Broad Applications of Distributed Lag Non‐Linear Model in Public Health: A Comprehensive Review

Abstract This study reviews the use of the distributed lag non‐linear model (DLNM) in public health research, focusing on environmental‐exposure, health–outcome relationships, and providing recommendations for future studies. Embase, PubMed, Web of Science, and Scopus databases were searched for literature published from January 2020 to November 2024 using the DLNM to analyze the environmental exposures and health outcomes. After screening, removing duplicates, and reviewing full‐text articles, eligible studies were assessed using the DLNM to examine the health effects related to environmental exposure, particularly temperature and other environmental factors. From 2,847 studies, 274 studies from 36 countries were selected for analysis, primarily from China (164), Europe (28), and North America (23). There were 174 exclusive climate data sources, no standardized heat thresholds, and 131 unique sources of air pollutant data. Among the 53 adverse health outcomes identified using the DLNM, morbidity was the most prevalent (), followed by hospitalization (), hospital admission (), and emergency room visits (). This review highlights the utility of the DLNM in capturing complex temporal relationships between environmental exposure and health, clarifying lagged effects. Despite the challenges of standardization and computational efficiency, ongoing developments are enhancing the utility of the DLNM. Future research should focus on advanced statistical techniques, such as machine learning and neural networks, and extend applications to other environmental health scenarios.