Causal inference plays an essential role in public health research, enabling the development of informed policy decisions, clinical guidelines, and population-level interventions. However, observational data pose substantial challenges, including complex data structures, missing data, heterogeneous populations, and post-treatment variables. This session focuses on Bayesian causal inference in public health, highlighting how Bayesian frameworks offer a rigorous and coherent approach to addressing these challenges. Bayesian methods naturally accommodate uncertainty, integrate prior scientific knowledge, and provide robust strategies for modeling missing data and latent structures, as well as for estimating a range of causal treatment effects. Consequently, Bayesian models make them particularly well-suited for diverse and complex public health applications, where data limitations are common and advanced statistical methods are essential. The session brings together statisticians, biostatisticians, and applied researchers to advance rigorous Bayesian causal methods in public health, promoting dialogue between methodological innovation and real-world applications to support reliable evidence for decision makers. Specifically, it will be an opportunity to explore innovative Bayesian models, including factor models and Bayesian nonparametric priors, and to deepen understanding of causal estimand identification, while comparing different causal inference settings and public health questions.