In linear least squares regression there exists a simple decomposition of the effect of an exposure on an outcome into two parts in the presence of an intermediate variable. This decomposition is described and then analogous decompositions for other models are examined, namely for logistic regression and proportional hazards models.
No takes yet. Share an insight, caveat, or question.
Christiana Kartsonaki (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: