ENSO-Driven Mortality: A Cross-Population Analysis Using the Modified Lee–Carter Model

Interannual climate variability poses increasing challenges to public health, with many extreme weather events closely linked to anomalous ocean–atmosphere interactions across the Pacific Ocean, Atlantic Ocean, and Indian Ocean. By taking the El Niño–Southern Oscillation (ENSO) in the Pacific as an entry point, this study incorporates the Southern Oscillation index (SOI), Niño3, Niño4, and Niño3.4 indices into the Lee–Carter model, a stochastic mortality forecasting framework that decomposes log mortality rates into age-specific effects and a time-varying mortality index capturing overall mortality trends. Under different mortality assumptions, the model evaluates how ENSO-related climate variability influences mortality rates in populations from four Pacific Rim countries [Australia, Canada, Japan, and the United States of America (USA)] and two regions of China: the Hong Kong Special Administrative Region (SAR) and Taiwan Province of China. Under a negative binomial specification, ENSO significantly influences mortality in a seasonally and regionally heterogeneous manner. For Australia, Canada, Japan, and Taiwan Province of China, mortality responds most strongly to SOI fluctuations. La Niña-like conditions are generally associated with higher mortality, particularly during winter in Australia and during summer in Canada, Japan, and Taiwan Province of China, while El Niño-like conditions elevate mortality risks for older adults in specific seasons. In Hong Kong SAR of China and the USA, the dual-index model shows that central-Pacific variability (Niño4) plays a dominant role in shaping the ENSO–mortality relationship, and its effects are season-dependent. The paper concludes with policy recommendations, including prioritizing protection for elderly populations, establishing a dynamic climate–mortality monitoring platform, and developing region-specific disaster prevention strategies.