Federated causal inference for effect estimation in healthcare: methods, challenges, and applications
Federated causal inference enables treatment-effect estimation across healthcare systems without centralizing patient-level data. We review methods for binary, continuous, and time-to-event outcomes through three lenses: estimand, distributed-computation protocol, and privacy mechanism. We compare major strategies, clarify when federated analyses can approximate pooled results, and highlight unresolved challenges in confounding, heterogeneity, survival estimands, privacy–utility trade-offs, validation, and software. This framework supports rigorous evaluation of federated causal methods in healthcare.
Authors
- Colin Magdamo (ORCID: https://orcid.org/0000-0001-8965-4630)
- Kyra Gan (ORCID: https://orcid.org/0000-0002-5147-8747)
- Mark W. Albers (ORCID: https://orcid.org/0000-0001-7855-3455)
- Meihan Wei (ORCID: https://orcid.org/0009-0009-5631-8855)
- Chengxi Zang (ORCID: https://orcid.org/0000-0002-8244-9551)
- Haoyang Li (ORCID: https://orcid.org/0000-0003-3544-5563)
- Fei Wang (ORCID: https://orcid.org/0000-0001-7948-990X)
- Max Sunog
- Zheqi Lyu
- Fuying Ding
- Simai Cheng
- Wenhan Zheng (ORCID: https://orcid.org/0009-0007-5092-7604)
- Jie Xu
Institutions
- Harvard University (US)
- Cornell University (US)
- University of Florida Health (US)
- University of Florida (US)
- Massachusetts General Hospital (US)
- Weill Cornell Medicine (US)
- Cornell Tech (US)
Publication Details
- Journal
- npj Artificial Intelligence
- Published
- 2026-10-17
- DOI
- https://doi.org/10.1038/s44387-026-00168-1
- Primary Topic
- Advanced Causal Inference Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00