Vaccination with BNT162b2 reduces transmission of SARS-CoV-2 to household contacts in Israel

Abstract

The individual-level effectiveness of vaccines against clinical disease caused by SARS-CoV-2 is well-established. However, few studies have directly examined the effect of COVID-19 vaccines on transmission. We quantified the effectiveness of vaccination with BNT162b2 (Pfizer-BioNTech mRNA-based vaccine) against household transmission of SARS-CoV-2 in Israel. We fit two time-to-event models - a mechanistic transmission model and a regression model - to estimate vaccine effectiveness against susceptibility to infection and infectiousness given infection in household settings. Vaccine effectiveness against susceptibility to infection was 80-88%. For breakthrough infections among vaccinated individuals, the vaccine effectiveness against infectiousness was 41-79%. The overall vaccine effectiveness against transmission was 88.5%. Vaccination provides substantial protection against susceptibility to infection and slightly lower protection against infectiousness given infection, thereby reducing transmission of SARS-CoV-2 to household contacts.

One-sentence summary: Vaccination reduced both the rate of infection with SARS-CoV-2 and transmission to household contacts in Israel.

Fig. 1.. Schematic representation of the data augmentation process for an example household.

Each infected household member is associated with: (A,D) a distribution for time from onset of infectiousness to testing; (B,E) a distribution for the infectious period; and (C,F) a distribution for the latent period. The filled ovals represent observed events, while the circles and stars represent unobserved events in the infection timeline. Panels (A–C) and (D–F) represent two possible sample sets from the delay distributions, each with a different index case.

Fig. 2.. Forest-plot of the age-adjusted vaccine effectiveness estimates across the 100 iterations of the delay distributions from the primary transmission model.

(A) Age-adjusted vaccine effectiveness against susceptibility to infection (VE_S); (B) age-adjusted vaccine effectiveness against infectiousness given infection (VE_I); (C) age-adjusted vaccine effectiveness against transmission (VE_T).

Fig. 3.. Forest-plot of the vaccine effectiveness estimates across the 100 iterations of the delay distributions from the alternative infection-hazard model.

Vaccine effectiveness estimates against susceptibility to infection are plotted (A) in the absence of infected household members (VE_S,0), or with at least one (B) unvaccinated (VE_S,u) or (C) fully vaccinated household member (VE_S,v). The vaccine effectiveness estimates of being exposed to a fully vaccinated versus an unvaccinated infectious household member are plotted given (D) individual j is unvaccinated (VE_I,u) or (E) individual j is fully vaccinated (VE_I,v).