FLAT-WHITE: from target trial emulation to federated high-throughput causal machine learning using decentralized real-world data
18 affiliations
- 1Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD USA
- 2The Center for Health AI and Synthesis of Evidence (CHASE), University of Pennsylvania, Philadelphia, PA USA
- 3The Graduate Group in Applied Mathematics and Computational Science, School of Arts and Sciences, University of Pennsylvania, Philadelphia, PA USA
- 4Department of Biostatistics, Epidemiology, and Informatics, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA USA
- 5Department of Neurology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA USA
- 6Leonard Davis Institute for Health Economics, University of Pennsylvania, Pennsylvania, PA USA
- 7Department of Biostatistics and Health Data Science, University of Pittsburgh, Pittsburgh, PA USA
- 8Merck & Co Inc, Rahway, NJ USA
- 9Weill Cornell Medicine, Cornell University, New York, NY USA
- 10Department of Biostatistics and Health Data Science, Indiana University, Bloomington, IN USA
- 11Department of Health Outcomes and Biomedical Informatics, College of Medicine, University of Florida, Gainesville, FL USA
- 12Department of Biomedical Informatics and Data Science, School of Medicine, Yale University, New Haven, CT USA
- 13Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
- 14Statistical Research and Data Science Center, Pfizer Inc, New York, NY USA
- 15College of Population Health, Thomas Jefferson University, Philadelphia, PA USA
- 16Real-World Evidence and Epidemiology, Chief Medical Office, Pfizer, Collegeville, PA USA
- 17Penn Institute for Biomedical Informatics, University of Pennsylvania, Philadelphia, PA USA
- 18Penn Medicine Center for Evidence-based Practice (CEP), Philadelphia, PA USA
Abstract
Randomized clinical trials remain the gold standard for evaluating treatment effects but are often infeasible for rapidly assessing drug repurposing opportunities, particularly in complex conditions such as Alzheimer’s disease. Target trial emulation (TTE) provides a principled framework for causal inference with electronic health records and, when coupled with modern machine learning, enables scalable analyses across distributed health systems. We present FLAT-WHITE (Federated Lossless Algorithm for Target Trial Emulation with Overlap Weighting for Healthcare Data Integration via Trusted and Efficient Communication), a novel federated causal machine learning pipeline for high-throughput treatment effect estimation without sharing patient-level data. FLAT-WHITE is a one-shot, lossless algorithm, requiring a single communication round while achieving results equivalent to centralized pooled analyses. With overlap weighting on the propensity score, it robustly addresses heterogeneity in treatment assignment and ensures valid causal inference in decentralized settings. We demonstrate FLAT-WHITE in a Alzheimer’s drug repurposing study, scanning 181 candidate drugs using data from four health learning systems encompassing approximately 150,000 patients with mild cognitive impairment. Findings were validated externally in 339,180 patients from the TriNetX global research network. FLAT-WHITE enables rapid, large-scale, and secure evaluation of treatment effects and establishes a foundation for collaborative clinical research across diverse healthcare systems.
Related
- Learning to Refer: Inference-Time Collaboration in Decentralized Federated Learning
- AI-Driven Dynamic Trust in Zero Trust Architecture: A Systematic Literature Review of Machine Learning, Federated Learning and Governance Challenges
- Hit Selection Using SSMD-Based Machine Learning Performance Metrics in High-Throughput Screening Assays
- Decentralized Bandit Learning of Markov Perfect Equilibria
- Decentralized Strategies for Finite Population LQG Social Control: A Reinforcement Learning Approach
BibTeX
@article{tong2026flatwhite,
title = {{FLAT-WHITE: from target trial emulation to federated high-throughput causal machine learning using decentralized real-world data}},
author = {Jiayi Tong and Lu Li and Dazheng Zhang and Kyra S. O’Brien and Yiwen Lu and Qiong Wu and Ting Zhou and Bingyu Zhang and Yuqing Lei and Huilin Tang and Naimin Jing and Haoyang Li and Zhenxing Xu and Yu Huang and Yu Hu and Yujia Zhou and Fongci Lin and Ying Jiang and Haitao Chu and Yong Chen and Fei Wang and Jiang Bian and Hua Xu and David A. Wolk and Yong Chen},
journal = {npj Health Systems},
year = {2026},
month = oct,
doi = {10.1038/s44401-026-00160-6},
url = {https://doi.org/10.1038/s44401-026-00160-6},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - FLAT-WHITE: from target trial emulation to federated high-throughput causal machine learning using decentralized real-world data
AU - Jiayi Tong
AU - Lu Li
AU - Dazheng Zhang
AU - Kyra S. O’Brien
AU - Yiwen Lu
AU - Qiong Wu
AU - Ting Zhou
AU - Bingyu Zhang
AU - Yuqing Lei
AU - Huilin Tang
AU - Naimin Jing
AU - Haoyang Li
AU - Zhenxing Xu
AU - Yu Huang
AU - Yu Hu
AU - Yujia Zhou
AU - Fongci Lin
AU - Ying Jiang
AU - Haitao Chu
AU - Yong Chen
AU - Fei Wang
AU - Jiang Bian
AU - Hua Xu
AU - David A. Wolk
AU - Yong Chen
PY - 2026
DA - 2026/10/27
JO - npj Health Systems
DO - 10.1038/s44401-026-00160-6
UR - https://doi.org/10.1038/s44401-026-00160-6
AB - Randomized clinical trials remain the gold standard for evaluating treatment effects but are often infeasible for rapidly assessing drug repurposing opportunities, particularly in complex conditions such as Alzheimer’s disease. Target trial emulation (TTE) provides a principled framework for causal inference with electronic health records and, when coupled with modern machine learning, enables scalable analyses across distributed health systems. We present FLAT-WHITE (Federated Lossless Algorithm for Target Trial Emulation with Overlap Weighting for Healthcare Data Integration via Trusted and Efficient Communication), a novel federated causal machine learning pipeline for high-throughput treatment effect estimation without sharing patient-level data. FLAT-WHITE is a one-shot, lossless algorithm, requiring a single communication round while achieving results equivalent to centralized pooled analyses. With overlap weighting on the propensity score, it robustly addresses heterogeneity in treatment assignment and ensures valid causal inference in decentralized settings. We demonstrate FLAT-WHITE in a Alzheimer’s drug repurposing study, scanning 181 candidate drugs using data from four health learning systems encompassing approximately 150,000 patients with mild cognitive impairment. Findings were validated externally in 339,180 patients from the TriNetX global research network. FLAT-WHITE enables rapid, large-scale, and secure evaluation of treatment effects and establishes a foundation for collaborative clinical research across diverse healthcare systems.
SN - 3005-1959
ER -