Bayesian inference is a tool for treating uncertainty, combining data and prior information from multiple sources and formats into updateable models, often of considerable complexity. Attention is increasingly being paid to the use of Bayesian inference in the study of host-pathogen systems, in which complex networks of between-individual interactions operate across a hierarchy of ecological strata. Despite growing interest, the adoption of hierarchical systems-models by ecologists remains rudimentary. Bayesian inference has been applied to wildlife disease networks at a population level, and to epidemiological diagnostic regimes at an individual level, but there exist very few attempts to integrate models and data that link individual-, group-, population-, landscape- and assemblage-levels of wildlife disease systems. Furthermore, the use of Bayesian techniques at an individual level has been limited, yet this is vital for uncovering the fine-scale interactions and latent variables typical of disease networks. This review explores the use of Bayesian hierarchical models in the study of host-pathogen systems, identifying the future research required to achieve the desired “whole-system” approach. We argue that the complexities and uncertainties underlying disease processes are best described within a Bayesian framework, contending that although the infrastructure to craft complex Bayesian hierarchical models exists, the actual application of these methods is limited within wildlife disease research.