Coronavirus October 2021—Part 15 Should You Get a Third Booster Shot?

In this post, we explore the pros and cons of getting a third booster shot (or second booster shot if you got the J&J vaccine) against COVID-19. As usual, if you’re less interested in how we got to our conclusions than you are in the conclusions themselves, feel free to skip to the BOTTOM LINE in each section and the CONCLUSION at the end.

Question: Should you get a third booster shot?

Answer: It depends on how likely you are to have a bad outcome if you contract COVID-19 as well as your specific goals in getting vaccinated.

Estimates of Continuing Vaccine Effectiveness

As the pandemic has continued, different studies have estimated different levels of continuing effectiveness of the vaccines against COVID-19 infection. Final results of the Moderna Phase III trial on its vaccine’s effectiveness up to March 26, 2021 showed an effectiveness at preventing COVID-19 infection of 93.2 percent and an effectiveness at preventing severe COVID-19 infection of 98.2 percent. However, the Delta variant didn’t become the predominant variant in the U.S. until after the period this paper analyzed, so its results might not be generalizable to the situation we’re now facing.

An Israeli study estimated the effectiveness of the Pfizer vaccine at preventing infection to have fallen to 50 – 58 percent (depending on your age) seven months out from vaccination and its effectiveness at preventing severe infection to be 86 – 94 percent (depending on your age) seven months out from vaccination. A Kaiser study likewise found Pfizer’s effectiveness at preventing infection with Delta had dropped to 47 percent five months out from vaccination but remained 94 percent effective at preventing hospital admissions six months out from vaccination. A study out of New York from the CDC looking at vaccine effectiveness for all three vaccines authorized in the U.S. (Pfizer, Moderna, and J&J) showed the effectiveness against COVID-19 infection declined from May 2021 to July 2021 from 91.7 percent to 79.8 percent. However, effectiveness against severe infection requiring hospitalization did not decline over that period, remaining around 95.3 percent. A study from the U.K. showed Pfizer’s effectiveness at preventing infection with Delta was 88 percent. Differences in study methodology and rollout of vaccines (in the U.K. for example, the interval between Pfizer doses was longer than the 3-week interval used in Israel, an adjustment now known to improve vaccine effectiveness) likely account for the differences in the decline of vaccine effectiveness between these studies.

Why has vaccine effectiveness at preventing infection with COVID-19 declined with time but its effectiveness at preventing severe illness and death has not? First, Delta, which is now overwhelmingly the predominant strain circulating in the U.S., may be about 50 percent more contagious than the original variant. It may have roughly 1000 times the concentration in an infected person’s nose and mouth than the original variant, so that when an infected person breaths, sneezes, or coughs on you, they’re delivering a far higher dose of virus and are therefore more likely to transmit the infection even if you’re fully vaccinated. If the levels of antibodies in your nose and mouth induced by your original vaccine regimen remain high enough, they can prevent an infection from ever gaining a foothold (even against the Delta variant—see below for why). But if they’ve declined below a certain point—the point at which they’re no longer concentrated enough to prevent an infection from gaining a foothold (a level that isn’t yet known)—you will develop an infection. And antibody levels always decline after infection or immunization (they need to: if they didn’t, blood would become too viscous to sustain life).

But—and this is a crucially important point—after either natural infection or immunization, memory B cells and T cells become ready to fight re-infection, likely indefinitely. One amazing study in 2008 looked at the blood of survivors of the 1918 flu pandemic, ages 91 – 101, and found that when their memory B cells were exposed to the 1918 flu strain, they generated neutralizing antibodies in a test tube, which then protected mice that were exposed the strain! Other studies—here and here—have shown that even variants are unable to escape T cells immunized against the original strain of SARS-CoV-2. In other words, for T cells, recognizing a SARS-CoV-2 variant is as easy recognizing someone you know even though they’ve changed their hairstyle or grown a mustache. When a previously-infected or immunized person does re-encounter COVID-19, it takes 3 – 5 days for their memory B cells and T cells to leap into action, likely explaining why mild breakthrough infections with COVID-19 happen but severe disease is rare. Finally, as time progresses, B cells learn to produce antibodies of better and better quality, so that even as your antibody levels decline, the antibodies that are left circulating and that are ramped up when you encounter a potentially infectious dose of virus are more effective at warding off infection. In fact, we’ve found that even though no one has been vaccinated against the Delta variant, your B cells eventually learn to make antibodies against it!

Bottom line: There seems to be little question that vaccine effectiveness at preventing infection wanes with time. How much it wanes, however, isn’t clear. It is clear, however, that vaccine effectiveness at preventing severe infection requiring hospitalization has not waned.

What Does Waning Effectiveness Mean in the Real World?

As we wrote previously, you can’t just look at a vaccine’s effectiveness and arrive at your real risk of contracting COVID-19 or being hospitalized from it. You have to start with the base rate of risk—the absolute risk—which varies from context to context. The base rate of risk of catching COVID-19 (if you’re unvaccinated) from someone who is asymptomatically infected who you pass by on the street for only seconds is as close to zero as you can get, while the base rate of risk of contracting COVID-19 from an infected spouse runs about as high as you can get at about 60 percent. To calculate your new absolute risk of contracting COVID-19 from your infected spouse if you’ve been vaccinated would require you to reduce the base rate of risk—60 percent—by the vaccine’s effectiveness at preventing infection, which, depending on how far in the past you were vaccinated, could be as low at 47 percent. This would mean at worst that your new absolute risk of contracting COVID-19 from your infected spouse would be 28.2 percent (compare to the new absolute risk if you got a third booster and enjoyed a 95 percent reduction in risk: 3 percent). In most other contexts in which you find yourself, however, the differences in absolute risk between not being boosted and being boosted would be narrower. For example, the risk of contracting COVID-19 from an ill family member who isn’t your spouse is somewhere around 26 percent (from older contract tracing studies, with a 50 percent increase added in for Delta’s extra infectiousness). A 47 percent reduction from vaccine-induced immunity would yield a new absolute risk of becoming infected to 13.8 percent. A 95 percent reduction from a booster shot would yield a new absolute risk of 1.3 percent.

We can calculate the risk of developing a severe case of COVID-19 in much the same manner. The differences that determine the base rate of risk aren’t, however, daily life circumstances but rather the risk factors for severe COVID-19: age, gender, and the presence of comorbid conditions like obesity, diabetes, immunocompromising states, and so on. These vary hugely, the highest risk coming from having severe kidney disease, which puts the risk of dying if you contract COVID-19 at 1.1 percent. Reduce that by 95 percent and you get an absolute risk of dying from COVID-19 if you have kidney failure and are vaccinated at 0.05 percent. But this represents a reduction in absolute risk of only 1.05 percent.

Bottom line: In the riskiest context, having been fully vaccinated without a 3rd booster leaves you at a mildly increased risk for contracting COVID-19 (the magnitude varying with your level of exposure) but not at an increased risk for severe COVID-19 or for dying from it.

Benefits of a Third Shot

The Israeli study above showed that the group of people who received a third Pfizer booster shot had an 11.3-fold reduction in the rate of COVID-19 infection compared to the group of people who only received the original two-shot series. However, the absolute difference in number of cases was only 86.6 per 100,000 person-days. The rate of severe illness was lower in the third booster shot group by a factor of 19.5—but again, the absolute difference in number of cases was low at only 7.5 per 100,000 person-days. This represents a vaccine effectiveness against infection of 95 percent—similar to the level achieved initially from the two-shot series. However, it only increases vaccine effectiveness against severe infection from 94 percent from the two-shot series to 95.3 percent with the third booster shot (calculation not shown).

A Phase III trial from Pfizer also showed a 95.6 percent efficacy in preventing infection when given on average 11 months after the second shot during a period where Delta was the prevalent strain (reported in the media only; trial data not available for review at the time of this writing). According to the press release, “In a trial with more than 10,000 participants 16 year of age and older, COVID-19 booster was found to have a favorable safety profile.”

Bottom line: A third booster of Pfizer will decrease your likelihood of contracting COVID-19 by a modest amount but will not significantly decrease the likelihood of severe infection because that’s already quite low from the original two-shot series. Also, the bump in protection will probably last only six to nine months (though we won’t know that for certain until enough time passes for us to observe it).

Risks of a Third Shot

The Israeli study did not assess for side effects. The Phase III Pfizer study did, however, and reported a “favorable safety profile” (though, again, the data isn’t available for review at the time of this writing). If you had a severe side effect to one or both of the first two (myocarditis, clotting), you should not get the third. Otherwise, a third shot is likely as safe as the first two.

What Third Shot Should You Get?

Data suggest mixing and matching vaccines may be slightly more effective than getting the same vaccine you had for your first series. One study looked at immunologic outcomes (that is, antibody responses) not real-world immunity, when people initially given the AstraZeneca vaccine were boosted with either Pfizer or Moderna vaccines. They found that the resultant antibody levels after a booster shot were higher if the second shot was Pfizer or Moderna instead of AstraZeneca. Another study of people who received one of the three available vaccines in the U.S. (Pfizer, Moderna, or J&J) and looked at antibody responses (again, not real-world immunity) in people who received additional booster shots that were the same as the first shots they were given (homologous boosting) in comparison to people who received additional booster shots that were different from the first shots they were given (heterologous boosting). Side effects were similar in frequency as reported for the primary series and not severe. Homologous boosting increased antibody levels 4.2 – 20-fold, but heterologous boosting (mixing and matching boosters) increased antibody levels 6.2 – 76-fold. Regardless of which primary vaccination subjects received (Pfizer, Moderna, or J&J), Moderna boosted antibody levels the most, followed closely by Pfizer and lastly by J&J. This suggests, but does not prove, that immunity might be better if you mix-and-match booster shots.

CONCLUSION: Should you get the third shot? It depends on your risk of a bad outcome if you contract COVID-19, your personal goals for vaccination, and your wariness about experiencing possible side effects. If you are significantly immunocompromised due to a disease or a medication, you should get a third shot given that a significant number of people in this category will not have developed antibodies at all to an initial vaccine series and are likely to have a moderate response to a third shot.

If you’re not immunocompromised but your goal is to reduce your risk of getting COVID-19 to the lowest possible level, then a third booster shot will do that (at least, temporarily). If your goal is to protect yourself against severe COVID-19, hospitalization, and death, a third booster shot likely adds little extra protection (as of this writing, according to CDC data, if you’re over 65—meaning in the highest risk group—and have been vaccinated against COVID-19 without a subsequent booster shot, your risk of dying if you contract COVID-19 is 0.004 percent). If you’ve had a natural infection before getting vaccinated, your risk of contracting COVID-19 is probably lower than if you just had an infection or were just immunized and you probably have even less to gain from a third booster shot (likely true as well if you were vaccinated first and then had a breakthrough infection, though we have no data as yet to prove this). If you are going to get a third shot, we’d recommend Moderna (which seems to bump antibodies the most), followed by Pfizer, followed by J&J.

  1. Coronavirus February 2020—Part 1 What We Know So Far
  2. Coronavirus March 2020—Part 2 Measures to Protect Yourself
  3. Supporting Employee Health During the Coronavirus Pandemic
  4. Coronavirus March 2020—Part 3 Symptoms and Risks
  5. Coronavirus March 2020—Part 4 The Truth about Hydroxychloroquine
  6. Coronavirus April 2020—Part 5 The Real Risk of Death
  7. Coronavirus April 2020—Part 6 Evaluating Diagnostic Tests
  8. Coronavirus April 2020—Part 7 The Accuracy of Our Antibody Test
  9. Coronavirus May 2020—Part 8 How to Reopen a Business Safely
  10. Coronavirus August 2020—Part 9 Masks, Vaccines, and Rapid Testing
  11. Coronavirus December 2020—Part 10 Should You Get the Pfizer Vaccine?
  12. Coronavirus December 2020—Part 11 Should You Get the Moderna Vaccine?
  13. Coronavirus April 2021—Part 12 Should You Get the Johnson & Johnson Vaccine?
  14. Coronavirus May 2021—Part 13 How Effective are the Vaccines in the Real World?
  15. Coronavirus August 2021—Part 14 The Delta Variant, Masks, and Vaccines

Sign up to get notified when a new blog post has been published.

Leave a Reply

  • Alex, great post. What about if you want to reduce odds of getting long haul COVID. Some studies I believe show that the long COVID instances don’t correlate with the severity of the initial infection. If this isn’t the case, the question is moot. If it is or might be the case, wouldn’t that warrant another section in your post, namely, reason to get booster to avoid a non-severe case, in attempt to reduce risk of long COVID. Thanks.

    Rob: The evidence does indeed indicate that the risk of long-COVID doesn’t correlate with the severity of symptoms, so it is indeed possible to get long-COVID with even mild infection. However, the two-shot series seems to reduce this likelihood by 50 percent. Because the rate of long-COVID seems to be around 20-30 percent, vaccination with the original regimens reduces the absolute risk of long-COVID to 10 – 15 percent. There is no evidence yet that a third booster (or second if you got J&J) further reduces the risk of long-COVID.

  • Excellent, thank you. I have had a 3rd Moderna, after initial 2 seven months ago. No side effects worth mentioning. I’m 93 with COPD, otherwise healthy. Was curious that so little explanation of age-related immune system insufficiency occurred. Without COPD, I fit into no category for a booster vaccination, not having a transplant, diabetes or any other immune suppressions listed. It’s academic now, since I got recently a 3rd shot, but the almost total omission of attention to lazy immune systems in the aged qualifying one for a booster puzzled me. Of course I don’t know if any of the 3 vaccinations got the attention of my old immune system! I am assuming so.

    Does strength or discomfort of immediate post-vaccination symptoms indicate enhanced or not stimulation of one’s immune system? Did the shot “take” better if one has a miserable post shot reaction? Just an idle question.

    Thank you. You are the best explainer of all, except maybe Ed Yong at The Atlantic.

    Alexandra: Post-vaccination symptoms have no relationship whatsoever with immunity generated. And thanks for the compliment!

  • Very informative, thank you.

    I got the Moderna booster motivated partly by the hope that I would be less infectious, or infectious for a shorter time, if I got a breakthrough infection. Is there any science on that yet?

    Iohannes: We know being vaccinated with the original regimen will decrease your infectiousness if you get a breakthrough infection (you’re as infectious on the first day of symptoms as someone not vaccinated, but infectiousness rapidly falls off in comparison to non-immunized individuals after that). We have no data on how third boosters affect infectiousness, if they reduce it even further.

  • Great post. Doctor, I’m curious if you have any thoughts on the guidance by the CDC for people in high risk areas/high risk jobs? For example, if people are traveling often for work (on planes) and attending large conferences or gatherings with hundreds of people, should they consider getting a booster to help prevent symptomatic infection? Thanks again for the great notes throughout the pandemic.

    Gino: Your risk of contracting COVID-19 will be higher if you encounter large groups of people on a regular basis rather than work mostly from home and encounter few. It’s difficult, however, to quantitate that risk. Our own view is that if you’re at low risk for an adverse outcome (e.g., you don’t have hypertension, kidney disease, obesity, and are younger than 65), a booster isn’t necessary. But again it really becomes a value judgment each person must make for him- or herself.