Evidence is building that immunity from Covid-19 infection is at least as strong as that from vaccination. Scientists are divided on the implications for vaccine policy.

The role of immunity from infection, which scientists have been trying to figure out since the outset of the pandemic, has gained fresh significance amid the controversy over vaccine mandates.

Vaccines...

Evidence is building that immunity from Covid-19 infection is at least as strong as that from vaccination. Scientists are divided on the implications for vaccine policy.

The role of immunity from infection, which scientists have been trying to figure out since the outset of the pandemic, has gained fresh significance amid the controversy over vaccine mandates.

Vaccines typically give rise to a stronger antibody response than infection, which might make them better at fending off the virus in the short term. Infection triggers a response that evolves over time, possibly making it more robust in the long term. A combination of both types appears to be stronger than either alone. But the jury is out on whether one form is stronger than the other, and whether their relative strength even matters for vaccine policy.

The comparison is further complicated by the emergence of new variants, such as that identified this month in southern Africa, which may be more contagious and be better at evading vaccines.

One thing is clear: Vaccination is a far safer, more reliable strategy for acquiring immunity, given the risks of serious illness or death from infection. But viewpoints splinter about whether people who have had Covid-19 before need a full course of vaccination, and whether documented prior infection should count as proof of immunity—as is the case in some other countries, including much of Europe.

Immunity from infection hasn’t been studied as extensively as vaccine-mediated immunity. But over the course of the pandemic, clues have emerged to suggest the two are at least equivalent.

The role of immunity from infection has gained fresh significance amid the controversy over vaccine mandates.

Photo: Spencer Platt/Getty Images

Several peer-reviewed studies conducted in the early part of the pandemic, before widespread vaccination, found that people infected during the first waves were around 80% less likely to test positive during the next surge. Those studies spanned healthcare workers in the U.K., the Danish population and patients at the Cleveland Clinic, a large health system with facilities mostly in Ohio and Florida.

A recent Israeli study found that people who had been vaccinated with two shots of the vaccine developed by Pfizer Inc. and BioNTech SE —the most commonly used there—were 13 times more likely to later get infected than those with a prior infection. The study, which hasn’t been peer reviewed, tracked confirmed infections between June and August this year for people who had been either vaccinated or infected in January or February.

It also suggested that immunity from infection is longer lasting than that from vaccination.

More real-world evidence would be needed to make the case that immunity from infection is superior to that from vaccination, said David Dowdy, associate professor of epidemiology at the Johns Hopkins Bloomberg School of Public Health.

A factor that may have exaggerated the protective benefit of infection in the Israeli study was that vaccinated people could be more likely to travel abroad and bring the virus back to their vaccinated families, pushing case numbers up in that group, he said.

Data from the U.K.’s Office for National Statistics showed that, between May and August, a prior infection offered around the same level of protection against the Delta variant as vaccination with either the Pfizer shot or the one developed by AstraZeneca PLC and the University of Oxford.

A woman receives her second Pfizer shot in Givatayim, Israel.

Photo: Oded Balilty/Associated Press

Some studies suggest the opposite. One, conducted by the Centers for Disease Control and Prevention, found that, among people hospitalized with a respiratory illness, Covid-19 was over five times more common among those who were unvaccinated and had an earlier infection compared with those who were fully inoculated and hadn’t had the virus before. Critics say the study, which hasn’t been peer reviewed, had flaws that likely overestimated the relative strength of vaccination.

The CDC said in a recent review of the current scientific evidence that both fully vaccinated and those previously infected with the virus have a low risk of subsequent infection for at least six months.

“It is complicated but…we’re at a state in the world where [vaccination and prior infection] seem equally protective,” said Monica Gandhi, professor of medicine and associate chief of the University of California San Francisco’s division of HIV, infectious diseases and global medicine.

The two forms of immunity appear to have different strengths. Vaccination with mRNA vaccines produced higher concentrations of neutralizing antibodies—the type that prevent the virus from entering cells—than infection, although levels waned in both groups, according to a recent paper published in the journal Nature by researchers at the Rockefeller University in New York.

The new coronavirus variant has sparked fears of further travel restrictions.

Photo: cj gunther/Shutterstock

Immune memory, however, appears to be stronger following infection. The Rockefeller research group found in an earlier study, also published in Nature, that the antibodies produced by memory B cells—which quickly multiply in subsequent encounters with the virus—continued to evolve at least a year after infection. The study on vaccinated people found that the antibodies produced by their memory B cells didn’t change much over time.

One possible reason for the difference, they said, was that pieces of virus remain in the body for weeks after infection, whereas vaccine particles fade away faster. The upshot: The immune memory of people who have been infected is ready to produce a broader array of antibodies than of people who have been vaccinated.

Michel Nussenzweig, the professor who led the Rockefeller research, said the papers suggest that vaccination likely offers better protection from infection but that this protection wanes rapidly. However, the quality of long-term immune memory, which is key to responding to infection and staying out of the hospital, is superior in people who have had an infection, he said.

So-called hybrid immunity—that in people who have had both vaccination and infection—has been shown to be strongest of all. The Rockefeller researchers found that vaccination boosted levels of antibodies in the blood and memory B cells in people who had been infected before. The effect also appears to work in the other direction: A study of vaccinated people who were infected during a July 4 holiday weekend outbreak in Cape Cod found that they produced high levels of antibodies and T-cells directed against the virus. That study, led by researchers at the Beth Israel Deaconess Medical Center in Boston, hasn’t been peer reviewed.

A drug store providing Covid-19 vaccine shots in New York City.

Photo: Spencer Platt/Getty Images

Questions remain, though, about whether people who have had Covid-19 need a full course of vaccination. A study from New York University found that although one dose of the Pfizer vaccine significantly increased antibody levels in people with a prior infection, a second dose produced a more muted response.

Another study from researchers at the Icahn School of Medicine at Mount Sinai in New York found that a single dose of the Pfizer or Moderna Inc. vaccines produced more antibodies in people who had previously had Covid-19 than two doses did in those who had never encountered the virus. It also found that people with prior infection report more unpleasant side effects from vaccination. The authors concluded that offering a single shot to those who had already had Covid-19 wouldn’t negatively affect their antibody levels and would spare them from needless pain. The NYU and Icahn studies haven’t been peer reviewed.

Some doctors say the mounting evidence on the role of immunity from infection supports a more nuanced approach to vaccine policy.

Among them is UCSF’s Dr. Gandhi, who supports a single dose of vaccine in people who have had the virus. She also thinks prior infection should carry weight when it comes to vaccine mandates. “Mandating [vaccination] so that someone [unvaccinated] loses their job if they have a proven prior infection is going too far,” she said.

Marty Makary, a professor at the Johns Hopkins University School of Medicine, also advocates a case-by-case approach to vaccination in people who have already had Covid-19, especially among children. “There’s no scientific basis for vaccinating people who had the infection,” he said. “It’s not clear to me that the benefits of vaccination in someone who has circulating antibodies outweighs the risk.”

Lockdowns, vaccine requirements and travel restrictions have swept Europe amid rising Covid infections and concerns over a variant detected in South Africa, highlighting new challenges ahead for the U.S. as officials want to avoid more shutdowns. The Wall Street Journal Interactive Edition

Yet others say universal vaccination—as recommended by the CDC—still makes sense. That is mainly because the vaccines are safe and have been shown to enhance the immune response of people who have been infected before.

One issue with a more targeted approach is that immune responses to infections vary, and there is no way to sort out people whose infection led to a strong response from those who didn’t. Although responses to vaccination also differ, the dose is fixed, making it less variable, they say.

“The risk of vaccination is extraordinarily low,” said Tom Frieden, former director of the CDC and chief executive of Resolve to Save Lives, a nonprofit initiative that works on strengthening epidemic preparedness. “The benefit is high and the uncertainty with infection makes it so that you can’t make that a replacement to vaccination.”

Write to Denise Roland at Denise.Roland@wsj.com