Causal theory and the etiology of periodontal diseases

Causal theory and causal models have proven to be helpful in understanding the multi-factorial causation of many diseases, including periodontitis. The design and application of causal models is essential to the appropriate design, conduct and analysis of studies on periodontal disease causation. This review discusses the principles of causation and causal theory, both in general and as they pertain to studies of the etiology of periodontitis. Specifically, we present an overview of models of causation, including the potential outcomes model, the sufficient component cause model, and directed acyclic graphical models. More thorough explanations and accompanying illustrations are provided for the sufficient component cause model and the directed acyclic graphical model. Topics related to the study of periodontitis, as illustrated by such models, include the principles of confounding and effect measure modification (interaction). Existing models of periodontitis etiology are critically reviewed in this context.