Back

Rural dengue dynamics: the interplay of climate, built environment, and agriculture in Costa Rica

Glidden, C. K.; Southworth, E. K.; Shragai, T.; Rojas-Araya, D.; Troyo, A.; Chaves-Gonzalez, L. E.; Marin, R.; Vargas Roldan, I.; Mordecai, E. A.

2026-02-17 epidemiology
10.64898/2026.02.12.26346219 medRxiv
Show abstract

Dengue is one of the worlds highest-burden arboviral diseases. Although classically considered an urban disease, many regions experience a substantial dengue burden in rural areas. The combined influence of long-term climate, short-term weather variation, local built environments, and land-use gradients on dengue dynamics in rural settings remains poorly understood, limiting our ability to predict shifting risk under global change. Here, we investigate these dynamics in Costa Rica to disentangle how these interacting socio-environmental factors shape rural dengue transmission. We first use 22 years of canton-level (admin-2) case data to establish that both dengue cases and incidence are consistently higher in rural than in urban districts. Then, using ten years of district-level (admin-3) monthly case data and a Bayesian hierarchical modeling framework, we identify the climatic and land-use features most strongly associated with dengue risk. Temperature underlies broad spatial patterns in dengues urban-rural distribution, while precipitation effects differ between coasts, reflecting intercoastal climate zone contrasts rather than interactions between urbanization and water availability. Given suitable climate, even modest levels of built infrastructure substantially increase risk, but the relationship plateaus at higher levels of building volume. Dengue risk is also elevated in areas with high agricultural crop cover at low and mid elevations but not at higher, cooler elevations. Together these results suggest that high risk of rural dengue in Costa Rica result from climate suitability aligning with baseline levels of built infrastructure, with agriculture potentially emerging as a distinct driver of rural dengue transmission.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.