What the heck is happening with COVID right now?
Amid the chaos and controversy that followed a federal judge’s decision to strike down the US mask mandate for travel earlier this week — a decision the Department of Justice is now appealing — it was easy to overlook the fact that mask policy wasn’t the only thing getting more perplexing by the day.
The pandemic itself has probably never been harder to parse.
On the one hand, cases have climbed in 41 states over the last two weeks, roused from their post-Omicron lows by an even more transmissible subvariant called BA.2.
On the other hand, BA.2 has been dominant in the US for about a month — and cases are rising about one-tenth as fast as they were in the first month of Omicron’s reign, increasing by 46% between March 20 and April 20 compared with 475% between Dec. 15 and Jan. 15.
Then again, the US is reporting only about 900,000 tests per day, on average — a fraction of the 2.5 million recorded at Omicron’s peak. Nor Americans switch to rapid at-home testing, or don’t bother to test at all, we’re undercounting infections more than before.
Still, hospitalizations recently bottomed out around 15,000 and have barely budged for weeks, while ICU numbers have plummeted to their lowest level since the beginning of the pandemic — another sharp contrast with winter, when hospitalizations more than doubled between Dec. 20 and Jan. 20 to an all-time high of 160,000.
This could be a sign, some experts say, that infection- and vaccine-induced immunity are blunting the latest variants’ most dire effects, a development they hope will usher in a less dangerous and disruptive phase of the pandemic.
Or, as other experts argue, it could just be the usual lag between cases and hospitalizations.
As we said: a bewildering moment. So rather than predict the future — never a wise move with the SARS-CoV-2 virus — here are three different ways this could play out in the weeks and months ahead.
The worst-case scenario
In recent days it’s become clear that BA.2, which has been slowly spreading across the US since December, is no longer the only subvariant in town. It’s also not the fittest. Instead, BA.2.12.1 — a subvariant of a subvariant — now appears to be outcompeting its predecessor in bellwether states such as New York and Massachusetts.
According to estimates of variant proportions from the Centers for Disease Control and Prevention, BA.2 has accounted for more than 70% of US infections since the beginning of April. But its growth has plateaued there. Meanwhile, BA.2.12.1 has surged from just 3.3% of infections in late March to 19% as of April 16.
In other words, BA.2.12.1’s share of cases has been roughly doubling every week — 2.5 times as fast as BA.2’s. This implies a significant “transmission advantage,” says Dr. Eric Topol, director of the Scripps Research Translational Institute — one that’s “similar in magnitude to the observed advantage of BA.2 over BA.1,” according to Trevor Bedford, leading virologist at Seattle’s Fred Hutchinson Center.
“The BA.2 wave in the United States is … morphing [in]to the BA2.12.1 wave,” Topol recently tweeted.
Why? Likely because BA.2.12.1 has an additional spike protein mutation (L452Q) that, in a slightly different form (L452R), “appeared to have an important role in promoting the spread of [the] Delta variant,” according to Bedford.
Meanwhile, the similarly mutated BA.4 and BA.5 sublineages are spreading rapidly in South Africa.
“The hypothesis is then that 452R/Q is conferring some additional intrinsic transmission advantage,” Bedford has explained. “Looking forward, I expect these 452R/Q sublineages to continue to expand.” The question is how much damage they do while they spread. The main reason BA.2 has moved so much more slowly than Omicron BA.1 in the US is that a huge percentage of the population — as much as half — just had Omicron BA.1, which confers a degree of immunity against reinfection. If BA.2.12.1 (and BA.4, BA.5) can sidestep some of that immunity, it could transform the current ripple of cases into more of a wave. And while vaccines and especially boosters have so far shown an extraordinary ability to shield recipients from severe illness and death, millions of seniors remain unvaccinated and/or unboosted; long COVID looms as a real concern; and it’s still unclear how much immunity wanes over time.
The worst-case scenario, then, is that these new sublineages exploit the United States’ persistent vulnerabilities — including a steady decline in masking — and trigger more serious illness than BA.2 alone would have.
The best-case scenario
The good news is that there’s no sign — yet — of anything in the US approaching the previous waves of hospitalization and death.
National numbers (which, again, have hit record lows) aren’t particularly useful here because BA.2 and BA.2.12.1 have spiked mostly in the Northeast. But the numbers from New York and Massachusetts (the leading BA.2.12.1 states) may be more illuminating.
Over the past two weeks, case counts have increased by 62% in New York and 51% in Massachusetts. Hospitalizations have increased as well. But crucially, ICU admissions have remained remarkably flat — and low. In New York, in fact, ICU numbers as a share of overall hospitalizations have (at 11%) never been lower. The same goes for Massachusetts (at 8%). Likewise, just 111 (or 29%) of the state’s nearly 400 hospitalized patients have been classified as “primarily hospitalized for COVID-19 related illness”; at Omicron’s peak, that number was above 50%. Another telling statistic: A full 62% of patients in Massachusetts currently hospitalized with COVID report having been fully vaccinated when they tested positive; before Omicron, that number was consistently closer to one-third.
For now, then, it seems that more patients in bellwether states are testing positive for COVID upon admission to the hospital because more people are testing positive for COVID in the surrounding communities — but BA.2 and BA.2.12.1 aren’t driving upticks in serious illness, probably thanks to vaccination.
Whether this pattern holds as BA.2.12.1 increases in prevalence remains to be seen. But according to the CDC, the new subvariant already accounts for more than 50% of infections in the New York region — and rates of case growth and test positivity may be starting to level off nonetheless. In Massachusetts, wastewater prevalence and cases both appear to have plateaued; meanwhile, cases in Washington, DC, may be starting to decline.
Emerging relatively unscathed from a BA.2/BA.2.12.1 bump could be a good sign for the future.
According to Bedfordthere are two plausible scenarios for the next year: (1) another “Omicron-like emergence event” in which a “new wildly divergent virus” evades existing immunity and upends society all over again, or (2) “evolution within BA. 2” to “further increase intrinsic transmission,” causing “lower attack rates,” largely “driven by drift + waning [immunity] + seasonality.”
Bedford considers the second, steady, flulike scenario “more likely” — and says that the more sublineages like BA.2.12.1 emerge, the more likely it will become.
“The more time that passes, the more confident we can be that another ‘Omicron-like’ emergence won’t occur,” he has explained.
If Bedford is right, that suggests future surges would look less like our huge winter Omicron wave than like whatever we’re experiencing now. And it would also increase the “likelihood we eventually switch to [a] wax with [an] Omicron backbone,” according to former Food and Drug Administration Commissioner Scott Gottlieb — meaning greater stability, predictability and protection against infection in the future.
The most likely scenario
Possibly somewhere in the middle. BA.2.12.1 accounts for only 20% of current infections in the Massachusetts region; apparently it takes off and undoes recent progress there. Or perhaps high vaccination and booster rates across the Northeast dull the variant’s ultimate impact there — while lower rates leave less COVID-cautious states more vulnerable this spring. Or maybe seasonality and warming weather help shield the South and West the way they did last spring and summer.
On Twitter, a respected COVID modeler who goes by the handle @JPWeiland posted a prediction earlier this week. Based on recent BA.2.12.1 growth rates, the modeler said the variant has “a more significant chance to break through the shift in seasons to create a real wave than BA.2 could muster,” but went on to say that because there is not a lot of anticipated immune escape with B.2.12.1, the wave should be inherently limited in size. “We won’t see close to the 1M cases/day in January, but we could see 200k. Hard to tell for sure.”
Indeed it is.