Vaccine effectiveness in reducing COVID-19-related hospitalization after a risk-age-based mass vaccination program in a Chilean municipality: A comparison of observational study designs

Background: Case–control studies involving test-negative (TN) and syndrome-negative (SN) controls are reliable for evaluating influenza and rotavirus vaccine effectiveness (VE) during a random vaccination process. However, there is no empirical evidence regarding the impact in real-world mass vaccination campaigns against SARS-CoV-2 using TN and SN controls. Objective: To compare in the same population the effectiveness of SARS-CoV-2 vaccination on COVID-19-related hospitalization rates across a cohort design, TN and SN designs. Method: We conducted an unmatched population-based cohort, TN and SN case–control designs linking data from four data sources (public primary healthcare system, hospitalization registers, epidemiological surveillance systems and the national immunization program) in a Chilean municipality (Rancagua) between March 1, 2021 and August 31, 2021. The outcome was COVID-19-related hospitalization. To ensure sufficient sample size in the unexposed group, completion of follow-up in the cohort design, and sufficient time between vaccination and hospitalization in the case–control design, VE was estimated comparing 8-week periods for each individual. Results: Among the 191,505 individuals registered in the primary healthcare system of Rancagua in Chile on March 1, 2021; 116,453 met the cohort study's inclusion criteria. Of the 9,471 hospitalizations registered during the study period in the same place, 526 were COVID-19 cases, 108 were TN controls, and 1,628 were SN controls. For any vaccine product, the age- and sex-adjusted vaccine effectiveness comparing fully and nonvaccinated individuals was 67.2 (55.7–76.3) in the cohort design, whereas it was 67.8 (44.1–81.4) and 77.9 (70.2–83.8) in the TN and SN control designs, respectively. Conclusion: The VE of a COVID-19 vaccination program based on age and risk groups tended to differ across the three observational study designs. The SN case-control design may be an efficient option for evaluating COVID-19 VE in real-world settings.