening. (10)

“We need to be ready to get back to work,” he said.

At an April 29 news conference, DeSantis announced an executive order to partially reopen much of the state. (11) The governor cited downward trends in flu- and COVID-like illnesses, and increased hospital capacity. He especially emphasized the declining rate of Floridians testing positive.

“The positivity rate is really significant, because that's an indication of how widely this is circulated, how aggressively it’s circulated,” the governor said. “And look, over the last two weeks, we've seen a consistent decrease, an average decrease in the positivity rate the last four or five days.”

Except that’s not what the state’s own data showed. Using internal DOH data, the Herald calculated Florida’s positivity rates and graphed the three-day rolling average of daily positivity, as advised by CDC guidelines, to smooth data irregularities. (12) These numbers would have been available to state officials in real time.

Percent positive (three-day rolling average)
= Statewide
= South Florida
= Statewide excluding S. FL
= Excludes large-batch test results

On May 4, all of Florida except for Miami-Dade, Broward and Palm Beach counties reopened.

CDC guidelines suggest that areas should see a downward trajectory either in positivity or in new cases over 14 consecutive days with no prolonged increases before reopening. (13)

A Herald analysis of 14-day linear trends found that as a whole, those 64 counties appeared to meet at least one of those two federal reopening criteria during most of April. (14)

That was no longer the case by early May.

On April 29, the same day DeSantis announced reopening, the state, excluding South Florida, was on the third day of a four-day sustained increase in average positivity. (15)

That uptick may have disqualified the state from using positivity as a basis for reopening on May 4 for at least another week per CDC guidelines. The counties also saw a rebound in new cases around the same time, according to the Herald analysis. (16)

The reopening of the 64 counties occurred anyway.

Indicators suggesting an increasing spread of the virus across the state predated reopening.

Though there is surely no single reason for the uptick, cellphone data show people outside of South Florida began to leave their homes more almost immediately after DeSantis began saying the pandemic was diminishing in mid-April. (17)

The increase in positivity came roughly two weeks later.

Since reopening, new cases have continued to go up in that part of the state, according to a Herald analysis of DOH data through June 6. (18)

Unlike the rest of the state, the three South Florida counties seemed to meet reopening criteria for positivity and new cases at the beginning of May. (19) Testing was increasing or even. New cases and positivity were consistently down.

However, by the time South Florida prepared to reopen, the region saw a sustained increase in positivity rates. (20)

Again, the increasing spread of the virus in South Florida predated reopening. (21)

DeSantis blamed large batches of test results from prisons and nursing homes, where the virus was rapidly spreading, for rising COVID-19 numbers in May. (22)

However, positivity was sometimes even higher after the batches were removed, according to a Herald analysis that removed batches of more than 20 tests from a single location in a single day. (23)

Even discounting batches, the state, excluding South Florida, still did not appear to meet federal reopening guidelines on positivity because of a sustained uptick in late April. (24)

In South Florida, a temporarily disqualifying four-day increase in positivity in early May is also still present after batches of tests were removed. (25)

Data obtained by the Herald end on May 13, so reporters could not independently analyze positivity rates or the effect of batch dumps after that date.

However, an analysis of DOH’s public data shows that since then three-day averages of new cases and positivity continued to rise across the state through early June — a potentially worrying trend as the state moves further into reopening.

El Nuevo Herald reporter Ana Claudia Chacin contributed to this report.

CITATIONS

  1. “...the state was aware of hundreds of potential cases, according to internal memos obtained by the Herald.

  2. “...not shared with the public or even local governments and hospitals, which were begging for guidance.
  3. “Many municipalities allowed spring break and other parties to go forward.”
  4. “DeSantis declared a state of emergency on March 9...”
  5. “The governor was contradicted the next day by Dr. Anthony Fauci...”

  6. “Miami-Dade County closed all nonessential businesses on March 19

  7. “...then issued a safer-at-home order a week later, on March 26.”
  8. “...residents had already started staying home, the Tampa Bay Times found...”
  9. “...DeSantis issued a statewide safer-at-home order on April 1.”
  10. “...encouraged by business leaders, DeSantis began to talk about reopening.”
  11. “...DeSantis announced an executive order to partially reopen much of the state.”

  12. “...as advised by CDC guidelines...”
  13. “CDC guidelines suggest that areas should see a downward trajectory either in positivity or in new cases over 14 consecutive days with no prolonged increases before reopening.”
  14. “...those 64 counties appeared to meet at least one of those two federal reopening criteria during most of April.”

  15. “...the state, excluding South Florida, was on the third day of a four-day sustained increase in average positivity.”

  16. “The counties also saw a rebound in new cases on May 2...”

  17. “Cellphone data show people outside of South Florida began to leave their homes more almost immediately after DeSantis began saying the pandemic was diminishing in mid-April.”

  18. “Since reopening, new cases have continued to go up for that part of the state.”

  19. “...the three South Florida counties seemed to meet reopening criteria for positivity and new cases at the beginning of May.”
  20. “...by the time South Florida prepared to reopen, the region saw a sustained increase in positivy rates.”
  21. “...the increasing spread of the virus predated reopening.”

  22. DeSantis blamed large batches of positive results from prisons and nursing homes...”
  23. “...positivity was sometimes even higher after the batches were removed...”

  24. ““...even discounting batches, the state, excluding South Florida, still did not appear to meet federal guidelines on positivity...”
  25. “In South Florida, a temporarily disqualifying four-day increase in positivity in early May is also still present after batches of tests are removed.”

This story was originally published June 12, 2020 at 8:00 AM.

Sarah Blaskey
Miami Herald
Sarah Blaskey is an investigative journalist for the Miami Herald, where she was part of the team that won the 2022 Pulitzer Prize for reporting on the collapse of a residential condo building in Surfside, FL. Her work has been recognized by the Scripps Howard Awards for excellence in local investigative reporting, the George Polk Award for political reporting and the Webby Awards for feature reporting. She is the lead author of “The Grifter’s Club: Trump, Mar-a-Lago, and the Selling of the Presidency.” She joined the Herald in 2018.
Aaron Albright
Miami Herald
Aaron Albright is an Experience Developer and data visuals journalist at McClatchy. He graduated from the University of Florida in 2017 with a Bachelor of Science degree in journalism. He has designed award-winning projects such as the Miami Herald’s “Perversion of Justice” and “Dirty Gold, Clean Cash” investigations.
Get unlimited digital access
#ReadLocal

Try 1 month for $1

CLAIM OFFER

A Numbers Game

The Miami Herald analyzed both public and non-public data in order to illustrate the spread of the novel coronavirus in Florida, examine reopening benchmarks and look at what comes next.