Abstract

The authors' reply: Figure 2 in our article shows descriptive statistics of the mass Covid-19 vaccination campaign in Chile. The figure does not show post-vaccination results; rather, it provides context to readers about the vaccination rollout. Figure 2A shows the cumulative number of participants who received one or two doses of vaccine and also shows outstanding compliance with the recommended vaccination schedule. Figure 2B shows the crude cumulative incidence of Covid-19 in three cohorts: participants who received two vaccine doses, those who received one dose, and those who remained unvaccinated on May 1, 2021 (the end of the follow-up period). The curves include all infections, regardless of the participant’s vaccination status or when vaccination occurred. They show the risk to which each cohort was exposed without adjustments. Because of the rapid vaccine rollout, the group that received two doses had some protection as compared with the other two groups (those who had received one dose and those who were unvaccinated). The one-dose group initiated vaccination approximately 40 days after the beginning of the campaign; 45% of the cases in this group occurred before vaccination. Figure 1 here shows the estimated cumulative incidence curves for a specific group based on the model. The surge in Covid-19 cases in Chile during the vaccination campaign was probably related to a rollback of Covid-19 restrictions, which resulted in increased mobility and behavioral changes in the population. Vaccine effectiveness results, both in our study and in a study in Turkey,1 and the striking differences in outcomes between fully immunized persons and unvaccinated persons2 suggest that the surge resulted from a highly transmissible virus circulating in a partially immunized population. Limited Covid-19 restrictions and a low proportion of the population being fully vaccinated may explain the high incidence of cases in countries where inactivated vaccines are being used. Finally, Table 2 in our article shows crude, unadjusted incidence rates for the four Covid-19 outcomes that were considered in our study: symptomatic cases of Covid-19, hospitalizations, ICU admissions, and confirmed deaths. The vaccination campaign in Chile prioritized persons at higher risk for severe illness, including older adults and persons with underlying conditions. Therefore, simple examination of the crude incidence rates may be misleading. Simpson’s paradox,3 which shows that the marginal and conditional association measures often differ, can explain the observed differences in incidence rates between groups according to immunization status. Our vaccine effectiveness estimates adjust for age, sex, region of residence, income, nationality, and underlying conditions associated with severe Covid-19. We provide several robustness checks in our article. In summary, there is robust evidence, both from our study and the Turkish study,1 that this inactivated SARS-CoV-2 vaccine is highly effective against severe Covid-19 and related hospitalization and death.