AUGUSTA — A mere 10 percent variation in how well Mainers practice social distancing and take other steps to reduce coronavirus transmission rates could mean the difference between a few hundred lives lost and 1,000-plus deaths.
That’s the grim calculus facing state epidemiologists, whose daily job is to ensure that there are enough hospital beds, ventilators and protective gear to handle either scenario – or a completely different one – when the anticipated surge of COVID-19 patients hits Maine.
“We don’t know which one we are going to be,” Dr. Nirav Shah, director of the Maine Center for Disease Control and Prevention, said of the various modeling scenarios. “What I do know is that no matter which one of these we are going to be, we’ve got a plan.”
Shah and two members of his planning team meet or talk daily to go over what the latest models tell them could be coming Maine’s way with the COVID-19 disease caused by the coronavirus. The Maine CDC reported 519 confirmed cases and 12 deaths as of Tuesday, but Maine is likely a week or two behind New York City, Boston and other areas of the country in the cycle of the virus.
The various models are inherently imprecise and uncertain. And the resulting, wide-ranging estimates, which Shah shared with a handful of reporters at the Maine Emergency Management Agency on Monday, illustrate the challenge of anticipating the course of a new, fast-moving infectious disease before scientists definitively know how it spreads in a modern, mobile society.
“The reason we do all of this is to plan, not to predict,” Shah said.
ESTIMATE OF 115 TO 1,000 DEATHS
For instance, a worst-case scenario suggests that Maine could see up to 12,000 new cases of COVID-19 per day at the peak of the outbreak if Mainers were only able to reduce the transmission rate of the virus by 40 percent, which would be very low compliance with “social distancing” mandates.
That figure would fall to 3,000 new cases per day at peak if Maine is able to reduce the transmission rate by 70 percent – and potentially to single digits if the aggressive control efforts are even more successful.
Similarly, the various models used by Shah and his team suggest Maine could experience anywhere from roughly 100 deaths to more than 1,000 by later this year – again, all depending on efforts to control the rate at which the virus jumps from person to person.
In any infectious disease, epidemiologists look to discern the contact rate or “reproduction number” that signals how many other people someone with COVID-19 will typically infect.
Scientists currently estimate that COVID-19’s rate is around 2.2 – one person passing the disease on to another two-plus individuals – but some studies have pegged the rate as high as 4.0 based on early data. Those seemingly small numerical changes, however, lead to exponentially different infection rates.
That number needs to be less than 1.0, Shah said, so that the outbreak will essentially “tire itself out” and public health officials will begin seeing the much-anticipated flattening of the infection curve. And while the various models out there have different projections on infection rates and deaths, Shah said, they all agree that “aggressive social distancing” is key.
“The course of the epidemic, fundamentally, in Maine will depend on how much public health interventions reduce the contact rate,” Shah said. “The virus is not going to change … so the only other lever that we have is societal mechanisms. In some cases that might be a vaccine, or it could be a drug. But right now, all we’ve got is social distancing.”
A widely viewed state-by-state model developed by researchers at the Institute for Health Metrics and Evaluation at the University of Washington, for instance, estimated that Maine could see 115 deaths by August. The institute estimates peak usage of Maine’s health care system will happen next week, on April 15, and that Maine should have enough ICU beds and ventilators to handle the surge.
The University of Washington model combines the latest data on COVID-19’s spread and mortality rates in China, Spain and Italy with local restrictions imposed by states, such as school closures and mandatory “stay-at-home” orders.
Those numbers change based on new data from countries further along the infection curve as well as state-level restrictions. For instance, Maine’s estimated death toll fell from 334 last week to 115 this week based on new data. But the researchers caution their estimates assume social-distancing measures will remain in place through May.
“If social distancing measures are relaxed or not implemented, the U.S. will see greater death tolls, the death peak will be later, the burden on hospitals will be much greater, and the economic costs will continue to grow,” Dr. Christopher Murray, director of the Institute for Health Metrics and Evaluation, said in a statement.
Using a weather analogy for the various models, Shah compared the University of Washington model to consulting an almanac to get an idea of what the weather will be like in Maine on, say, July 17.
Two of the other models that Shah and his team use – produced by Johns Hopkins University in Maryland and Imperial College of London – would be more akin to using a supercomputer to analyze complex meteorological data in order to produce a weather prediction.
The Johns Hopkins model estimates that Maine could see between 200 and 300 deaths if the transmission rate were reduced by 70 percent, but more than 1,000 deaths if that “contact rate” was only reduced by 60 percent. Johns Hopkins also provides more local modeling opportunities by combining county-level infection rates with demographic data from the census and travel data culled from anonymous cellphone tracking programs.
Shah said his team uses “a combination of models that use different techniques – each with strengths and limitations” in order to come up with their planning scenarios.
BUT IS IT ENOUGH?
All of this is geared toward making sure Maine has sufficient numbers of intensive care unit beds, ventilators, respiratory therapists and personal protective gear to keep up with the disease, regardless of which scenario plays out.
Both the Washington State and the Johns Hopkins models suggest Maine should have enough critical care hospital beds and ventilators to handle the estimated surge. But when asked about those projections, Shah remains cautious and replies, “We are still planning.”
“By design, we are not looking for THE answer,” Shah said. “We are looking for the widest range of answers that we believe are scientifically defensible, so that we can plan accordingly.”
Shah has been reluctant – even resistant – to provide estimates of sicknesses and deaths in Maine even as public health officials and governors in other states have done so. Instead, he uses his daily briefings to discuss the latest case numbers in Maine and, as of last week, the growing death toll while providing updates on the testing, protective equipment acquisition, and the availability of intensive care beds and ventilators.
In his meeting with reporters Monday, Shah said it is important to understand the complex data inputs as well as the inherent uncertainty of modeling an outbreak such as this to put estimates into proper context.
But he credited Maine residents with largely abiding by the “stay-at-home” mandate imposed by Gov. Janet Mills last week and maintaining the recommended 6-foot buffer zone with others when they have to go out in public. The “grim” estimates of COVID-19 illnesses and deaths reflected in models underscore the need to keep that up, he said.
“This is serious stuff,” Shah said. “I’d rather be straight with people that this is a serious disease. The fatality rate is not anything to trifle with. It spreads a lot more easily than the flu.”
Send questions/comments to the editors.
