Addressing Misclassification Bias in Vaccine Effectiveness Studies with an Application to Covid-19 (preprint)

Safe and effective vaccines are crucial to control Covid-19 and to protect persons who are at high risk of complications or death. Test-negative design is a popular option for evaluating the effectiveness of Covid-19 vaccines, but the findings could be biased by several factors, including imperfect sensitivity and/or specificity of the test used for the SARS-Cov-2 infection.We propose a simple Bayesian modeling approach for estimating vaccine effectiveness that is robust even when the diagnostic test is imperfect.We use simulation studies to demonstrate this robustness to misclassification bias for estimating Covid-19 vaccine effectiveness, and illustrate the utility of our approach using real-world examples