1
$\begingroup$

I'm working on a project dealing with transplants and comparing outcomes of first time (primary) transplants with re-transplants. I'm trying to decide which set of variables to include for propensity score matching.

For example, for the category of age, would it be better to match based on age at the time of the first transplant for both cohorts or for the age at the time of the first transplant for the primary transplant group and age at the time of retransplant for the retransplant group?

Improve this question
$\endgroup$
2
  • 2
    $\begingroup$ I don't know which are your control and treatment group, but typically the answer would be whichever variables your theory and the literature suggest would satisfy the ignorability and positivity assumptions, and check balance before the matching. Additionally, why do you have the survival tag? $\endgroup$
    – dubafek
    Commented Apr 10, 2022 at 23:03
  • 1
    $\begingroup$ Are you sure that you want to use propensity score matching? That runs a risk of throwing away useful data and leads to problems that other ways to control for confounding can help to avoid. With other adjustment techniques, you could in principle control for all of those factors and not have to choose which to match. Please read this page for a superb, expert explanation. $\endgroup$
    – EdM
    Commented Apr 12, 2022 at 20:45

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.