Waning of two-dose BNT162b2 and mRNA-1273 vaccine effectiveness against symptomatic SARS-CoV-2 infection is robust to depletion-of-susceptibles bias

ABSTRACT

Concerns about the duration of protection conferred by COVID-19 vaccines have arisen in postlicensure evaluations. However, “depletion of susceptibles” bias driven by differential accrual of infection among vaccinated and unvaccinated individuals may contribute to the appearance of waning vaccine effectiveness (VE) in epidemiologic studies, potentially hindering interpretation of estimates. We enrolled California residents who received molecular SARS-CoV-2 tests in a matched, test-negative design case-control study to estimate VE of mRNA-based COVID-19 vaccines between 23 February and 5 December 2021. We analyzed waning protection following 2 vaccine doses using conditional logistic regression models. Additionally, we used data from case-based surveillance along with estimated case-to-infection ratios from a population-based serological study to quantify the potential contribution of the “depletion-of-susceptibles” bias to time-varying VE estimates for 2 doses. We also estimated VE for 3 doses relative to 0 doses and 2 doses, by time since second dose receipt. Pooled VE of BNT162b2 and mRNA-1273 against symptomatic SARS-CoV-2 infection was 91.3% (95% confidence interval: 83.8-95.4%) at 14 days after second-dose receipt and declined to 50.8% (31.2-75.6%) at 7 months. Accounting for differential depletion-of-susceptibles among vaccinated and unvaccinated individuals, we estimated VE was 53.2% (23.6-71.2%) at 7 months among individuals who had completed the primary series (2 doses). With receipt of a third dose of BN162b2 or mRNA-1273, VE increased to 95.0% (82.8-98.6%), compared with zero doses. These findings confirm that observed waning of protection is not attributable to epidemiologic bias and support ongoing efforts to administer additional vaccine doses to mitigate burden of COVID-19.