Seasonal pattern of dengue infection in Singapore: A mechanism-based modeling and prediction

Seasonal variations in dengue infections are predominantly observed in tropical countries, where regular climate cycles influence vector populations and lead to substantial spikes in dengue incidence during specific periods each year. In Singapore, dengue cases typically peak in July, coinciding with the onset of the Southwest Monsoon season. Here, we developed a mechanistic SIR-SEI mathematical model that incorporates seasonal patterns through periodic vector recruitment rates to forecast future dengue outbreaks. By approximating the time-dependent infection rate parameter with a 10-term Fourier series, we predicted a rise in dengue cases in early 2024, later confirmed by actual case reports. Simulation results demonstrated strong alignment with observed data, showing a relative error of 4.99% within the maximum data range. Additionally, equivalence testing rejected the null hypothesis of dissimilarity, indicating that the model’s predictions closely fit the actual data. Further analysis revealed a 10-week lag between increased infection rates and subsequent incidence spikes, accompanied by an effective reproduction number that confirms dengue’s endemic status in Singapore. The model’s predictive accuracy and epidemiological insights highlight its potential as an effective early warning system, supporting government efforts in dengue outbreak prevention and management.