This chapter discusses diagnostic reasoning from the perspective of causal inference. The computational framework that provides the foundation for the analyses–probabilistic inference over graphical causal structures–can be used to implement different models that share the assumption that diagnostic inferences are guided and constrained by causal considerations. This approach has provided many critical insights, with respect to both normative and empirical issues. For instance, taking into account uncertainty about causal structures can entail diagnostic judgments that do not reflect the empirical conditional probability of cause given effect in the data, the classic, purely statistical norm. The chapter first discusses elemental diagnostic inference from a single effect to a single cause, then examines more complex diagnostic inferences involving multiple causes and effects, and concludes with information acquisition in diagnostic reasoning, discussing different ways of quantifying the diagnostic value of information and how people decide which information is diagnostically relevant.