A Causal Inference Framework for Leveraging External Controls in Hybrid Trials

TL;DR

This paper develops a formal causal inference framework for integrating external control data into hybrid trials to improve treatment effect estimation, addressing exchangeability assumptions and proposing efficient estimators.

Contribution

It introduces a novel causal framework, graphical criteria, and doubly-robust estimators for combining internal and external controls in clinical trials.

Findings

Framework successfully applied to spinal muscular atrophy trial data.

Proposed estimators demonstrate good finite-sample performance.

The approach improves efficiency in estimating treatment effects.

Abstract

We consider the challenges associated with causal inference in settings where data from a randomized trial is augmented with control data from an external source to improve efficiency in estimating the average treatment effect (ATE). Through the development of a formal causal inference framework, we outline sufficient causal assumptions about the exchangeability between the internal and external controls to identify the ATE and establish the connection to a novel graphical criteria. We propose estimators, review efficiency bounds, develop an approach for efficient doubly-robust estimation even when unknown nuisance models are estimated with flexible machine learning methods, and demonstrate finite-sample performance through a simulation study. To illustrate the ideas and methods, we apply the framework to a trial investigating the effect of risdisplam on motor function in patients with spinal muscular atrophy for which there exists an external set of control patients from a previous trial.