EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Privacy-enhancing sequential learning under heterogeneous selection bias in multi-site electronic health records data

Ritoban Kundu, Maxwell Salvatore, Kumar Kshitij Patel, Lucila Ohno-Machado iD, Hyunghoon Cho, Xu Shi iD, Bhramar Mukherjee iD

Journal of the American Medical Informatics Association · 2026

DOI: 10.1093/jamia/ocag083 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, 2026. Verfasst von Ritoban Kundu und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biostatistics, Epidemiology & Informatics, University of Pennsylvania, Philadelphia, PA 19104, United States. Hauptthemen laut MEDLINE: Electronic Health Records, Neoplasms, Confidentiality, Machine Learning. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, NIH HHS, MI-CARES.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

To develop privacy-enhancing statistical methods for estimating disease risk parameters across multiple electronic health record (EHR) sites with heterogeneous selection mechanisms, avoiding individual-level data sharing. We illustrate their utility via a cross-biobank analysis of smoking and 97 cancer subtypes using NIH All of Us (AOU) and Michigan Genomics Initiative (MGI) data sites. Materials and Methods Distributed health platforms often render centralized algorithms infeasible due to patient privacy protection. We propose Sequential Pseudo-Likelihood (SPL) and Sequential Augmented Inverse Probability Weighting (SAIPW) to adjust for selection bias using summary statistics shared across sites and external population information. SAIPW employs flexible auxiliary models for multiple robustness. We compared SPL and SAIPW against unweighted and centralized/meta-learning benchmarks in simulations, applying them to harmonized MGI (n = 50 935) and AOU (n = 241 563) data.

Results

Unweighted estimators exhibited substantial bias. SPL and SAIPW yielded unbiased estimates with valid coverage, with SAIPW remaining robust to selection model misspecification. Both approaches showed negligible efficiency loss relative to centralized methods. Meta-learning methods proved unstable for rare outcomes. Real-data analyses consistently identified strong associations between smoking and lung, bladder, and larynx cancers.

Discussion

These findings highlight the necessity of adjusting for site-specific selection biases in distributed health networks. SPL and SAIPW offer practical, scalable solutions that bypass the instability of meta-analysis for rare events, successfully harmonizing diverse biobanks while strictly enhancing patient privacy.

Conclusion

Our framework enables valid, privacy-enhancing inference across EHR sites subject to heterogeneous selection, facilitating scalable, distributed research using real-world data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Electronic Health RecordsNeoplasmsConfidentialityMachine LearningHumansSelection BiasAlgorithmsSmokingLikelihood Functions

Stichwörter der Autor:innen

decentralized learningelectronic health recordsmultiple robustnessselection biassequential learning

Autor:innen und Institutionen

  • Department of Biostatistics, Epidemiology & Informatics, University of Pennsylvania, Philadelphia, PA 19104, United States.
  • Department of Genetics, University of Pennsylvania, Philadelphia, PA 19104, United States.
  • Foundations of Data Science, Yale University, New Haven, CT 06511, United States.
  • Biomedical Informatics & Data Science, Yale University, New Haven, CT 06510, United States.
  • School of Medicine, Yale University, New Haven, CT 06510, United States.
  • Department of Biostatistics, University of Michigan, Ann Arbor, MI 48109, United States.

ORCID-Profile: Lucila Ohno-Machado, Xu Shi, Bhramar Mukherjee

Förderung

  • NCI NIH HHS
  • NIH HHS
  • MI-CARES
  • National Science Foundation and MI-CARES
  • National Science Foundation
  • the National Cancer Institute
  • National Cancer Institute

Zitieren

APA

Kundu, R., Salvatore, M., Patel, K. K., Ohno-Machado, L., Cho, H., Shi, X., & Mukherjee, B. (2026). Privacy-enhancing sequential learning under heterogeneous selection bias in multi-site electronic health records data. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocag083

Vancouver

Kundu R, Salvatore M, Patel KK, Ohno-Machado L, Cho H, Shi X, et al. Privacy-enhancing sequential learning under heterogeneous selection bias in multi-site electronic health records data. Journal of the American Medical Informatics Association. 2026. doi:10.1093/jamia/ocag083

BibTeX
@article{Kundu2026,
  author = {Kundu, Ritoban and Salvatore, Maxwell and Patel, Kumar Kshitij and Ohno-Machado, Lucila and Cho, Hyunghoon and Shi, Xu and Mukherjee, Bhramar},
  title = {Privacy-enhancing sequential learning under heterogeneous selection bias in multi-site electronic health records data},
  journal = {Journal of the American Medical Informatics Association},
  year = {2026},
  doi = {10.1093/jamia/ocag083}
}
Kostenlos starten →