Skip to content
flowss
Start freeChecking your session
Causal DAG templates

Causal DAG — randomised trial

Why randomisation breaks the confounder path.

Causal DAGResearch & AcademicCausal DAG syntax docs
Template previewCausal DAG
Randomisation severs confoundingETreatmentOOutcomeCPatient characteristicsRRandom assignmentExposureOutcomeConfounderMediatorUnobservedThe exposure is unconfounded: every backdoor path is already blocked, so the crude association estimates the total effect. 4 variables, 3 edge…No adjustment is needed: the exposure is unconfounded in this diagram, so the crude association between `Treatment` and `Outcome` already estimates the total effe…

Make it your own.

title "Randomisation severs confounding"
exposure   E "Treatment"
outcome    O "Outcome"
confounder C "Patient characteristics"
node       R "Random assignment"

R -> E
E -> O
C -> O