Disentangling snakebite dynamics in Colombia: How does rainfall and temperature drive snakebite temporal patterns?

Author summarySnakebite envenoming is a neglected tropical disease characterized by its high burden on the rural population and high mortality if antivenom is not administered. The ecology of this health problem is not well-understood; however, approaches to address the temporal are growing. So far, we know that rainfall can play an important role in driving snakebite incidence seasonality at a national scale. Moreover, geographical areas with high rainfall are more prone to have high snakebite risk, but the spatial heterogeneity of the temporal association (i.e., if there are different seasonal patterns of rainfall-incidence association in different geographical areas of a country) is just starting to emerge in the literature. By formulating and fitting compartmental models to data, we generated a flexible framework that relies on temporal resolved datasets and a compartmental mathematical model to understand the effect of climatic covariates (such as rainfall and temperature) driving snakebite dynamics in space and time. We applied this framework to Colombia and found that dry seasons cause a decrease in snakebite incidence: Rainfall only drives snakebite dynamics in regions with marked dry seasons. Thus, rainfall is a limiting resource of the system, and its effect is not spatially homogeneous. On the other hand, the temperature had no significant effect driving snakebite incidence. Our modeling approach can also be used to estimate the effect of climate anomalies on snakebite incidence and has the potential to be used as a tool to monitor snakebite incidence. The role of climate driving zoonotic diseases' population dynamics has typically been addressed via retrospective analyses of national aggregated incidence records. A central question in epidemiology has been whether seasonal and interannual cycles are driven by climate variation or generated by socioeconomic factors. Here, we use compartmental models to quantify the role of rainfall and temperature in the dynamics of snakebite, which is one of the primary neglected tropical diseases. We took advantage of space-time datasets of snakebite incidence, rainfall, and temperature for Colombia and combined it with stochastic compartmental models and iterated filtering methods to show the role of rainfall-driven seasonality modulating the encounter frequency with venomous snakes. Then we identified six zones with different rainfall patterns to demonstrate that the relationship between rainfall and snakebite incidence was heterogeneous in space. We show that rainfall only drives snakebite incidence in regions with marked dry seasons, where rainfall becomes the limiting resource, while temperature does not modulate snakebite incidence. In addition, the encounter frequency differs between regions, and it is higher in regions where Bothrops atrox can be found. Our results show how the heterogeneous spatial distribution of snakebite risk seasonality in the country may be related to important traits of venomous snakes' natural history.