Skip to Main Content

As Covid-19 vaccines are being rolled out across the U.S., Americans seem to be heaving a collective sigh of relief.

Yes, it will take months to get the vaccine to everyone. Yes, there were tremendous gaps in the Trump administration’s plans to distribute the vaccines, including promising doses that didn’t exist.

But it seems as if there is light at the end of the tunnel. As long as we maintain social distancing, keep wearing masks, and washing our hands, it feels to many as though we can hold on until we get vaccinated.


I’m sorry to be writing the words that follow, but here they are: We can’t vaccinate our way out of this pandemic. And the myopic focus on achieving herd immunity through mass vaccination may even make it tougher for America — and the world — to defeat Covid-19.

Don’t get me wrong: Mass vaccination is essential. But herd immunity is a numbers game. It is defined as the point at which community spread of a disease stops because unprotected individuals are surrounded by a “herd” of people who are immune to infection, making it difficult, if not impossible, for infected people to pass on the disease.


Many experts have said we will achieve herd immunity when about 70% of the population is immune to SARS-CoV-2, the virus that causes Covid-19, either through vaccination or by having had Covid-19.

How do we reach that number?

It’s harder than it seems. For starters, while the Pfizer/BioNTech and Moderna vaccines showed about 95% efficacy in the clinical trials, vaccine effectiveness — how well a vaccine performs under real-world conditions — is likely to be lower for several reasons. One is that the people who participate in clinical trials are an imperfect representation of the whole population. They tend to be healthier, and younger. Real-world factors such as vaccine transportation and storage can also reduce vaccine effectiveness.

Say the Moderna and Pfizer vaccines now being given across the country achieve 90% effectiveness. Vaccinating 70% of U.S. residents puts us at 63% immunity. So, we’ll need to vaccinate a full 80% of the population to reach the herd immunity threshold.

Additional vaccines are starting to be approved. Some of them have lower efficacy. For instance, the AstraZeneca vaccine has about 70% efficacy, and Johnson & Johnson has reported that its one-dose vaccine has 66% efficacy. Their real-world performance could be lower still. If these vaccines become part of the mix in the U.S., actual protection will be lower than the estimated 90% we’d get from just the Moderna and Pfizer vaccines.

There are other barriers to achieving herd immunity. Vaccine uptake — how many people actually get vaccinated — is far below the level we need, in part because Covid-19 beliefs have been politicized in the U.S. and a percentage of the population doesn’t even believe the disease is real. In a Kaiser Health News survey released near the end of January, 13% of Americans said they would “definitely not” get vaccinated, 7% would take the vaccine only if it was “required,” and another 31% would “wait and see how it’s working” before getting vaccinated. Not encouraging numbers for those hoping for a quick journey to herd immunity.

Even when ample vaccine supplies are restored — perhaps by President Biden invoking the Defense Production Act — other factors will further drive down the number of people who get vaccinated. Eligibility factors currently exclude approximately 25% of U.S. residents from Covid-19 vaccination. The Pfizer vaccine can be administered only to those age 16 and up; for the Moderna vaccine, it’s those 18 and up. This represents approximately 20% of the population. Furthermore, although the CDC says that pregnant people may get vaccinated, it stops short of a clear recommendation. The decision is a “personal choice” left up to individuals and their health care providers.

Excluding those currently ineligible for vaccination against SARS-CoV-2 due to age or other conditions leaves 75% of Americans with no restrictions on vaccination. Factoring in the 13% of Americans who definitely don’t want the vaccine and the 7% who would get it only if it was required means just 49.5% of Americans would have immunity in the near future. If half of those who are in a wait-and-see mode don’t get vaccinated — another 15% of the population — then we are looking at just 40% vaccine coverage of the currently eligible population, far below the 70% needed for herd immunity. And that’s even before considering that real-world vaccine effectiveness will be below clinical trial levels.

The young people who aren’t cleared to get the Moderna and Pfizer vaccines have proven to be highly efficient asymptomatic spreaders of Covid-19. Leaving this population unprotected will enable the disease to continue to spread widely.

Finally, we don’t yet know the durability of the immune response to the various vaccines. It may persist. Or it may wear off, leaving people vulnerable after they’ve been vaccinated and creating conditions for new outbreaks.

If my years of global health work on the HIV/AIDS epidemic has taught me anything, it’s that even the best laid plans can’t anticipate every challenge. To vaccinate 75% of the U.S. population, approximately 248 million people — that’s nearly 500 million doses — are needed. And it means we need to be vaccinating nearly 2 million people a day so all of them are immune by the fall of 2021. As I write this, we’re vaccinating only about 1 million people a day. At that pace, Reuters estimates it would take until April 2022 for 75% of Americans to receive at least their first vaccine dose.

And that’s only if everything goes well logistically (it won’t) and if there are no further mutations in SARS-CoV-2 that make combating it more difficult (there will be).

It’s time to stop promoting the myopic belief that the unrealistic goal of herd immunity can be achieved in 2021 and start looking to reinforcing all aspects of the health care response as we start to concede that Covid-19 will become an endemic disease that will continue to lurk in the population. For the foreseeable future, that means continued physical distancing; occupancy limits in restaurants and other retail establishments; replacement of physical menus with smart phone-based menus to prevent surface spread of the virus, and more.

We’ll also need to monitor people who have been vaccinated to gauge the durability of the immune system’s response and whether booster shots are necessary, as they are for tetanus and diphtheria. Finally, our nation’s public health infrastructure will need to be bolstered, putting in place new protocols to monitor for new variants of the virus as soon as they emerge.

Can we defeat Covid-19? We can and we will. But setting sights on a near-term goal of achieving herd immunity ignores the math that governs the spread of disease. That approach is going to take a while. To get past Covid-19, we need to use all the tools available.

Iain MacLeod is the co-founder and CEO of Aldatu Biosciences of Watertown, Mass., which develops novel viral diagnostics, including those for pathogens such as SARS-CoV-2, and a research associate at the Harvard T.H. Chan School of Public Health.

  • You are seeming to ignore the fact that efficacy is related to both minor symptoms and severe symptoms. The lower efficacy is so because it takes into account fevers and minor symptoms. For severe symptoms all of those you noted have over 90% efficacy for stopping hospitalizations.

  • Reference for this statement, please, “The young people who aren’t cleared to get the Moderna and Pfizer vaccines have proven to be highly efficient asymptomatic spreaders of Covid-19. “?

  • I have been trying to tell my friends such things, but some respond with the ignorant, “Lets get together for a show in June.” I appreciate you laying everything out so succinctly.

  • The author notes that pediatrics are not eligible for the vaccine right now, but that will change soon — it is being tested in those populations.

    He also seems to think that vaccinating 2 million people per day would be difficult, but we have reached over 1 million without straining the system. Sure it will take a while to ramp up, but there is no reason to think it is not possible. The limiting factor is vaccine supply, but vastly more vaccine will be available by June.

    And even if we don’t achieve true herd immunity, if we are able to get 60% or so of the population vaccinated (including the most vulnerable) plus those who have had the disease, the spread will be limited and will not strain our health care system.

    If we end up in a situation in the fall where a bunch of 20 and 30 somethings are spreading the disease amongst themselves but not getting seriously ill, then we are in good shape.

  • good job. you just increased vax hesitancy with loads of wild speculation and increased people’s despair – let’s just not get the vax, give up, tired of being isolated. do you even think about what you’re saying when you write? are you in communities at all, seeing how people are actually living?

  • Unfortunately, this article makes numerous speculative claims while providing little evidence:

    1) The author fails to point out that the CDC estimates that actual infections are up to 7 times higher than the reported numbers. Therefore, if 27 million people have tested positive, close to 189 million people may have already contracted Covid-19, meaning that there was already substantial immunity in the population even before mass vaccination started. (Other estimates of current levels of immunity are lower but still relatively high; see for a detailed discussion).

    2) In places like the Dakotas, where there have been very few restrictions, and the primary driver of a reduction in transmission is likely to be population immunity, daily new cases have been declining since mid-November. Indeed, in North Dakota, cases have fallen by a whopping 95% in less than three months. In the US as a whole, the effective reproduction number (R) has been less than 1 since mid-January, with a decline in new cases of over 60% since the peak.

    3) The estimate that 60% of the population must be immune for herd immunity is based on the assumption that R is 2.5, in which case 1-(1/2.5) of the population must be immune for R to decline to 1. However, since April, R has been very close to 1 in the US (see Rt chart on this page: When cases have gone up, R has been slightly above 1; when they’ve gone down, R has been slightly below 1. Even a small level of additional immunity can bring R from e.g. 1.2 to under 1 (an additonal 1-1/1.2 of the susceptible population would need to be immune, or approximately 17%). Of course, R may be close to 1 partly because of the precautions that are currently in place. However, the decline in the Dakotas, discussed above, suggests that the low value of R may be largely a result of population immunity.

    • Thanks so much for the detailed science in your response! Even a lay person can see how completely illogical this article is and how it misses all the relevant variables and makes unsupported statements

    • Thanks for your insightful thoughts, I really appreciate them.

      Population immunity (in your example, a natural form of herd immunity) was clearly a factor in driving down the rate of new infections in the Dakotas so dramatically. After such a huge surge in positivity rate, the pool of susceptible individuals dropped dramatically thus limiting the spread of SARS-CoV-2. The question is: what is the durability of that natural immunity? Can we rely on it? I’d love to say that we can. Correlates of protection against SARS-CoV-2 infection have yet to be fully understood. Yes, the data look promising ( although there are a growing number of reports of re-infection with SARS-CoV- associated with an insufficient antibody response during the first infection. And natural immunity might not be the most robust: there are several viruses (e.g. HPV and VZV) for which vaccination generates stronger immune responses and more-effective protection against disease than natural infection.

      At this point – given what we know about humans’ response to other coronaviruses – the protection offered by natural immunity is not a certainty. How many times have we all had the common cold? Can we really take the risk of not vaccinating those that have been previously infected with SARS-CoV2 without knowing the *quality* of the immune response that they harbour? My argument is that we can’t hinge our entire strategy on a single public health approach, be it natural or vaccine-derived immunity. To your point – which is a great one – we do need to figure out a way to factor in natural immunity into vaccine rollout, and the CDC alludes to this in their most recent interim guidelines ( “(t)hus, while vaccine supply remains limited, persons with recent documented acute SARS-CoV-2 infection may choose to temporarily delay vaccination…”. This comes with the important caveat, “recognizing that the risk of reinfection, and therefore the need for vaccination, might increase with time following initial infection”.

      An ideal complement would be an antibody (or other immunological assay) to gauge both the strength and quality an individual’s existing immune response and to allow public health efforts to be focussed appropriately (and no, my company does *not* develop antibody or other immunological tests). Although this would’ve been ideal before vaccine rollout, it’s still not too late. Just like vaccinations for other pathogens, we may need booster shots in the future, which is why investment in multiple public health approaches will help steer us out of than pandemic rather than relying solely on natural or vaccine-derived immunity, the durability of which is still unknown.

Comments are closed.