In Reply We agree with Dr Re’em and colleagues that serology findings may not be a robust marker of SARS-CoV-2 infection at the individual patient level. Samples with indeterminate results (ie, optical density ratio ≥0.8 and <1.1) were discarded, thereby reducing the risk of false-negative results but not false-positive results. Since the sensitivity and specificity of the test in our study1 were assessed in another setting, we acknowledge that we cannot properly estimate the prevalence of past infection in our study population. However, whatever the prevalence one may assume in our population, participants with positive (vs negative) results would be more likely to have had COVID-19. Serology testing is thus a useful tool to study the correlates of SARS-CoV-2 infection at the population level. This conclusion holds even assuming a sensitivity of 61.2%, encompassing non-seroconversion, antibodies waning, and non-detection as reported by Perez-Saez et al.2 Moreover, introducing the date of serology testing (ie, May or June 2020 [n = 5698] vs after June 2020 [n = 21 125]) in model 2, we found no interaction between serology results and the date of serology testing.