One lazy Sunday afternoon, my cell phone vibrated with a new text message. One of my relatives in Italy wanted my take on an Italian news article about protection against SARS-CoV-2 infection by the COVID-19 vaccines versus protection resulting from a natural infection.

As a former scientist and now a journalist who frequently covers vaccine and infectious disease research, my first instinct was to emphasize the effectiveness and safety of the COVID-19 vaccines. Getting vaccinated against SARS-CoV-2 will always be safer than taking chances with a natural infection.

But as I read through the article (with some help from Google Translate), I realized that the question was more nuanced than simply getting vaccinated or not. The article was an opinion piece in favor of exempting people who had recovered from COVID-19 infection from vaccination. The writers based their argument on results from a recent CDC report that analyzed COVID-19 case and hospitalization rates of people in California and New York. The CDC scientists found that before the rise of the delta variant, COVID-19 cases were lowest in people who had been vaccinated against COVID-19 with no evidence of a prior COVID-19 infection. After the delta variant became the dominant one, however, case rates rose in this population. The rate of COVID-19 infections was lower in people who had recovered from a prior COVID-19 infection — regardless of vaccination status — than in those who were vaccinated but had no prior COVID-19 infection.

These results make sense with what we know about the SARS-CoV-2 variants so far. Researchers have shown that there is a slight drop in the effectiveness of the COVID-19 vaccines against the delta variant, but the vaccine is still effective at preventing severe disease. Because the COVID-19 vaccines are designed so that our immune system mounts a response against the viral spike protein, vaccines may be less effective against variants that have mutations in the spike protein. During a SARS-CoV-2 infection, the body develops antibodies against many different SARS-CoV-2 proteins, not just the spike protein. This means that the body will likely recognize SARS-CoV-2 if it sees it again, even if the spike protein has mutated in that particular variant.

In light of these results, the article argued that a prior COVID-19 infection offers protection that is just as good at preventing SARS-CoV-2 reinfection as vaccination is, meaning that recovery from COVID-19 infection could stand in for employer or government-mandated vaccination requirements. Without commenting on vaccine requirements in specific countries, I told my relative that there are some important caveats to consider before pushing for policy changes based on these data.

The CDC study did not assess any potential protective effects of booster doses of the COVID-19 vaccines or how the omicron variant, now the most dominant SARS-CoV-2 variant worldwide, fares against immunity from vaccination or natural infection. When a new variant emerges, scientists will need to address these same questions again in another new context.

Like many in the scientific community who have fielded questions from family and friends about SARS-CoV-2 and the pandemic for multiple years now, I explained what the study showed and what it didn’t. We’re still learning more about this virus every day, from the biological changes occurring in each new variant to how long our immune protection against those variants will likely last. I double-checked my rusty Italian and sent back a text that, I hope, made the science surrounding this ever-evolving pandemic a little clearer.