Many Adults Who Thought They Had COVID-19 Actually Didn't

About half of unvaccinated adults who thought they had COVID-19 were proven wrong by antibody tests, researchers found.

In a survey of people recruited on social media, 45% of people who thought they'd had COVID, but never confirmed it, actually had no anti-spike antibodies, reported Dorry Segev, MD, PhD, of Johns Hopkins University in Baltimore, and colleagues, in a JAMA research letter.

Overall, antibodies were detected in 99% of people who said they had a positive COVID-19 test result, in 55% who believed they had COVID-19 but who were never tested, and in 11% who thought they never had SARS-CoV-2 infection.

The study took place before the Omicron wave, in which all forms of immunity have shown to be less durable. Other limitations noted by the authors included a lack of direct neutralization assays, as "antibodies alone do not directly equate to immunity," they wrote.

Segev and co-authors recruited healthy unvaccinated adults through a Twitter post and a Facebook ad from Sept. 11, 2021 to Oct. 8, 2021. Participants answered a questionnaire about demographics, COVID status, and mask use. They were divided into three groups:

• COVID-confirmed: people who reported a test-confirmed COVID infection
• COVID-unconfirmed: people who believed they had COVID, but were never tested
• No-COVID: people who did not believe they had COVID and never tested positive

All groups were invited to undergo antibody testing. Overall, 816 adults underwent screening and filled out the questionnaire. Respondents had a mean age of 48, 52% were women, and 82% were white.

The COVID-confirmed group had 295 participants; of these, antibody testing confirmed 99% actually had COVID. A median of 8.7 months (range 0-20) had passed since this group's self-reported COVID diagnosis. The median level of antibodies against the SARS-CoV-2 spike protein receptor-binding domain (RBD) among those who tested positive was 205 U/mL, and "levels did not differ by months since diagnosis," according to the researchers.

Among 275 people in the COVID-unconfirmed group, the median anti-RBD level among those who tested positive was 131 U/mL. Of 246 people in the no-COVID group, the median anti-RBD level among those who tested positive was 82 U/mL.

Segev's group acknowledged that COVID diagnoses were self-reported in this study, and there may be "an unknown degree of selection bias" due to the survey's public recruitment efforts. In addition, the study population was healthy and mostly white, which could limit its generalizability.

Comment