What If Herd Immunity Is Out of Reach?

Photo: Federic J. Brown/AFP via Getty Images

What if it never really ends, just recedes?

There are, at the moment, a number of encouraging signs about the near-term course of things: Caseloads and hospitalizations are falling dramatically, perhaps as a sign of seasonal effects turning a corner; vaccine deployment, while still suboptimal, is improved from a month ago; there has been good news about additional vaccines, with AstraZeneca (already approved in the U.K. but facing an FDA roadblock here) reporting fantastic results against severe disease; and vaccine shipments are said to on the way, with Novavax promising 100 million American doses by the spring.

A few months ago, these developments might have suggested the true endgame of the pandemic was in sight—and indeed the likely vaccination of 100 million or more by late spring does suggest a dramatic change in the country’s experience of the disease, with those vaccinated feeling safe from hospitalization and death and the disease in retreat. But thanks to a combination of higher herd-immunity estimates, stubbornly high vaccine “hesitancy,” and the arrival of new coronavirus variants that render existing vaccines less effective, the second year of the American pandemic is beginning to look less like a page-turning, book-slammed-shut bang and more like a long and indefinite whimpering into the future — in which many are protected but the disease, undefeated, still circulates, perhaps forever. That the coronavirus would become endemic, like the common cold, has always been one possible outcome, though less appealing than true elimination. The arrival of new variants has made that kind of near-term future, with enduring reservoirs of virus throughout the country, seem less appealing still.

This is, at base, a matter of the math of herd immunity — whether the vaccines we have, combined with the natural protection already acquired through disease exposure, can produce sufficient population-level immunity that the coronavirus actually dies away. Most of the vaccine trials were focused on reduction of severe disease, and so we do not yet have a clear sense of how effective they will be in stopping transmission—though most experts believe they will put a pretty big dent in caseloads. How big? In December, Harvard’s Marc Lipsitch estimated that the current bundle of vaccines would likely prove between 50 and 70 percent effective against transmission. What does that mean in terms of herd immunity? A sort of median estimate of the natural reproduction rate (or “R0”) for the classic COVID-19 strain is around 3 — on average, each person infected in a totally unexposed population would infect three others. Assuming an R0 of 3 yields an estimated herd-immunity threshold of 67 percent exposure, which, Bloomberg’s Justin Fox points out, would require between 96 percent and 134 percent of the population be vaccinated to achieve herd immunity. At the end of last year, Anthony Fauci somewhat controversially revised his own estimate of the threshold of herd immunity, first from between 60 and 70 percent to “70, 75 percent,” then to “75, 80, 85 percent,” then to “80-plus percent,” and then all the way to 90 percent. Mathematically, you simply cannot achieve 90 percent protection from a vaccine that offers even 70 percent protection, let alone 50 percent, and while the vaccines might somewhat outperform Lipsitch’s back-of-the-envelope calculations, the new variants are driving their efficacy in the other direction, pushing herd immunity even further out of reach. Perhaps the vaccines will surprise us, offering more protection against transmission than has been expected—and in this goal the country may be aided by the naturally lower rates of transmission among the very young. A bit more than a quarter of the country has already been exposed to the virus, which means we may already have 80 million or more Americans with protection. But that is only a true floor on which to add additional immunity through vaccination if none of those previously infected line up to get shots—and since the majority of them probably don’t even know they had the disease in the first place, and no effort is being made to target vaccine doses to the un-exposed, the contribution of those 80 million towards herd immunity is likely to be somewhat smaller than their numbers suggest.

Technically, of course, even vaccinating 96 percent of the country would be doable, though almost certainly it would take the U.S. into 2022 to achieve it. But the challenge grows steeper when you turn from the abstractions of math to the much messier world of human behavior. While it is certainly likely that vaccine resistance will gradually shrink as more and more Americans safely receive shots, with infection rates beginning to fall as a result, the scale of present-tense skepticism is an immense roadblock. According to a large Kaiser Family Foundation poll published in December, only 41 percent of Americans say they will “definitely” agree to be vaccinated, with another 30 percent saying they “probably” will, 12 percent saying “probably not,” and 15 percent saying “definitely not.” That gives a rough ceiling of 85 percent of the American population signaling at least some openness to vaccination, less than Fauci’s estimate of threshold of 90 percent needed for herd immunity (though, presumably, some of the already-infected could help close that gap). And of course certain groups are more skeptical than others — in rural America, Kaiser found, only 64 percent would definitely or probably take the vaccine and 35 percent probably or definitely wouldn’t. More recently, an eye-opening report by the CDC suggested that among those working at nursing homes and other long-term-care facilities, only 37 percent agreed to a first dose.

And then there are the variants, which seem already to be reducing the efficacy of our existing batch of vaccines — with presumably more variants to come. Already, what has been called the U.K. strain seems to have picked up a new mutation first observed in the South African and Brazilian strains, for instance, which may further reduce the effectiveness of vaccines and natural immune protection. And in most of the world, limited genetic surveillance of the disease means we are probably missing some new variants of the disease. Of course, it is important to remember that rendering existing vaccines less effective is not the same thing as rendering them ineffective—and that even with reduced efficacy, the vaccines we have do seem to offer meaningful population-level protection against the new variants. Already many of the manufacturers are working on retooling their products to offer additional protection against the new strains — indeed, this ability to tweak the existing vaccine platforms has been one big selling point of the new class of vaccines produced for the first time, with breathtaking speed, in this pandemic. But at least according to initial reports, those tweaked vaccines won’t be available until 2022 — almost a year from the arrival of this group of variants.

This is not to say that the brief window of vaccine optimism will yield only to a future of eternal, or even periodic, lockdown. Things are going to get a lot better, and probably pretty quickly. The vaccines are very effective in protecting against severe disease, at least when caused by the classic strain, which means that anyone who receives them will likely feel safe and protected from the scariest outcomes — again, at least against the classic strain. Herd immunity is not binary, which means that well before its threshold is reached, the spread of the disease will begin to slow, perhaps even precipitously — in fact, there is some hope that we are already beginning to see such population exposure bending the curve of cases downward. And some mathematical modelers have long argued that calculations of herd-immunity thresholds based simply on R0, like Fauci’s, are crude overestimates, overlooking the superspreader dynamics of the disease (whereby the vast majority of new infections are produced from a tiny minority of cases and the median sick person doesn’t infect anyone else at all). But many of them have been making those arguments since the beginning of the pandemic, suggesting that herd immunity was just around the corner in communities and countries that then saw large second and third waves — and in fact there have been terrible subsequent waves even in places, like Brazil’s Manaus, estimated to have had much higher levels of exposure (in the case of Manaus, exposure was estimated at 76 percent, in the vicinity of most estimated herd-immunity thresholds). Seasonal effects may help quite a lot, suppressing the disease for roughly three-quarters of the year.

But it does mean that what has long been the dream of most Americans enduring the pandemic — a point at which “all of this is over,” with COVID-19 as much a historical artifact for us as, say, SARS is in East Asia — may never come to pass precisely as imagined. Instead, in the medium term and perhaps even the long term, a likelier endgame is one in which large portions of the population are protected, from at least severe disease produced by at least some variants, but, with immunity falling short of the herd threshold, the disease continues to circulate — infecting even some of those who’ve been vaccinated, threatening the lives of those who haven’t, and continuing to evolve, perhaps in some scary ways. For most of those who’ve received a vaccine, the disease will fade into the background, joining the ranks of other endemic diseases, but as a social fact the coronavirus will nevertheless remain.

What will that world look like? In Denmark, they are already planning on issuing immunity passports, which would allow those who’ve been vaccinated to travel and socialize and do business in ways that others still can’t — a sort of immunity apartheid system of the kind that many warned about at the beginning of the pandemic but appears more and more difficult to avoid and that may ultimately persuade some number of vaccine skeptics to receive doses after all. There could also be harder measures taken, with companies requiring vaccine compliance from employees or even a national vaccine mandate — though each of these measures would run up against health-privacy protections in the U.S., where even masking regulations have proved difficult to enforce. The fate of those measures of vigilance is unclear, too — if less than half of nursing-home workers are vaccinated, will long-term-care residents and staff be required both to mask up and to social distance in an ongoing way, as they have for the last year, in circumstances described as solitary confinement, which may have already produced a mental-health crisis? Especially in places with low vaccination rates, schools may deploy temperature checks and rapid spit-tests — or may disregard those protocols and risk sporadic outbreaks. And any time a new strain arises, in the American reservoir or abroad, there may be renewed panic and vigilance — with even already-vaccinated people waiting those many months it will take to roll out a tweaked vaccine to feel truly safe again.

What If the COVID Pandemic Never Really Ends?