Correlation-weighted communicability curvature as a structural driver of dengue spread: A Bayesian spatial analysis of Recife (2015–2024)

We investigate whether functional connectivity in urban road networks explains dengue incidence in Recife, Brazil (2015–2024), beyond traditional adjacency-based spatial dependence. For each neighborhood, we compute the average communicability curvature , a graph-theoretic measure capturing multiscale accessibility through redundant network paths. The curvature metric is incorporated into Negative Binomial models, fixed-effects regressions, SAR/SAC spatial models, and a hierarchical INLA/BYM2 specification. Across all frameworks, curvature emerges as the strongest and most stable predictor of dengue risk. In the BYM2 model, the structured spatial component collapses ( ϕ ≈ 0 ), indicating that spatial variation traditionally attributed to CAR adjacency effects is largely absorbed by functional network connectivity. Rather than eliminating spatial dependence, the results suggest a reparametrization of space: dengue diffusion in Recife is structured less by geometric contiguity and more by network-mediated urban connectivity. • Introduces correlation-weighted communicability curvature • Curvature reduces spatial effect in BYM2-INLA models • Annual curvature shifts reveal transmission architecture • NDWI significant but affected by cloud cover noise • Integrates urban networks and Bayesian spatial modeling