Table 4.

Five examples of right-censoring bias

Structural Representation of Biases from Right-Censoring
Bias		Description
1	Confounding by common cause of treatment and attrition
2	Treatment affects censoring
3	No treatment effect when outcome causing censoring: Bias is not expected under the strict null
4	Treatment effect when outcome causes censoring and there is a true treatment effect. Expect a biased effect estimate for the target population on at least one causal contrast scale
5	Treatment effect and effect-modifiers differ in censored (restriction bias without confounding)

Key:

A denotes the treatment;

Y denotes the outcome;

U denotes an unmeasured confounder;

denotes randomisation into treatment;

asserts causality

biased path for treatment effect in the target population.

indicates a latent variable X measured by proxy X′.

indicates a path for bias linking A to Y absent causation.

indicates that conditioning on X introduces bias.

indicates effect modification of by X.

indicates effect modification of by