LIMA — Ohio was bracing for a surge in new coronavirus cases that would have quickly overwhelmed hospitals by the end of April. But the surge was much smaller than originally anticipated, so much so that Gov. Mike DeWine will unveil his plan Monday to partially reopen non-essential businesses.
How did the state go from predicting 10,000 new cases per day to its latest model, which expects new cases to plateau at less than 2,000 cases per day if Ohioans continue to practice strict social distancing?
“We underestimated the impact of our distancing measures,” said Mike Oglesbee, director of The Ohio State University’s infectious disease institute and a professor of virology and comparative pathology. “That’s a good thing. It’s only human to say: wait a minute. It wasn’t as bad as you said it was, therefore we don’t believe any of it.”
DeWine acted earlier than most governors, prohibiting spectators from attending the Arnold Sports Festival in March before the state confirmed its first known cases of community spread on March 9. Colleges began postponing classes. Conferences and large events were canceled, even before mass gatherings were officially banned. Schools closed. Restaurants were forced to pivot to curbside pickup or shutter their stores indefinitely, finally culminating in the state’s stay-at-home order that shut down some non-essential businesses and altered the way many Ohioans went about their daily lives.
The purpose behind the collective social distancing measures, public health experts explain, was to slow the spread of the disease —which when not contained, spreads exponentially — so hospitals had time to build special COVID-19 units and avoid the worst case scenarios unfolding in places like Italy and New York City.
But the disease modeling used to guide these policy decisions has caused confusion, as the actual number of cases in some states like Ohio has been lower than once expected.
There are several prominent approaches to the disease modeling, from the SRI model, which considers susceptible, infected and recovered populations to predict the disease’s spread, to the University of Washington model which calculates backwards from the number of deaths.
The Ohio State University relied on a different model that looked at contact networks between individuals and communities to estimate how many people were susceptible.
Oglesbee said the raw data was drawn from patients who were symptomatic and had tested positive for the virus, which biased the sampling toward higher-risk populations. But researchers could see the disease was spreading exponentially based on new cases detected in early March.
“If that trend continued,” he said, “we would have hit that quite high initial surge.”
The models were later revised, Oglesbee said, when it became apparent that social distancing measures were working. But he said modelers were initially skeptical that the epidemic curve was flattening, as there were still testing shortages and backlogs.
“Some people could look at that and say, ‘Wait a minute. Now you’re saying 600 instead of 1,600 (new cases per day), so you must have no idea what you’re talking about,” Oglesbee said. “It’s really dangerous to be grasping onto one specific number in the forecast. The most important thing is the trend.”
The initial focus, he said, was hospital capacity. But now the modeling has pivoted to the recovery phase as Ohio gets ready to relax some of its social distancing policies.
Testing will continue to be an issue as more Ohioans are allowed to return to work.
Researchers at Harvard believe the U.S. needs to deliver at least 5 million coronavirus tests per day by June for states to begin to relax their collective social distancing measures. Daily testing capacity should then increase to at least 20 million per day by late July, they say, allowing the U.S. to test 2-6% of the population each day so more people can return to work.
They suggest focusing those efforts on the essential workforce first before remobilizing other sectors of the economy, but more supplies are needed to do so effectively.
“We are still at a place where testing is limited enough that we cannot even test all symptomatic individuals,” said Ross Kauffman, an assistant professor and director of public health at Ohio Northern University, “so we certainly are not at a place where we can be testing all of the essential workers and doing so on a regular basis in order to ensure that they are and remain healthy.”
That may start to change. On Friday, DeWine announced that the state will be able to run at least 7,200 tests per day by Wednesday, eventually reaching 22,000 tests per day by the end of May as more companies agree to produce swabs and reagents needed to perform widespread diagnostic testing.
Reach Mackenzi Klemann at 567-242-0456.