New government-funded research this week is expected to offer some comfort to people who survived covid-19. This suggests that they have a low risk of reinfection of the coronavirus, at least about three months later.
Researchers at the National Cancer Institute teamed up with commercial testing labs and two health data collection companies for this study, Published Wednesday at JAMA Internal Medicine.
They analyzed unidentified data from more than 3 million Americans who did a commercial antibody test for SARS-CoV-2, the coronavirus that causes covid-19, sometime between January and August 2020. The antibody tests, although they are not perfect, they indicate whether someone has had a recent recent infection. These people were divided into those who had antibodies and those who did not, based on tests. Next, the researchers analyzed how many people in both groups subsequently underwent a PCR test for covid-19, which aims to diagnose an active infection.
About 10% of people in each group started to have a CRP test. More people with antibodies tested positive for the virus in the first 30 days after the antibody test than those without antibodies. But this is not surprising, since detectable traces of the virus can remain in the body for months, even after the symptoms have passed and the person is no longer infected. Therefore, it is likely that these positive PCR results would generally detect the first infection. When the researchers specifically looked at the rate of positive testing after the first month and especially more than 90 days later – enough time for a positive PCR test to indicate true reinfection – the results were encouragingly different.
After three months or more, only 0.3% of people with a previous positive antibody test tested positive for coronavirus again, compared with 3% of people with a negative antibody test. In other words, having a previous infection was associated with a much lower risk of infection three or more months later.
G / O Media can receive a commission
“People who have recovered from covid-19 should be sure that being positive for antibodies is associated with some protection against a new infection,” said study author Douglas Lowy, chief deputy director of the NCI, in an email.
However, the findings have their limitations. On the one hand, they cannot tell us exactly how much protection a past infection will provide against reinfection, or how long it should last (although other research has suggested which can take years). Another factor that this study cannot explain is the recent emergence of variants of the coronavirus. Some – like that first identified in South Africa last year – they are believed to increase the risk of reinfection, as they can partially prevent the immune response created by a previous infection or vaccination.
Still, there is no research showing that any variants currently spreading can completely escape from natural immunity or provided by someone’s vaccine. Our immune system has many weapons against a family germ, and most reinfections are likely to end up being milder than the first time.
Even before these new variants existed, however, there were documented cases of reinfection, including cases where the symptoms were worse on the second attempt. And the new study’s findings still suggest that reinfection does happen, though rarely. Therefore, no one should assume that they are immune to covid-19 just because they survived a previous infection without any problems. Ultimately, the best way to keep everyone safe from covid-19 is to vaccinate as many people as possible, including those who have already had the viral disease, according to Lowy. It is a remedy that involves much less risk than contracting a natural infection.
“People who have recovered from covid-19 must still plan vaccination when they have the opportunity,” he said.
The NCI plans to continue to fund research that will track the prevalence of reinfection in the general public, along with studies that will examine how our immune response to the virus can change over time and against new variants.