Researcher: 10% of state may have been infected with COVID-19

About 10 percent of Connecticut’s population has been infected with the coronavirus.

That’s according to Pedro Mendes, director of the University of Connecticut’s Richard D.

Berlin Center for Cell Analysis and Modeling.

Since the start of the pandemic, Mendes has been modeling the outbreak across the state and specifically drilling down in Hartford County. His intent, he said, is

to plan for the resource needs of John Dempsey Hospital.

There have been a total of 62,028 positive coronavirus tests in Connecticut as of Friday. That’s 1.7 percent of the total state population. But how many people have had the virus but have not been tested is something of an open question.

Mendes’ 10 percent estimation is based on his computer model, and includes “people who have had the disease and have not died,” he said.

“That includes asymptomatics” — people who have contracted the virus but have not displayed any symptoms — which is complex as asymptomatic patients typically don’t get tested.

The model is an estimation. Mendes said there is a deviation of plus or minus 5 percent but, even if 15 percent of the state’s population has obtained some measure of immunity, that would be well below what is needed for so-called “herd immunity,” in which enough people in any given population are immune to stop transmission of the virus.

The number of people who would need to have immunity to prevent the spread of the virus falls somewhere in the range of 70 percent. Mendes said it’s “at least 75 percent” of the population.

Dr. Anthony Fauci, speaking last week on Connecticut-based radio show “Conversations on Health Care,” said it would need to be 70 percent, through a “combination of vaccines and natural immunity.”

Yale researcher Virginia Pitzer said earlier in the pandemic that the target should be 67 percent — that a vaccine would need to be 67 percent effective at creating immunity and administered to 100 percent of the population, or 100 percent effective and administered to 67 percent of the population, or somewhere in between.

But how many people with immunity is not the only concern.

“The other thing about herd immunity, you actually need for people to stay immune for a long time,” Mendes said.

His model also is predicting an increase in the number of cases of and hospitalizations from COVID-19 in Connecticut over the next few weeks.

The worst-case scenario — which Mendes said was very unlikely — is “a peak similar to the one we had in April.”

Predictions get less solid the farther out you go, and Mendes only forecasts three weeks out. Beyond that, he doesn’t trust his model.

“It looks like a spaghetti plot like they do with the hurricanes,” he said.

As of Friday, there were 184 people in the hospital with a COVID-19 infection. Mendes’ most probable model puts that at 225 by the end of this week.

But again, it’s difficult to predict. A lot depends on “superspreader events,” in which a few people pass the virus on to a crowd. Mendes said they were “stochastic,” or randomly determined.

“Sometimes they happen, sometimes they don’t happen at all,” he said.