Nonlinear and Delayed Effects of Rainfall and Temperature on Dengue Incidence in Eastern Visayas, Philippines: A Bayesian Distributed Lag Nonlinear Model Approach
Background: Dengue remains a major public health challenge in Eastern Visayas, Philippines, yet climate-driven transmission dynamics remain underexplored in the region. Methods: This retrospective ecological time-series study analyzed weekly surveillance and meteorological data from 2019 to 2023. Nonlinear exposure–response relationships and lagged associations between rainfall, temperature, and dengue incidence were modeled using a Distributed Lag Nonlinear Model (DLNM) within a Bayesian framework fitted via Integrated Nested Laplace Approximation (INLA), assessing lag effects up to 10 weeks. Results: Weekly rainfall strongly influenced dengue risk, peaking markedly at 200 mm with a cumulative Relative Risk (RR) of 19.4 (95% CrI: 10.3–36.5). Higher precipitation levels showed a downward risk trajectory, consistent with larval flushing. Temperature displayed a suggestive bimodal pattern with elevated risk at 24°C and 28–29°C across 4–6-week lags, though near-average temperatures (e.g., 28.5°C) showed no statistically strong cumulative risk changes.Conclusion: Dengue risk in Eastern Visayas is driven by precise rainfall thresholds and thermal lag windows. Incorporating 200 mm precipitation and 28–29°C temperature triggers into proactive climate-based surveillance—pending external validation—can enhance regional public health preparedness and local outbreak response.