One Health approach to explore the association of leptospirosis in rural Brazil with rodent species richness

Reservoir-host species richness has been linked to the emergence of zoonotic infectious diseases, including leptospirosis. Myocastor coypus is a potentially important reservoir host of Leptospira in wild habitats. Nonetheless, studies on the associations between these ecological factors and leptospirosis occurrence are limited. We evaluated the associations of rodent species richness and predicted M. coypus habitat suitability with leptospirosis occurrence in rural Brazil. Leptospirosis cases in rural Brazil from 2015 to 2019 were used as the outcome variable. We applied the maximum entropy modeling method (MaxEnt) to predict the habitat suitability of M. coypus and used stacked species distribution models to estimate rodent species richness. Logistic regression models that considered spatial autocorrelations were used to assess associations. In the primary Besag–York–Mollié (BYM) spatial model, municipalities in the highest tertile of rodent species richness had 77% higher odds of leptospirosis occurrence than those in the lowest tertile (odds ratio [OR] = 1.77, 95% Bayesian credible interval [BCI]: 1.33–2.21). These findings suggest that higher rodent species richness is associated with greater odds of leptospirosis occurrence in rural Brazil. Further studies should investigate this association across multiple spatiotemporal scales using information on reservoir-host abundance, infection, and human exposure.