Privacy-preserving causal mediation analysis using distributed electronic health record networks

2026-07-20

AI Summary

This study addresses the challenge of conducting causal mediation analysis across institutions due to privacy and governance constraints that restrict sharing of individual-level electronic health records (EHRs). To overcome this limitation, the authors propose the first federated framework integrating renewable learning with counterfactual causal mediation analysis, enabling collaborative estimation of natural direct and indirect effects without exchanging raw patient data—only low-dimensional summary statistics are shared. By unifying federated learning with causal inference, the method accurately replicates centralized analysis results in both simulations and real-world EHR data from 32,146 patients. Empirical findings reveal that body mass index (BMI) plays a relatively weak mediating role in the pathway through which GLP-1 receptor agonists reduce HbA1c levels.

Abstract

Electronic health record (EHR) networks provide unprecedented opportunities to study treatment mechanisms at scale, but mediation analyses across institutions are often hindered by privacy and governance constraints that restrict sharing of patient-level data. We developed a privacy-preserving federated mediation framework that enables estimation of natural direct and indirect effects without exchanging individual-level records across participating sites. The proposed approach integrates renewable learning with counterfactual causal mediation analysis, allowing institutions to collaboratively investigate treatment mechanisms using only low-dimensional summary statistics. Both simulation studies and the real-world application demonstrated that the federated estimator closely reproduced pooled-data results while preserving patient privacy. We applied the method to 32,146 patients in the Indiana Network for Patient Care to evaluate the extent to which body mass index (BMI) mediates the effect of GLP-1 receptor agonist on glycated hemoglobin (HbA1c) reduction. The BMI-mediated pathway accounted for only a small proportion of the overall treatment effect, suggesting that most glycemic improvement occurred through mechanisms other than weight loss.

Problem

• privacy-preserving
• causal mediation analysis
• electronic health records
• federated learning
• data sharing constraints

Innovation

• federated mediation analysis
• privacy-preserving
• causal mediation
• electronic health records
• counterfactual inference