Just weeks into its staggering ascent in the United States, Omicron appears to maybe, maybe, be taking its leave of a few big urban centers up and down the East Coast. Documented coronavirus infections seem to be leveling off, even falling, in cities such as Boston, New York, and Washington, D.C.—a possible preview of what the country’s been waiting on tenterhooks for: the beginning of the end of the Omicron wave.
The pattern fits with what recent models predict. National case counts will hit a maximum this month, maybe a touch later. (Some think that the peak is already behind us.) It’s all a bit squishy still, but epidemiologists such as Justin Lessler of the University of North Carolina at Chapel Hill are “pretty confident” that the American apex is nigh. Peak could then give way to plunge, as it did in South Africa. It’s tempting, then, to imagine Omicron loosening its vice grip on the United States just as quickly as it latched on. February will be better; March, rosier still. Americans will get something like a Hot Post-Omi Spring.
A symmetrical, V-shaped rise and fall is a very nice and neat story. It is also probably wrong.
Before I stuff my foot completely inside my own mouth, let me be clear: This is not a Full-Blown Pandemic Prediction™. I personally do not know exactly what is on the other side of the Omicron peak. Neither do the experts. Actually, no one does. The back ends of curves can mirror the fronts, but they don’t have to—it depends on us and our immunity, on the virus and its hijinks, and on the frequency and intensity at which host and pathogen continue to collide. The decline could be sharp and fast, or sputtering and slow. It could start off steep, then lose steam. It could plateau—or even reverse course and tick back up.
What we can say is that the higher a wave crests, the longer and more confusing the path to the bottom will be. We need to prepare for the possibility that this wave could have an uncomfortably long tail—or at least a crooked one. “I do think the decline is unlikely to be as steep as the rise,” Saad Omer, an epidemiologist at Yale, told me.
During outbreaks, the only truly certain things are those “in hindsight,” Shweta Bansal, an infectious-disease modeler at Georgetown University, told me. And even the recent past is cloudy right now. We’ve lacked the test-and-trace infrastructure to fully track Omicron’s spread, which has seriously messed with our ability to forecast what the virus might do next. Most scientists are not even all that certain about where we stand in relation to the peak. And “the further into the future we want to project, the more uncertainty there is,” Lauren Ancel Meyers, the director of the University of Texas at Austin’s COVID-19 Modeling Consortium, told me.
Even with the murkiness ahead, how we exit this wave will certainly be affected by how we entered it. On its record-shattering sprint upward, Omicron had certain advantages: The virus seems to thrive in the upper airway and become contagious fast; it’s ace at dodging a lot of the antibodies in vaccinated and previously infected people, giving it a larger pool of hosts to work with than Delta. In the United States, Omicron also arrived at an especially opportune time: Americans, many of them older, unvaccinated, or with a chronic health condition, were sick of masking, and were barreling into their holiday-heavy winter. The fleet-footed virus slammed into a susceptible population that, behaviorally, was quite amenable to slathering it around. That dangerous combination spurred our wave, then skyrocketed it.
When this tide turns hinges on when Omicron starts to run out of new people to infect—either because it has burned through everyone it can or because we, through our behaviors, starve it of hosts. Cases crater; the curve, in turn, crashes. A version of this seems to have unfolded in South Africa, where recorded cases peaked around mid-December, then fell, and fell, and fell. (The United Kingdom, whose wave is a couple of weeks behind South Africa’s, seems poised to turn a corner too.)
How those foreign free falls play out is instructive, “but we also need to recognize that the U.S. is not South Africa,” Maia Majumder, a computational epidemiologist at Harvard, told me. Even subtle differences in host populations can massage a wave into a different shape—a rounder pinnacle, a more leisurely wane. Yes, the United States’ population is more vaccinated than South Africa’s, but it’s also older. (And lots of Americans over 65 aren’t boosted.) The two countries’ health profiles, medical infrastructures, and approaches to controlling SARS-CoV-2 differ; so do the behaviors of their residents. Omicron also caught South Africa as it was heading into summer; the United States may have a tougher time unsticking itself from the virus during colder months. And Delta, which was already driving surges of its own before Omicron arrived, hasn’t yet disappeared here.
The United States is also an especially sprawling and diverse place, as Samuel Scarpino of the Rockefeller Foundation’s Pandemic Prevention Institute pointed out on Twitter. Viruses thrive on human interconnectedness, and an early surge in big cities can front-load cases so much that the national narrative booms, then starts to bust. After we pass the summit, “I think at least the initial downslope will be precipitous,” Yonatan Grad, an infectious-disease expert at Harvard, told me. But as the virus continues to trickle into more rural, sparsely populated parts of the country, that story gets more complicated: a smattering of regional peaks could slow and lengthen the overall decline. We tend to talk about “the peak” as if it’s one monolithic thing, but it’s an aggregate of asynchronous outbreaks; each community will experience its own, unique Omicron spike, Grad said. The national trajectory depends heavily on “how long it takes to percolate into different parts of the country,” Natalie Dean, a biostatistician at Emory University, told me. A cliff-like drop might give way to a series of rolling hills. (Scarpino thinks that South Africa’s decline, which has recently slowed, may now be exhibiting this geographical flattening effect.)
How we react to the curve could also stretch it out, and that’s the biggest wild card of all. When people hear that we’ve skittered past the top of a peak, “psychologically, they loosen up,” UNC’s Lessler told me. (This is something that many epidemic models don’t account for.) Masks come off. Schools, workplaces, and leisure venues reopen. People rejoin social circles, or kick-start new ones. Smaller shifts such as these, multiplied by millions, can turn a waterfall decline into molasses. “So much of susceptibility is tied up in behavior,” Majumder said. And as people get further out from their most recent vaccination or infection, their risk of catching the virus goes back up.
A lethargic decline is a costly one. Already, health-care systems around the country are being pummeled by record-breaking cases. In many states, hospitals are hitting capacity; people are struggling to access care for all sorts of sicknesses. Hospitalization and death waves are smaller in magnitude than infection waves, and lag behind them, but they’re “much more protracted,” UT’s Meyers said. The sheer height of our infection peak is already poised to haunt us. There have been so many infections that cases, hospitalizations, and deaths won’t return to November’s pre-Omicron levels—let alone the numbers of last year’s early-summer lull—for a long time. “It’s going to get much worse before it gets better,” Meyers said. Even if the United States’ curve turns out to be symmetrical, half of this wave’s infections, and more than half of its hospitalizations and deaths, are still ahead, past the peak of cases. Adding any more weight to the curve’s far side just makes that picture uglier.
On the more optimistic flip side, behavior can also curb transmission—enough to keep the overall number of infections lower than it might otherwise be. We heard this lesson early on in the pandemic, when cases were first rising at alarming rates: mask up, hunker down, flatten the curve. It’s still true now. The hope is that the lower the peak, the fewer unnecessary infections can occur after it, Lessler said.
A horizontal squish does delay the peak and stretch out the wave. But it also buys us time to vaccinate more people and roll out treatments, and reduces the burden on the health-care system at any single point. We missed our chance for an early pancaking effect in many big cities, but smaller, rural parts of the country can still take heed, and it’s probably especially important that they do so. Those regions tend to have lower vaccination rates and lack “the capacity for a fast-running surge,” Anne Sosin, a public-health researcher at Dartmouth College, told me. If they’re not buffered from their own Omicron waves, the variant could concentrate in the parts of the country that can least afford to absorb it.
What lies beyond the peak isn’t out of our control either. The decline can be sped up by the same mitigation behaviors that temper the rise, Majumder said. Curves can get flatter. They can also get shorter. And minimizing cases on the wave’s far side will still blunt the impact on the health-care system, and lessen the variant’s social toll. The key here, then, is to avoid seeing “past the peak” as a cue to relapse into riskier behavior. “The start of a decline is not sufficient to think we’re out of the woods,” Georgetown’s Bansal said. Every step we take now will determine how long we stay high up on this curve and, eventually, where we land—as well as what condition we’ll be in when we arrive at the bottom.