A: No. Getting vaccinated is by far the safest way to develop immunity to SARS-CoV-2.
If you’ve already been infected, getting vaccinated still provides you with the best protection-in fact likely *better* than either previous infection or the vaccine alone.
A recent study from Israel created waves with findings that people with previous COVID-19 infection had a lower risk of a new infection than fully vaccinated individuals.
We’ll get to the study details below, but most troubling was those who took this study as evidence that “natural infection is better”…and thus a reason *not* to get vaccinated. This Nerdy Girl (Jenn) even got a media inquiry to get my opinion on purposefully infecting people with COVID akin to chickenpox parties (HELL NO!).
If we squint *really* hard, we can see how one might get there, but this is a REALLY BAD TAKE.
Remember, having a re-infection is *conditional* on having survived your original infection. It also means you’ve already suffered from the disease itself, potential hospitalization and Long Covid.
💥Preferring natural infection to vaccination is like saying you want to become immune to your thumbs being cut off a second time by cutting them off yourself.💥
❓What did this study find?
The data came from a patient database from Israel’s 2nd largest Health Maintenance Organization. The authors compared infection risk from June 1 to August 15th in people who had been previously infected, those previously infected with one vaccine dose, and those NOT previously infected with two doses of vaccine. All vaccination in Israel was with the Pfizer/BioNTech vaccine.
The study found:
➡️ Fully vaccinated people had 6-13X the odds of infection compared to previously infected but unvaccinated people, depending on the model used.
➡️ The risk for previously infected patients with one dose of the vaccine was the lowest of all.
Since this was a real life (or “observational”) study rather than a randomized clinical trial, there could be important differences between these groups that would bias the findings. For example, the most vulnerable people may have worked hard to shield themselves from COVID prior to getting vaccinated. If the frailest were less likely to have been previously infected, this would make it look like natural infection was more protective than it really is.
These types of biases are difficult to overcome, but the authors used a technique called “matching” so that each previously infected patient is compared to a vaccinated patient of similar age, sex, and region of residence. This doesn’t magically fix all these problems, as the groups may still differ in ways that aren’t accounted for such as underlying conditions or preventive behaviors like distancing and masking.
A related challenge is “survival bias.” To be in the previously infected group, one must have *survived* their original infection. In the vaccinated-only group, those who are most vulnerable and may have died from an unvaccinated covid infection are still alive and susceptible.
Finally, and this is a big one—the infections are based on voluntary rather than routine testing. This means that if previously infected patients are less likely to get tested, their re-infections won’t show up in the medical records and it looks like they are not getting infected.
Overall, the study is intriguing, but the research design and analysis leave a lot to be desired. The study is somewhat inconsistent with other data that we have on this question.
➡️ A recent well-designed study from the UK found that the two doses of Pfizer/BioNTech or Astrazeneca vaccine protected against infection *as well or better* than previous infection.
This data was from the large ONS Coronavirus Infection Study, which is representative of the UK population and randomly tests people at regularly. Among other advantages this avoids the selection bias due to voluntary testing in the Israeli study.
➡️ A recent CDC study from Kentucky found that AMONG those previously infected, those who were not vaccinated had >2X times the odds of re-infection compared with those who were fully vaccinated. This study does not compare re-infection to the vaccinated-only group, but it confirms that previously infected individuals benefit from vaccination. These data also suffer from several limitations including a small sample (246 infections) and voluntary testing.
We also have many immunological studies showing stronger and more consistent antibody response with vaccination compared to previous infection (See example here.).
➡️ This immunological data also shows that previous infection + vaccination is the most powerful immune combination.
The Israeli data does contribute to evidence that only one vaccine dose might be needed for protection in those previously infected. Some countries including Israel and France are taking this approach, whereas others such as the UK and US do not seem to have seriously considered it (don’t ask us, we are not sure why!).
➡️ Previous infection DOES provide good protection against re-infection. That’s great news given how many infections there have been worldwide.
➡️ That does NOT mean getting your *initial* immunity naturally is better. It means you were lucky enough to survive, but at great risk.
➡️ Vaccination allows you to get your immunity while minimizing your risk of severe disease or death.
➡️ Natural infection + vaccination is looking like the strongest protection of all.
💥 If you’ve never had COVID— get vaccinated.
💥 If you’ve had COVID– get vaccinated and boost your protection.
Those Nerdy Girls
UK ONS Coronavirus Infection Study
Kentucky re-infection study from CDC/MMWR
Single-dose for previously infected
The power of “hybrid” immunity (infection +vaccination)
Previous DP post on natural vs. vaccine immunity
Credit for “cutting of thumbs” analogy as seen on twitter