Robust inference for risk heterogeneity under group imbalance
Abstract
Population-level heterogeneity is ubiquitous in biomedical data, where differences across demographic or clinical subgroups can substantially alter risk patterns. For example, in intensive care unit (ICU) studies, the mortality risk associated with specific admission diagnoses can vary across ethnic groups. Existing approaches for detecting risk heterogeneity are often sensitive to baseline model misspecification and regularization bias, both of which commonly arise in practice. In this paper, we propose a robust framework for inferring risk heterogeneity between two populations using Neyman orthogonality, which yields estimators that are locally insensitive to nuisance parameter estimation error. The proposed estimator is consistent and asymptotically normal, and simulation studies demonstrate that in finite samples our method substantially reduces bias and improves inferential stability compared with standard likelihood-based approaches. In an application to the eICU Collaborative Research Database, our method reveals clinically meaningful ethnicity-specific heterogeneity in admission diagnoses for in-hospital mortality that standard likelihood-based methods fail to detect.
Related
- Bayesian inference of risk differences for multi-group bilateral data
- On the Gradient Heterogeneity Dynamics of Adversarially Robust Federated Regression
- Heterogeneity-calibrated Byzantine-robust distributed composite quantile regression
- Robust Inference for Dyadic Data with Spatially Dependent Nodes
- Global Inference for High Dimensional Quantile Regression with Heterogeneity
BibTeX
@misc{xu2026robust,
title = {{Robust inference for risk heterogeneity under group imbalance}},
author = {Mengqi Xu and Subha Maity and Joel Dubin},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2606.00797},
eprint = {2606.00797},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.00797},
}
RIS
TY - UNPB
TI - Robust inference for risk heterogeneity under group imbalance
AU - Mengqi Xu
AU - Subha Maity
AU - Joel Dubin
PY - 2026
DA - 2026/10/13
PB - arXiv
DO - 10.48550/arxiv.2606.00797
UR - https://arxiv.org/abs/2606.00797
AB - Population-level heterogeneity is ubiquitous in biomedical data, where differences across demographic or clinical subgroups can substantially alter risk patterns. For example, in intensive care unit (ICU) studies, the mortality risk associated with specific admission diagnoses can vary across ethnic groups. Existing approaches for detecting risk heterogeneity are often sensitive to baseline model misspecification and regularization bias, both of which commonly arise in practice. In this paper, we propose a robust framework for inferring risk heterogeneity between two populations using Neyman orthogonality, which yields estimators that are locally insensitive to nuisance parameter estimation error. The proposed estimator is consistent and asymptotically normal, and simulation studies demonstrate that in finite samples our method substantially reduces bias and improves inferential stability compared with standard likelihood-based approaches. In an application to the eICU Collaborative Research Database, our method reveals clinically meaningful ethnicity-specific heterogeneity in admission diagnoses for in-hospital mortality that standard likelihood-based methods fail to detect.
ER -