Fig. 2From: How unmeasured confounding in a competing risks setting can affect treatment effect estimates in observational studiesDirected Acyclic Graph showing the relationship between some of the parametersBack to article page