Maine appears to have risen to the top of the nation judging by one measure that tries to gauge how extensively the coronavirus is spreading.
Yet even as infections surge, Maine is still outperforming most other states on the indicators that epidemiologists use to track the pandemic on a day-to-day basis and that they say present a fuller picture of the virus’ spread. The state is still seeing far lower rates of daily new cases, hospitalizations and positive test results than most of the nation.
Commonly referred to as R, the reproduction number is meant to gauge the average number of people that will catch the coronavirus from a single infected person, with a number greater than one meaning that an outbreak is growing and less than one meaning it will eventually diminish.
By that measure alone, the virus has recently been spreading at a greater rate in Maine than in just about every other state, according to one online model developed by researchers from the Harvard T.H. Chan School of Public Health and another developed by a group of software developers that includes the founders of Instagram. They have each assigned Maine a recent reproduction number of around 1.45.
But scientists offer a number of caveats about the value of the reproduction number, which can’t be calculated in real time and must instead be modeled using older data.
Those models rest on “a fair number of assumptions,” according to Dr. Peter Millard, a former epidemiology staffer at the U.S. Centers for Disease Control and Prevention and an adjunct professor with the University of New England.
While it’s generally helpful to know if the reproduction number is greater than one, Millard said, it will not help to identify the latest trends that epidemiologists must quickly respond to and publicize to contain the pandemic, such as an outbreak in a nursing home or the spread of the virus through a rural town.
“I don’t think it’s a reliable statistic that would have much value in Maine,” Millard said.
Models of the reproduction number can also be easily swung in states such as Maine, New Hampshire and Vermont that are now seeing upticks of the coronavirus after having relatively low numbers of cases throughout the pandemic, according to Robert Long, a spokesperson for the Maine Center for Disease Control and Prevention.
“That’s not bad or wrong. But it means that a state like Maine, which was stable at 20-30 cases/day, will appear to be ‘spiking’ when we go to 40-60 cases, to say nothing of 150-200,” Long said. “That’s what happens when you look at relative percentage differences more than anything else.”
For the last two weeks, Maine CDC Director Nirav Shah has been publicly presenting the data that underpin those mathematical models, according to Long. He said that a “more germane” measure would present the incidence of the virus on a per-capita basis.
Public health experts frequently point to growth in new cases and hospitalizations, as well as the portion of tests coming back positive, as some of the most telling measures of the pandemic’s impact on a state.
On Tuesday, despite the state’s steadily rising case numbers, Maine’s seven-day rolling average of new daily coronavirus cases was still one of the lowest in the nation. Maine recorded an average of 1.45 new cases each day for every 10,000 people over the past week, placing it 43rd in the nation, according to data from Johns Hopkins University’s Coronavirus Resource Center.
The seven-day average of Mainers hospitalized with COVID-19 has risen from 10 hospitalizations per million people in late October to 37 per million now, according to the COVID Tracking Project, but that’s still well below the national rate of more than 150.
Maine’s test positivity rate has averaged out to 2 percent over the last seven days, according to data tracked by Johns Hopkins University. That rate has been going up in recent weeks but is still the second lowest in the nation, with only Vermont having a lower one.
North Carolina ranked just behind Maine in terms of new daily cases, yet it has one of the lowest R values in the country in both the models from Harvard and the software developers. Vermont, with the lowest rate of new daily cases, has the third highest R value in the software developers’ model, behind only Maine and New Hampshire.