AppraisalResearch Note: Estimating the complier average causal effect when participants in randomised trials depart from allocated treatment
Under a Creative Commons license
open access
Cited by (0)
Footnotes: aThese assumptions are: no unmeasured confounding of the relationship between the exposure and the outcome; all participants have a non-zero probability of receiving all levels of the exposure; the outcome that was observed for a participant is the same as the potential outcome given their actual level of exposure.
Ethics approval: Nil.
Competing interests: Nil.
Source(s) of support: Nil.
© 2021 Australian Physiotherapy Association. Published by Elsevier B.V.