Fig. 1From: A DAG-based comparison of interventional effect underestimation between composite endpoint and multi-state analysis in cardiovascular trialsDirected acyclic graph for the causal system between treatment (X), being free of any event at time t (S t ) and t+Δ (S t+Δ ), and unobserved variables (Z) that are unrelated to treatment (for example by randomization) and affect the event rate. Figure according to Aalen et al. [20]Back to article page