Modeling the impact of the El Niño–Southern Oscillation on the dynamics of dengue outbreaks in Argentina between 2018 and 2024

Abstract

Dengue fever has become an epidemiological concern worldwide, with 531,617 cases reported in Argentina during the 2023–2024 season. The El Niño–Southern Oscillation (ENSO) phenomenon has been linked to changes in dengue dynamics. This study assessed the relationships among ENSO, local climatic variables, and dengue outbreaks in Argentina from 2018 to 2024. A nationwide spatiotemporal analysis was conducted using dengue cases and monthly mean temperature, relative humidity, and rainfall. The Southern Oscillation Index (SOI) was used as the ENSO indicator (El Niño phase ≤ −0.5, neutral phase = −0.5 to 0.5, and La Niña phase ≥ 0.5). We fitted Hierarchical Bayesian models with Integrated Nested Laplace Approximation (INLA). A Negative Binomial distribution of dengue cases was used to account for overdispersion. Spatial dependence was modeled by using the Besag–York–Mollié 2 (BYM2) structure, adjusted for population. The final model achieved a Nagelkerke pseudo-R 2 of 0.537. Exposure-response curves of posterior probability and 95% credible intervals suggested that dengue risk increased when monthly mean temperatures were between 20° and 30° Celsius during the El Niño or neutral phase, with less consistent effects during the La Niña phase. Humidity >50% was associated with increased dengue risk during the La Niña phase, with less steep or absent associations during El Niño or neutral phases. Negative SOI scores were associated with an increased risk of dengue after controlling for the covariates. Dengue outbreaks in Argentina were independently associated with temperatures between 20 and 30 °C during the El Niño and neutral phases, humidity > 50% during the La Niña phase, and the El Niño phase of the ENSO. Incorporating ENSO indices into predictive models could enhance early warning systems and timely public health interventions in Argentina.