UConn researcher: 30% of CT COVID immune

About 30 percent of Connecticut’s population may be immune from the coronavirus, according to one researcher.

“But that's not enough,” Pedro Mendes said.

Mendes, a researcher and professor in computational biology at the University of Connecticut, has been modeling the spread of the coronavirus in Connecticut since the start of the pandemic.

As a baseline, he said 75 percent immunity is necessary to stop transmission of the virus, but that number will shift as more infectious variants take hold.

“That number depends on how infectious is the virus,” Mendes said. “With the new variant, the B.1.1.7, that number goes up, and it's probably more towards 80, 85 percent.”

Mendes’ estimates of immunity include natural immunity, obtained from having contracted and survived the virus, and through vaccinations. He’s using something called a SIR model, estimating susceptible, infected and recovered people.

“It has the group of infected people, those who have active infection,” he said. “Therefore, by contact with susceptibles, they can also infect them, and then there's a group of recovered. So those are people that have recovered from the infection, and they're no longer infectious, and we assume they have some immunity.”

To do that, Mendes said his model uses “clues.” Starting with the number of confirmed infections at any given time, there is also an estimated infection rate and a known number of hospitalizations and deaths.

“That's like a puzzle,” he said. “All those pieces have to adjust together to make it work.”

That model also helps him estimate what may be the current true spread of the virus in the state.

According to Mendes, there are 14,000 people in Connecticut actively fighting the coronavirus, asymptomatic and severe infections, and everything in between.

Using state data and a complex computational model, Mendes estimates the number of known active infections to be 8,810 across Connecticut. But the real number is probably 5,000 more, a total of 13,607, Mendes said.

“Let's put it this way, we know of a little bit more than half of the people who are infected,” he said.

Mendes’ model does have some limitations. For example, that estimation of 30 percent immunity ignores the relatively small number of reinfections, people who have been infected twice, and breakthrough infections, people who have become infected after being vaccinated.

It also ignores age breakdowns. So, while close to 100 percent of the state’s nursing home residents have been vaccinated, and are therefore immune, zero percent of the state’s children have been vaccinated.

The model also assumes the same strength and duration of immunity for all vaccines and for immunity for recovered patients.

But the model improves the more people get tested, and Mendes praised the state for its focus on diagnostic testing.

“The key factor is the testing, the number of people tested, which is one of the best things that has happened in Connecticut. We are testing a large number of people,” he said. “And even though now it's smaller, and I wish they were still testing 40,000 people a day, we are still testing more than other states are. And that's one of the biggest defenses because you know, once you identify someone as positive, they take more precautions. They stay home.”