IE 11 is not supported. For an optimal experience visit our site on another browser.

When will coronavirus cases peak? It's getting harder for experts to predict

“If these trends continue to go up, I think this wave could continue through the winter," epidemiologist Loren Lipworth said.
Medical workers from New York handle test samples at a coronavirus test site in Houston on July 17, 2020.
Medical workers from New York handle test samples at a coronavirus test site in Houston on July 17, 2020.Go Nakamura / Getty Images

The United States passed a grim milestone Thursday, exceeding 4 million confirmed coronavirus cases and nearly 145,000 deaths nationwide. As the virus continues to spread in more than 30 states, experts say it could be a while before the country sees the end of this surge of infections.

In the early months of the pandemic in the U.S., the number of daily deaths peaked around mid-April, with more than 2,300 reported fatalities, before gradually falling to less than 1,000 deaths per day. Daily cases, which had fallen to a plateau of around 20,000 in mid-May through early June, have reached record highs in recent weeks.

Full coverage of the coronavirus outbreak

But the changing demographics of the latest outbreaks across the country, combined with inconsistent mitigation strategies by states, are making it more challenging for scientists to predict when the worrying new upward curve may start to level out.

“The trends that we see across the U.S. don’t look like they’re peaking anytime soon,” said Loren Lipworth, an epidemiologist at Vanderbilt University Medical Center in Nashville, Tennessee. “If these trends continue to go up, I think this wave could continue through the winter.”

Early on in the pandemic, states such as New York, California, New Jersey, Washington and Louisiana were among the hardest hit, but as cases and deaths have declined or plateaued in these regions, new infections — and subsequent fatalities — are rising elsewhere. As much of the country experiences sharp upticks in new COVID-19 cases, many states are now also starting to see their hospitalization and fatality rates rise.

“Overall, it’s a small proportion of confirmed COVID cases that end up in the hospital, and then a portion of those who die,” Lipworth said. “But even if it’s just 1 or 2 percent, as the number of cases skyrocket, that number really adds up.”

Public health officials had warned that there could be a second wave of coronavirus infections in the fall, coinciding with the flu season, but Lipworth said it seems likely that the country will still be squarely in the first wave by the end of the year.

“I don’t think we can even start thinking about a second wave until we can say confidently that the first wave has ended, and that’s certainly not the case in most parts of this country,” she said.

Lipworth added that the stages of a pandemic are not rigidly defined, but the U.S. would need to have the current outbreak under control before epidemiologists declare an end to the first wave. This hasn’t happened yet, she said, because even some states that saw cases and deaths level off or drop are now seeing both numbers creep back up again.

California, for instance, enacted strict lockdowns early on in the pandemic and slowed transmissions throughout the state. But now, new infections are being reported at record-high levels, and the state recently surpassed New York for the most confirmed coronavirus cases in the nation, at more than 422,000.

The diverse experiences of states have made it hard to model the trajectory of the pandemic on a broad, national basis, according to Jaline Gerardin, an epidemiologist at Northwestern University Feinberg School of Medicine in Chicago, who has worked extensively on modeling and disease projections within Illinois.

But as states experience surges in new infections, it’s likely there will be upticks in hospitalizations and deaths within several weeks, she said.

“The cadence that we saw with the first peak was that there was a couple weeks lag,” Gerardin said. “We had a very severe and sharp intervention — sheltering in place — on March 21, and then we saw hospital admissions peak around early April, so just two to three weeks later. And then deaths didn’t peak until six weeks later, at around early May.”

She said she expects similar lags this time around, but she added that a number of dynamics have changed since March that make it tricky to project when this surge will peak.

For one, younger people — those in their 20s, 30s and 40s — are making up a larger percentage of new cases in many states. In Florida, for example, the median age among new coronavirus infections was 65 in March. Now, people ages 25 to 34 make up the largest percentage of positive cases statewide, and the median age of new cases dropped to the 30s.

Download the NBC News app for full coverage of the coronavirus outbreak

This younger demographic could alter the projections of hospitalizations and deaths across states, but Lipworth said the severe testing shortages in March and April mean scientists still don’t have a clear picture of how various age groups were affected during the initial surge.

“Younger demographics tend to be the groups that are more likely to ease restrictions, which is one reason we’re seeing the distribution shift,” she said. “But we’re also testing more people now, and we’re testing more asymptomatic people, which is more heavily weighted to younger people.”

Dr. Jennifer Stevens, director of the Center for Healthcare Delivery Science at the Beth Israel Deaconess Medical Center in Boston, said understanding how behaviors change over the course of the pandemic, and subsequently what effect those changes have on transmissions, is a key part of refining local and regional models.

“That’s one whole bucket of work, understanding the choices that people are making and the implications of increased human interactions," Stevens said. "Ideally that will give us some lead time to prepare as health care workers.”

But finessing models based on behavioral changes is tricky work because researchers don’t have a good sense of what role milder restrictions have on slowing transmissions, according to Gerardin.

"We know that hitting things with a big hammer — like sheltering in place in March and April — will have a big effect, but we don’t know in great detail how different parts of the hammer were responsible for that effect," Gerardin said. "That means we don’t really know what effect different phases of reopening will have on transmissions."

Yet, even with the many unknowns that persist, scientists say having more uniform mask policies could go a long way in containing the virus within communities.

“If everybody in this country wore a mask,” Lipworth said, “we as epidemiologists could confidently say that we could make an impact and flatten these curves.”

Follow NBC HEALTH on Twitter & Facebook.