Authentic Multinational Federated Time-to-Event Analyses Among People with HIV in Latin America

2026-08-06

AI Summary

This study addresses the challenge of conducting centralized time-to-event analyses in multinational HIV cohort studies, which is often hindered by barriers to cross-border individual-level data sharing. The project presents the first implementation of a federated time-to-event analysis framework in a real-world distributed setting across Latin America. Each participating center locally handles missing data via multiple imputation and fits stratified cause-specific Cox models, exchanging only aggregated summary statistics. Leveraging a surrogate likelihood approach and standardized coordination protocols, the method accurately reproduces centralized cumulative incidence functions without sharing individual patient data. The resulting Cox estimates demonstrate high concordance with those from pooled analyses and outperform conventional meta-analytic approaches, thereby validating the feasibility and effectiveness of federated learning for high-quality, privacy-preserving multinational HIV research.

Abstract

Multinational HIV cohort studies face regulatory barriers to cross-border sharing of individual participant data, limiting centralized pooled analyses. Federated statistical methods, which exchange only aggregated information, offer a privacy-preserving alternative but have rarely been examined in real-world distributed environments for HIV research. Here, we evaluate the feasibility and analytical performance of a communication-efficient federated framework within the Caribbean, Central, and South America Network for HIV epidemiology. Virologic failure and major regimen change after antiretroviral therapy initiation were analyzed as separate outcomes; for each, we estimated cumulative incidence functions (CIFs) and fit stratified cause-specific Cox proportional hazards models via a surrogate likelihood-based federated implementation. Each site imputed missing data, conducted local analysis, and shared only summary statistics according to a coordinated computation protocol. The federated approach exactly reproduced centralized CIFs and closely approximated centralized Cox regression estimates, outperforming conventional meta-analysis for both outcomes. These findings demonstrate that authentic federated analysis is feasible for multinational HIV research and can yield results closely aligned with centralized analysis while preserving data privacy. Post-hoc feedback from local analysts, however, indicated that broader adoption will require managing the logistical and coordination overhead and ensuring harmonized data collection and quality control across sites.

Problem

• multinational HIV research
• data privacy
• federated analysis
• time-to-event analysis
• regulatory barriers

Innovation

• federated analysis
• surrogate likelihood
• time-to-event analysis
• data privacy
• multinational HIV cohort