Robust estimation of heterogeneous treatment effects in randomized trials leveraging external data

Rickard K.A. Karlsson, Piersilvio De Bartolomeis, Issa Dahabreh, Jesse H. Krijthe
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:1783-1791, 2026.

Abstract

Randomized trials are typically designed to detect average treatment effects but often lack the statistical power to uncover individual-level treatment effect heterogeneity, limiting their value for personalized decision-making. To address this, we propose the QR-learner, a model-agnostic learner that estimates conditional average treatment effects (CATE) within the trial population by leveraging external data from other trials or observational studies. The proposed method is robust: it can reduce the mean squared error relative to a trial-only CATE learner, and is guaranteed to recover the true CATE even when the external data are not aligned with the trial. Moreover, we introduce a procedure that combines the QR-learner with a trial-only CATE learner and show that it asymptotically matches or exceeds both component learners in terms of mean squared error. We examine the performance of our approach in simulation studies and apply the methods to a real-world dataset, demonstrating improvements in both CATE estimation and statistical power for detecting heterogeneous effects.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-karlsson26a, title = { Robust estimation of heterogeneous treatment effects in randomized trials leveraging external data }, author = {Karlsson, Rickard K.A. and De Bartolomeis, Piersilvio and Dahabreh, Issa and Krijthe, Jesse H.}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {1783--1791}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/karlsson26a/karlsson26a.pdf}, url = {https://proceedings.mlr.press/v300/karlsson26a.html}, abstract = { Randomized trials are typically designed to detect average treatment effects but often lack the statistical power to uncover individual-level treatment effect heterogeneity, limiting their value for personalized decision-making. To address this, we propose the QR-learner, a model-agnostic learner that estimates conditional average treatment effects (CATE) within the trial population by leveraging external data from other trials or observational studies. The proposed method is robust: it can reduce the mean squared error relative to a trial-only CATE learner, and is guaranteed to recover the true CATE even when the external data are not aligned with the trial. Moreover, we introduce a procedure that combines the QR-learner with a trial-only CATE learner and show that it asymptotically matches or exceeds both component learners in terms of mean squared error. We examine the performance of our approach in simulation studies and apply the methods to a real-world dataset, demonstrating improvements in both CATE estimation and statistical power for detecting heterogeneous effects. } }
Endnote
%0 Conference Paper %T Robust estimation of heterogeneous treatment effects in randomized trials leveraging external data %A Rickard K.A. Karlsson %A Piersilvio De Bartolomeis %A Issa Dahabreh %A Jesse H. Krijthe %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-karlsson26a %I PMLR %P 1783--1791 %U https://proceedings.mlr.press/v300/karlsson26a.html %V 300 %X Randomized trials are typically designed to detect average treatment effects but often lack the statistical power to uncover individual-level treatment effect heterogeneity, limiting their value for personalized decision-making. To address this, we propose the QR-learner, a model-agnostic learner that estimates conditional average treatment effects (CATE) within the trial population by leveraging external data from other trials or observational studies. The proposed method is robust: it can reduce the mean squared error relative to a trial-only CATE learner, and is guaranteed to recover the true CATE even when the external data are not aligned with the trial. Moreover, we introduce a procedure that combines the QR-learner with a trial-only CATE learner and show that it asymptotically matches or exceeds both component learners in terms of mean squared error. We examine the performance of our approach in simulation studies and apply the methods to a real-world dataset, demonstrating improvements in both CATE estimation and statistical power for detecting heterogeneous effects.
APA
Karlsson, R.K., De Bartolomeis, P., Dahabreh, I. & Krijthe, J.H.. (2026). Robust estimation of heterogeneous treatment effects in randomized trials leveraging external data . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:1783-1791 Available from https://proceedings.mlr.press/v300/karlsson26a.html.

Related Material