COVID-19 Simulator Predicts Sharp Spike in New US Cases Without New Stay-at-Home Orders

Could the United States have 100,000 new COVID-19 cases per day by the end of the summer?

This was the worry voiced by Anthony Fauci, MD, Director of the National Institute of Allergy and Infectious Diseases in his testimony to the US Senate this week. Dr. Fauci’s concerns were sparked by a recent uptick in daily new COVID-19 cases that now has the US averaging 40,000 new cases per day.

A COVID-19 Simulator developed by researchers at Massachusetts General Hospital, Harvard Medical School, Georgia Institute of Technology and Boston Medical Center predicts a similarly grim scenario by the end of August if states don’t take more direct action to slow the spread of cases, says Jagpreet Chhatwal, PhD, lead investigator of the study team.

The spike in new COVID-19 cases comes after a period where the average new rate of daily cases appeared to have stabilized at 20,000 per day. That number was a little deceiving, however, because some states were experiencing rising numbers of cases while others with high case loads such as New York and Massachusetts were declining—thus giving the appearance of a stability.

“If trends continue with daily cases rising sharply, we are going to hit 100,000 cases by the end of August,” says Chhatwal, a senior scientist at the Mass General Institute for Technology Assessment and an assistant professor at Harvard Medical School. “The question is—what can we do to avoid it? The simulator can provide insights.”

By bringing back stay-at-home orders until the end of August in states seeing a rise in cases, the simulator projects the national daily new case count to drop below 15,000 per day—a decrease of 85%.

'Pausing Is Not Enough'

Many states with increasing rates of new infections such as Texas, Arizona and Florida are starting to pause reopening efforts, but pausing is not enough, says Chhatwal. “We have to scale back.”

According to the simulator, if Texas continues along the same course through August, the daily new infection count could rise from 5,000 to 18,000. With a stay-at-home order, Texas could reduce its daily new case count to 2,500 by the end of August.

In states like Massachusetts and New York, where the average number of new cases is still on the decline, it may be possible to keep the virus in check without resorting back to stay-at-home orders, Chhatwal says. However, policymakers in those states should keep a close eye on what is happening elsewhere.

While it is not surprising to see the numbers of new infections rising as restrictions are relaxed, the question is how much of an increase is acceptable before we start to shut down again, says Turgay Ayer, PhD, a co-investigator and the director of business intelligence and health care analytics at the Center for Health and Humanitarian Systems at Georgia Institute of Technology. “In a world with no vaccine, strict social distancing is still the most powerful control we have, and we may have to go back to exercising stay-at-home orders,” Ayer says.

“We also need to balance the state economy with persistent risk of COVID-19 infection in the community—an outbreak in just one county may make it necessary to shut down the entire state before things go out of control, and hence early detection of such local outbreaks is of focal importance.”

The Mass General-Georgia Tech team first developed and launched their simulator in April as a way to apply their skills in infectious disease modeling to the unfolding pandemic. The simulator allows users to model the impact of different interventions at the national or state level from now through October.

The team recently received funding from the National Science Foundation to provide county-by-county modeling of interventions as well. Given that different counties within the same state can have widely different disease scenarios (think upstate New York versus New York City, for example), a county-by-county picture will allow for more targeted shutdowns that could help state officials better balance the interests of public health and the economy.

“As the situation in each jurisdiction and our understanding evolves, we hope to further enhance our COVID-19 Simulator to inform relevant policies driven by data,” says Chhatwal.