Extending inferences from a randomized trial to a new target population

TL;DR

This paper reviews and compares methods for generalizing causal inferences from randomized trials to broader populations, addressing challenges when treatment effect modifiers influence participation.

Contribution

It introduces and evaluates methods for extending trial results to new populations using outcome and participation models, including doubly robust approaches.

Findings

Methods perform well in simulations

Software implementations are provided

Application to coronary artery disease trial demonstrates practical utility

Abstract

When treatment effect modifiers influence the decision to participate in a randomized trial, the average treatment effect in the population represented by the randomized individuals will differ from the effect in other populations. In this tutorial, we consider methods for extending causal inferences about time-fixed treatments from a trial to a new target population of non-participants, using data from a completed randomized trial and baseline covariate data from a sample from the target population. We examine methods based on modeling the expectation of the outcome, the probability of participation, or both (doubly robust). We compare the methods in a simulation study and show how they can be implemented in software. We apply the methods to a randomized trial nested within a cohort of trial-eligible patients to compare coronary artery surgery plus medical therapy versus medical therapy alone for patients with chronic coronary artery disease. We conclude by discussing issues that arise when using the methods in applied analyses.