A comparison of methods for estimating the average treatment effect on the treated for externally controlled trials

TL;DR

This paper compares various statistical methods for estimating the average treatment effect on the treated in single-arm trials with external controls, highlighting their biases, variances, and robustness under different assumptions.

Contribution

It provides a comprehensive simulation-based comparison of multiple methods for ATT estimation, emphasizing the advantages of doubly robust approaches.

Findings

Doubly robust methods generally exhibit smaller biases.

Nonmatching methods tend to have smaller standard deviations.

Violations of assumptions significantly impact method performance.

Abstract

While randomized trials may be the gold standard for evaluating the effectiveness of the treatment intervention, in some special circumstances, single-arm clinical trials utilizing external control may be considered. The causal treatment effect of interest for single-arm trials is usually the average treatment effect on the treated (ATT) rather than the average treatment effect (ATE). Although methods have been developed to estimate the ATT, the selection and use of these methods require a thorough comparison and in-depth understanding of the advantages and disadvantages of these methods. In this study, we conducted simulations under different identifiability assumptions to compare the performance metrics (e.g., bias, standard deviation (SD), mean squared error (MSE), type I error rate) for a variety of methods, including the regression model, propensity score matching (PSM), Mahalanobis distance matching (MDM), coarsened exact matching, inverse probability weighting, augmented inverse probability weighting (AIPW), AIPW with SuperLearner, and targeted maximum likelihood estimator (TMLE) with SuperLearner. Our simulation results demonstrate that the doubly robust methods in general have smaller biases than other methods. In terms of SD, nonmatching methods in general have smaller SDs than matching-based methods. The performance of MSE is a trade-off between the bias and SD, and no method consistently performs better in term of MSE. The identifiability assumptions are critical to the models' performance: Violation of the positivity assumption can lead to a significant inflation of type I errors in some methods; violation of the unconfoundedness assumption can lead to a large bias for all methods... (Further details are available in the main body of the paper).