Publications

LEVERAGING LARGE LANGUAGE MODELS TO ACCELERATE SYSTEMATIC REVIEWS IN ENVIRONMENTAL HEALTH

Abstract

Systematic reviews are a cornerstone of evidence-based science, synthesizing findings across multiple studies to provide comprehensive summaries of current knowledge [1, 2, 3]. In environmental health, where complex exposures interact with diverse populations and health outcomes, systematic reviews play an essential role in characterizing causal relationships, identifying knowledge gaps, and supporting policy decisions [4, 5, 6]. The rigor of systematic reviews grounded in transparent, reproducible methods [6] makes them invaluable for establishing causal inference in environmental epidemiology, where experimental designs are often infeasible or unethical [7, 8, 9].
Timely systematic reviews are crucial for science-based decision-making [9, 7], particularly in rapidly evolving fields such as environmental health. As new studies emerge and our understanding of emerging contaminants and health outcomes deepens, decision-makers must have access to current syntheses of evidence. Delays in conducting reviews can result in policies based on incomplete or outdated information, potentially affecting public health outcomes and environmental protection efforts. The ability to rapidly synthesize evidence is therefore not merely a convenience but a prerequisite for responsive, informed governance.

Date
2026
Authors
Deborah Khider, Roselyn Tanghal