Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 77 - Causal Inference When Resources Are Limited
Type: Topic-Contributed
Date/Time: Monday, August 9, 2021 : 10:00 AM to 11:50 AM
Sponsor: Biometrics Section
Abstract #317556
Title: Causal Inference for Organ Transplantation: Challenges in Studying a Finite Resource and an Application in Lung Transplantation
Author(s): Ruth Keogh*
Companies: London School of Hygiene and Tropical Medicine
Keywords: Causal inference; Survival; Longitudinal data; Organ transplant

Organ transplantation is a treatment option for severe disease. Understanding its impact on survival relies on observational data such as national transplant registers, as randomized trials are infeasible. In this work we specify target trials to articulate different causal questions and discuss how to answer them using longitudinal data on patient measures, transplant status, and survival. The motivation is a study of the impact of lung transplantation on survival in cystic fibrosis. Causal questions include what the effect is of joining the transplant wait-list and what the effect of transplant is in those who received one. We consider both population-average and personalised effects. Different approaches to addressing time-dependent confounding and estimating the causal quantities of interest and will be discussed. These include methods based on forming a sequence of emulated trials within the longitudinal data and using inverse probability weighting in the estimation, and g-estimation. Organs are a finite resource and organ allocation depends on a match between certain donor and recipient characteristics. I will discuss the challenges this brings and how they can be addressed.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program