A masked woman makes a heart symbol with her hands.
Enlarge / Dr. Silvia Gelvez gestures after receiving a dose of the Pfizer-BioNTech vaccine against COVID-19, at Colombia University Clinic in Bogota on February 18, 2021.

As long as there have been vaccines against COVID-19, there have been arguments for why people shouldn’t get those vaccines. One of the more persistent—and hairier—arguments is that people who have already been infected with the pandemic coronavirus, SARS-CoV-2, don’t need a vaccine. An infection will generate immune responses similar to those generated by vaccines, the thinking goes. So, why waste coveted vaccine doses on people who already have immune responses against the virus—which may also needlessly put those people at risk of vaccine side effects, however rare?

It’s a reasonable question, and there is legitimate scientific debate about it. There are also different approaches to the issue in terms of public health policy. In Israel, for example, people who have recovered from COVID-19 after testing positive on a PCR test can get a vaccination “Green Pass” that’s valid for up to six months. The pass allows them entry into various places just as it does for people who are fully vaccinated. In the European Union, some member states offer a similar “Digital COVID Certificate” to people who have recovered from COVID-19 and received just one dose of a two-dose mRNA vaccine regimen.

In the US, however, public health officials are unequivocal in their approach: people are categorized as either vaccinated or unvaccinated, regardless of prior infection. It’s an approach with many strengths, including robust scientific data supporting vaccination for people who have recovered. That data—which we’ll get into below—has consistently shown that immune responses from natural infections are extremely variable, thus unreliable. Vaccines, on the other hand, have repeatedly proven to generate highly protective immune responses.

The vaccines are also remarkably safe, with few types of serious side effects that only occur extremely rarely. One of the most concerning side effects is myocarditis (inflammation of the heart muscle). But even there, the rate of myocarditis in the most at-risk group (males ages 12 to 29) is only estimated to be 41 in a million, and the cases are generally mild.

Comparing that with actual COVID-19 infections—which can cause severe disease even in young, healthy people and may cause persistent, months-long symptoms in up to half of people infected—there’s no contest. Vaccines are safer. And they’re just as safe for people who have previously recovered. People with past COVID-19 cases are no more likely to have serious side effects from vaccines than people who haven’t been previously infected, though they may have more side effects.

The US approach also has logistical benefits. Simple categories of “vaccinated” and “unvaccinated” skip over the messy and difficult step of figuring out who has been infected and when. From the early stages of the pandemic, the US has struggled—and is still struggling—to roll out accurate, widely available tests for SARS-CoV-2. Many people who have been infected never officially tested positive. Others assumed they were infected when they may have actually had one of many other respiratory infections. And antibody tests that look for evidence of past infections are notoriously inaccurate.

Though opponents argue that mass vaccination is driven by “evil corporations” out for prodigious profits at all costs, the fact is that vaccines are extremely safe and offer recovered people strong, lasting protection against a virus that has already killed more than 700,000 Americans.

Efficacy and variability

That’s not to say that there aren’t weaknesses to the US’s approach. For one thing, the approach can make vaccines look bad. In many instances, vaccine effectiveness is gauged by comparing COVID-19 case rates between vaccinated and unvaccinated people. But in the US, the unvaccinated include people who have no immunity and recovered people, who have some immunity and are, thus, expected to have fewer infections. This waters down case rates in the unvaccinated group and ends up lowering the vaccine efficacy estimates.

Still, the vaccines efficacy estimates are extraordinarily good. A recent study found that the Pfizer-BioNTech mRNA vaccine was holding steady with 90 percent efficacy against COVID-19 hospitalization for at least six months. A separate study found that the Moderna mRNA vaccine was 93 percent effective against hospitalizations among people without immunocompromising conditions. Johnson & Johnson’s vaccine was 71 percent effective.

And again, many vaccine efficacy numbers don’t account for past infection and may be artificially lower because of that. How much lower? It’s unclear. Since the beginning of the pandemic, researchers have noted time and again that immune responses generated by SARS-CoV-2 infections vary wildly, with some of the weaker responses seen in people with mild disease and stronger responses in people with severe disease.

In one study Ars reported on back in June of last year, researchers looking at SARS-CoV-2 antibodies in people who had recovered found that the difference between the highest and lowest levels varied by a factor of over 1,000. The researchers saw even more variability when they looked at neutralizing antibodies—those known to bind to the virus and prevent it from infecting cells. Neutralizing antibody levels in recovered people varied over a range of 40,000-fold, and up to 20 percent of people didn’t have any detectable level of neutralizing antibody.

Antibodies

Of course, antibodies are not the entirety of the immune responses that determine if a person will get infected or not and, if they do, how severe their infection will become. However, antibodies can provide a reasonable gauge of how well someone is protected. A study late last year that tracked 12,500 healthcare workers found that the higher the antibody levels, the lower the risk of infection. And in May of this year, researchers found “a remarkably strong” relationship between neutralizing antibody levels and vaccine protection.

A fundamental difference between the immune responses generated by vaccines and natural infection is their specificity. In a natural infection, whole SARS-CoV-2 viruses infect cells in the respiratory tract. Responding immune cells can target any number of facets of those whole viruses. This creates a relatively large diversity of antibodies that bind to different bits of SARS-CoV-2. The vaccines, meanwhile, only serve up to the immune system key snippets of SARS-CoV-2—namely the virus’ spike protein. This is the protein that SARS-CoV-2 uses to enter human cells, and it’s a key target of neutralizing antibodies. All antibodies in vaccinees will target the spike protein. Though vaccinees have less antibody diversity than previously infected people, they have high levels of highly targeted antibodies. Think of it as the difference between hunting a tiny virus with a shotgun and a sniper rifle.

With variable immune responses after infection comes variable real-world data on how well past infection protects against reinfection, which has led to the different public policy approaches. In a study conducted at Cleveland clinic and posted online in June, researchers found that among 52,238 employees, there were no differences in COVID-19 case rates between employees who were unvaccinated but previously infected, vaccinated and previously infected, and vaccinated people with no previous infection. “Individuals who have had SARS-CoV-2 infection are unlikely to benefit from COVID-19 vaccination,” the authors concluded.

Yet, in another study published in August by the Centers for Disease Control and Prevention, researchers looked at the vaccination status of more than 200 Kentucky residents who had tested positive for SARS-CoV-2 in 2020 and then tested positive again during May and June 2021. The CDC researchers found that people previously infected but unvaccinated were 2.34 times more likely to get reinfected than people who were previously infected and fully vaccinated.

Delta difference

The time frame for the CDC study coincides with the rise of the delta coronavirus variant in the US, which may also play a role in protection levels from past infection. In a French study published in July in Nature, researchers examined antibodies in 56 unvaccinated people who had recovered from a SARS-CoV-2 infection prior to the rise of delta. Six months after their infection and amid the rise of delta, the researchers found that their neutralizing antibody levels were 4 to 6 times lower against delta than they were against earlier variants.

The researchers next looked at a different group of 42 people who had gone a year since a SARS-CoV-2 infection. Of those 42, 26 were still unvaccinated and 21 had received one dose of a vaccine. At that point, the unvaccinated 26 had extremely low levels of neutralizing antibodies against any SARS-CoV-2 variants, particularly delta. Many people had no detectable levels of neutralizing antibody against delta. The vaccinated group, meanwhile, had high levels of neutralizing antibody similar to or above the levels seen in people who were fully vaccinated.

That finding has played out in several studies. A March study from researchers in Washington state, for instance, found that one dose of an mRNA vaccine in people who had recovered boosted levels of neutralizing antibody against all SARS-CoV-2 variants up to a thousandfold. And several other studies have found that vaccine doses after infection cause sky-high spikes in antibody levels. Some data has also suggested that antibody levels in the vaccinated recovered are even higher than people who have only been vaccinated.

Overall, the variable immune responses to infection, lower neutralization against delta, and the clear boost in protection from a very safe, highly effective vaccine make a strong argument for vaccinating the recovered.