More than 3 million people in the U.S. are estimated to be contagious with the coronavirus

More than 3 million people in the United States have active coronavirus infections and are potentially contagious, according to a new estimate from infectious-disease experts tracking the pandemic. That number is significantly larger than the official case count, which is based solely on those who have tested positive for the virus.

The vast — and rapidly growing — pool of coronavirus-infected people poses a daunting challenge to governors and mayors in hard-hit communities who are trying to arrest the surge in cases. Traditional efforts such as testing, isolation of the sick and contact tracing can be overwhelmed when a virus spreads at an exponential rate, especially when large numbers of asymptomatic people may be walking around without even knowing they are infectious.

To put the 3 million-plus figure in perspective: It is close to 1% of the population. It is about equal to the number of public school teachers in the country, or the number of truck drivers. If the University of Michigan's football stadium were packed with a random selection of Americans, about a thousand of them would be contagious right now.

Columbia University epidemiologist Jeffrey Shaman said his team's model estimated that 3.6 million people are infected and shedding enough virus to infect others. That's a 34% week-to-week increase that followed a 36% increase in the previous seven-day average, he said.

The estimate does not include an approximately equal number of latent infections among people who caught the virus in recent days and can't pass it on yet because it is still incubating.

"It's bad; it's really, really bad," Shaman said. "We're running into Thanksgiving now and that's only going to make it worse. We're going to go through a lot of people being infected between now and the end of the year, unfortunately."

Separately, modelers at the University of Washington's Institute for Health Metrics and Evaluation estimated Tuesday that approximately 3.2 million people have been infected just since Election Day, Nov. 3, a figure significantly larger than the approximately 1.95 million official cases tracked over the same period by the Washington Post through reports from state health departments.

The IHME model forecasts continued daily increases for a month and a half, estimating that 245,000 people would become newly infected on Tuesday alone.

"When do you want to hit the brakes? That's the question," said Ali Mokdad, an epidemiologist at IHME who is among many scientists and doctors urging action by the government and general public to reverse the trend lines. "When you have a fire, you send the firetruck. You don't wait and say, 'OK, let me wait a little bit, maybe that fire isn't going to spread that much' … We already moved into exponential growth. Just hit the brakes as soon as you can."

This fall wave of infections and hospitalizations is different in several ways from the one last spring. The outbreaks are widespread now, with 49 states showing rising coronavirus hospitalizations, in contrast to the spring's concentration of cases in the Northeast and a few large cities. Doctors are better at treating severe cases of COVID-19, the disease caused by the virus, and so it is less likely to be fatal.

But the biggest difference is that this fall wave is still swelling, and is probably many weeks from cresting. This pandemic is following the seasonal pattern of the 1918 influenza pandemic, the worst such plague in U.S. history, in which the autumn wave was worse than the first one in the spring of that year. Even after more than 11 million confirmed infections since early in the year, more than 8 in 10 people in the United States remain susceptible to the coronavirus, experts estimate.

The non-tested cohort includes people who have no symptoms, but will in coming days. It also includes people who will never develop symptoms. And it includes people who have symptoms but don't want to deal with the consequences of a positive test, such as being forced to miss work or become isolated from their social network.

What happens next depends on public awareness of the reality of the emergency and the willingness of state and local officials to respond in a manner commensurate with the crisis while also acceptable to people suffering from pandemic fatigue.

Despite lower mortality rates, the dramatic rise in infections that began in September has seen a delayed echo in the rise of the death toll. More than a thousand people are dying daily on average, and the country is nearing 250,000 deaths since the start of the pandemic.

Although the surge of infections this fall has in some measure been due to colder weather and people congregating indoors, it has also been because of human behavior — specifically, the willingness of millions of people in the United States to ignore public health guidelines on facial coverings and social distancing. What is happening across the country is not inevitable, experts say.

Local and state leaders know they have to act, and those actions will be unpopular with many people. There is little appetite for severe shutdowns and closures, but the current trends in infections and hospitalizations suggest that incremental measures may simply be too little too late.

The fall surge also is happening in a transition period in which there is no clear national leadership on how to respond.

President Donald Trump, who refuses to concede that he lost the election, remains the point man for a largely hands-off approach to the contagion. Trump has embraced the view of his adviser Scott Atlas, a radiologist, that it is better to focus on protections for highly vulnerable people and otherwise allow the virus to spread among the rest of the population to hasten herd immunity.

The nation's leading infectious-disease experts say that strategy will lead to many tens of thousands of avoidable deaths because it is impossible to separate the most vulnerable from the rest of society.

"This nonsense of herd immunity means we give up," Mokdad said. "It's unethical. It means we let people die. That's not acceptable."

On Sunday, Atlas took to Twitter to urge residents in Michigan to "rise up" against Democratic Gov. Gretchen Whitmer's new restrictions designed to control the contagion. Many Trump supporters have also bought into his argument that the pandemic has been exaggerated by the news media and scientists as a way of hurting his re-election chances.

Before the election, Trump predicted that news coverage of the pandemic would end on Nov. 4. The opposite has been true — the pandemic remains on the front pages of newspapers and leads the nightly news broadcasts. But the cultural divide remains, with the wearing or non-wearing of masks seen as a political identifier.

The current strategy in many states — patchwork restrictions, combined with widespread public disregard of public health guidelines — portends a brutal winter for the country in which hospitals could be strained to their limits or beyond, something that is already happening in the Dakotas and other communities of the Upper Midwest and Great Plains, as well as El Paso.

"We're getting the herd immunity strategy whether we want it or not," said Andrew Noymer, an epidemiologist at the University of California at Irvine. "We have a totally out- of-control epidemic and we are taking baby steps."

The only period comparableto the current one in terms of the size of the infected cohort was in late March and early April. Tests were hard to come by early in the pandemic. Robert Redfield, director of the Centers for Disease Control and Prevention, estimated that the number of infected people was 10 times the official count.

Now the multiplier is surely lower — the model developed by Columbia University researchers uses 5.5. By that measure, as many as 10 million people in the United States have been infected in the past two weeks.

Researchers at IHME estimate the number of daily new infections by working backward from the known numbers of deaths. The institute relies on an estimated infection fatality rate of 0.6% across all populations, Mokdad said.

These are models, and they rely on a series of assumptions about a virus for which much remains unknown — including how long, exactly, a person who is infected will remain contagious.

But the big picture is clear.

"We're going in the wrong direction," Shaman said.