Abstract Vaccines can provide protection against infection or limit disease severity. Vaccine efficacy (VE) is typically evaluated independently for different outcomes, but this does not provide insight into the mechanism of the protective effect and can cause biased estimates of VE. We propose a new conceptual framework and statistical implementation for VE estimation for infections with multiple possible outcomes of infection: joint analysis of multiple outcomes in vaccine efficacy trials (JAMOVET). JAMOVET is a Bayesian hierarchical regression model that controls for biases and can evaluate covariates for VE, the hazard of infection, and the probability of progression. We applied JAMOVET to simulated data, and data from COV002 (NCT04400838), a phase 2/3 trial of ChAdOx1 nCoV-19 (AZD1222) vaccine. Simulations showed that biases are corrected by explicitly modeling disease progression and imperfect test characteristics. JAMOVET estimated ChAdOx1 nCoV-19 VE against infection (${\mathrm{VE}}_{in}$) at 55% (95% credible interval [CrI] 35-70) and progression to symptoms (${\mathrm{VE}}_{pr}$) at 44% (95% CrI 26-59). This implies a VE against symptomatic infection of 75% (95% CrI 62-85), consistent with published trial estimates. JAMOVET is a powerful tool for evaluating diseases with multiple dependent outcomes and can be used to adjust for biases and identify predictors of key outcomes.