Facing criticism from all sides – as befuddled parents, teachers and public health experts struggled to make sense of the governor’s “tweaks” to the Harvard-inspired school reopening plan – West Virginia abandoned its original metric in mid-September.
Now, the state is using what it calls the “West Virginia Model.” And rather than basing school reopenings solely on the number of per capita COVID-19 cases, counties can reopen by increasing the number of negative test results. According to state officials, the change is designed to promote more testing.
“Flood us,” Gov. Jim Justice said in a press briefing Wednesday.
People did. The number of tests went up, and the virus “reproductivity rate,” a measure of the rate of virus spread, dropped. The state once had the highest rate in the nation, and now has one of the lowest.
The result? School boards have reopened classrooms, to the delight of some parents and student athletes. But teachers are furious.
And now, a system that lets counties effectively choose whether to base their reopening on the rate of new cases or the positivity rate means counties can remain “green” even with elevated case numbers. And experts say it’s just as flawed as the previous iteration.
The metric “doesn’t make any sense,” said Dr. Cyrus Shahpar, an epidemiologist and former team leader at the U.S. Centers for Disease Control’s Global Rapid Response Team. He now works at a global public health nonprofit, advising local governments on how to craft policies to fight the coronavirus pandemic.
West Virginia’s original reopening plan was based on a metric devised by researchers at the Harvard Global Health Institute. They calculated the number of positive cases per 100,000 people in each county. If that number, averaged over a week or two, climbed above 10, then a county was at “elevated risk.” The state would then order schools closed.
But after the plan delayed reopening in some counties for weeks, parents and coaches were fed up. They took the battle to Kanawha County court in a pair of legal challenges, one by a parent and another by a football star.
Meanwhile, the death toll continued to climb: as of Thursday, 370 people have now died in West Virginia from the coronavirus.
Judges eventually threw both lawsuits out, but the governor bowed to pressure to accelerate the process anyway.
The new plan adds a second way for counties to reopen schools: decreasing their percentage of positive tests. This is calculated by taking the number of positive test reports and dividing it by the total number of tests administered in a county.
To move into a lower risk category – like from “orange,” where distance learning is required, to “gold,” which allows in-person school – counties must either drop the number of positive cases or reduce their positivity rate.
And just like that, more orange counties became gold.
Dr. Clay Marsh, the state’s coronavirus czar, explained the rationale behind the changes in an interview with Mountain State Spotlight. “When we were using the infection incidence rate, people just stopped getting tested, and we were only testing the people that were sick,” he said.
Now, he said, “we hope we will identify the ‘super-spreaders.’”
Gaming the system
This change in how the risk is measured led to a rapid increase in tests – a record of nearly 10,000 on Oct. 1 – but also allegations of gaming the system.
One official, a school board member in Putnam County, got tested twice in two days.
“Got both back already, I’ll go every day if that’s what it takes,” Christian Wells wrote on Facebook on Sept. 25.
Justice’s comments and the ensuing social media fracas led Marsh to issue a series of clarifications. “We do not support repetitive testing only focused at driving negative tests,” he wrote on Twitter.
In an interview, Marsh added, “multiple tests in a single day or two days – that doesn’t seem right.”
The state is following the CDC’s recommendation of reporting the percent of positive tests, but some states, including Pennsylvania, report only one test per person even if that person got tested more than once.
Shahpar, the ex-CDC epidemiologist, recommended West Virginia report that metric as well.
Marsh said the state was working on it. Right now, he said, “we don’t have any automated systems that help identify each person in a way that we can easily do that.”
The changes have ignited another legal battle, this time with the West Virginia Education Association. The union filed legal action against the governor earlier this week, seeking to reverse the changes and return to the Harvard metric. It accused the governor of “manipulating” the state’s metric to move the state closer to what Justice has called the “finish line” – a full reopening.
In interviews with Mountain State Spotlight, a pair of leading public health experts also criticized the state’s decision to allow an improving positivity percentage to override data on the number of new cases.
Shahpar said the state was right to take into account the positivity percentage. But, he cautioned, it shouldn’t be the only metric.
“You have two numbers that you’re looking at,” he said. “You probably want to wait until both numbers are good.”
Tomás Pueyo agreed. Pueyo, a Silicon Valley tech executive, wrote a lengthy and prescient analysis of the pandemic in March that catapulted him into the spotlight as one of the nation’s most-cited coronavirus experts – short only, perhaps, of the CDC’s Dr. Anthony Fauci.
His thesis, that fighting the pandemic would require first a “hammer” – wide-reaching shutdowns – and then a lengthy “dance” of targeted suppression measures, is frequently cited as inspiration by Marsh, the state’s coronavirus czar.
But the suppression measures recommended by Pueyo, such as enforcing quarantine of positive cases and isolating all of their contacts, have not been fully implemented in West Virginia or nearly anywhere else in the United States. Unlike some other countries, the United States failed to stop the spread of the virus, where it has now killed over 212,000 people.
Pueyo acknowledged that it does “make a lot of sense to actually tweak these things” as we learn more about the virus, but “the thing that makes zero sense is choosing one or the other: positivity or prevalence.”
“That’s just stupid. You need both,” he went on.
Marsh defended the state’s changes, pointing to the rise in testing and the falling death rate since the new metric was implemented. A dropping positivity rate would eventually lead to a drop in cases, he argued.
“If we go back and choose the worst of the two rates… we would shoot ourselves in the foot again and go back to where we were,” he said.
Map tweaks spur big change in Kanawha learning
As of Thursday afternoon, Kanawha County was orange on Harvard’s map due to elevated numbers of new cases in recent weeks. But it was yellow on the state’s map.
As a result, the state’s largest school system had returned to in-classroom instruction for a few days earlier in the week. It was one of the last counties to offer in-person learning — a blended model with a few days of in-person instruction — since schools returned during the pandemic.
This blended approach was supposed to continue through Oct. 16. But the same day students returned to classrooms, the Kanawha County Board of Education voted without warning that the county would begin five days of in-person instruction Oct. 12. There would be no days off for teachers and personnel to disinfect their classrooms.
Teachers, understandably, were shocked, as the board’s decision followed a heavily modified map.
Robert McCloud, a science teacher at Riverside High School in eastern Kanawha County, said in an email, “I understand that tweaks are necessary.”
But, he added, “the tweaking of numbers… has eroded both my personal and public confidence in the system.”
The positive metric, he explained, “was never intended to work for this purpose” and isn’t an accurate way of measuring virus spread in county hotspots.
“We are playing with fire,” he wrote.