Fig 1.
Geographic distribution of snakebite cases in Mexico (2018).
(A) Total number of reported snakebites per state. (B) Population-adjusted snakebite incidence per 100,000 inhabitants. Colors indicate intensity of occurrence, with darker shades representing higher values. The base layer of the map showing Mexican municipalities was obtained from: https://idegeo.centrogeo.org.mx/geovisor. (INEGI, 2024). Terms of use: https://www.inegi.org.mx/contenidos/inegi/doc/terminos_info.pdf.
Table 1.
Parameter estimates from the zero-inflated Poisson conditional autoregressive model of snakebite occurrence in Mexico (2018). Estimates include posterior means and 95% credible intervals (2.5% and 97.5% quantiles) for fixed effects, zero-inflation parameters (Ω), and variance components (τ² for spatial variation, σ² for nonspatial variation). Environmental suitability (ES) was calculated for B. asper. SLI: Social Lag Index.
Fig 2.
Model validation: observed (SBI) versus predicted snakebite incidence (PSBI) rates.
The solid line represents the 1:1 relationship. Labeled points indicate municipalities with substantial deviation from model predictions: Zapotitlán de Méndez, Puebla (blue), San Luis del Cordero, Durango (purple), San Miguel Yotao, Oaxaca (green), Santa María Ixcatlán, Oaxaca (yellow) and Santa Ana Ateixtlahuaca, Oaxaca (brown).
Fig 3.
Spatial distribution of predicted snakebite risk after controlling for hospital-related reporting bias.
This map shows model predictions when hospital density is standardized to zero across all municipalities while maintaining spatial autocorrelation effects, thereby revealing the underlying risk pattern independent of healthcare access. Grey represents municipalities with zero counts of predicted snakebites or predicted snakebite incidence. The base layer of the map showing Mexican municipalities was obtained from: https://idegeo.centrogeo.org.mx/geovisor. (INEGI, 2024). Terms of use: https://www.inegi.org.mx/contenidos/inegi/doc/terminos_info.pdf.