Quantifying the invasion risk of West Nile virus: Insights from a multi-vector and multi-host SEIR model

Martina Ferraguti; Afonso Dimas Martins; Yael Artzy-Randrup

doi:10.1016/j.onehlt.2023.100638

Quantifying the invasion risk of West Nile virus: Insights from a multi-vector and multi-host SEIR model

One Health. 2023 Oct 8:17:100638. doi: 10.1016/j.onehlt.2023.100638. eCollection 2023 Dec.

Authors

Martina Ferraguti^{1

2

3}, Afonso Dimas Martins⁴, Yael Artzy-Randrup²

Affiliations

¹ Departamento de Biología de la Conservación y Cambio Global, Estación Biológica de Doñana (EBD), CSIC, C/Américo Vespucio, 26, E-41092 Seville, Spain.
² Department of Theoretical and Computational Ecology (TCE), Institute for Biodiversity and Ecosystem Dynamics (IBED), University of Amsterdam, Science Park 904, 1098 XH Amsterdam, the Netherlands.
³ CIBER of Epidemiology and Public Health (CIBERESP), Madrid, Spain.
⁴ Department of Population Health Sciences, Faculty of Veterinary Medicine, University of Utrecht, Yalelaan 7, 3584 CL Utrecht, the Netherlands.

Abstract

The invasion of vector-borne diseases depends on the type of specific features of the vector and hosts at play. Within the Culex pipiens complex, differences in ecology, biology, and vector competence can influence the risk of West Nile virus (WNV) outbreaks. To determine which life-history traits affect WNV invasion into susceptible communities the most, we constructed an epidemiological Susceptible-Exposed-Infectious-Recovered model with three vector (eco)types, Culex pipiens pipiens, Cx. pip. molestus, and their hybrids, and two vertebrate hosts, birds (as amplifying hosts) and humans (as dead-end hosts). We investigated how differences in feeding preferences and transmission rates influenced WNV transmission across different habitats and two seasons (Spring versus Summer), to investigate the impact of increasing mosquitoes on the WNV transmission risk. Our results showed that vector feeding preferences and the transmission rate between mosquitoes and birds were the parameters that most influenced WNV invasion risk. Even though our model did not predict WNV invasion across any of the studied environments, we found that natural habitats displayed the highest susceptibility to WNV invasion. Pipiens (eco)type acted as the primary vector in all habitats. Hybrids, contrary to common opinion, showed minimal involvement in WNV transmission. However, it is important to interpret our study results with caution due to the possibility of idealized spring and summer seasons being reflected in the field-collected data. Our study could be a tool to enhance current vector surveillance and control programs by targeting specific vector types in specific environments, especially in natural habitat, which are most responsive to environmental shifts. The joint approach based on epidemiological modelling based on field collected data can help to reduce wasted time and economic costs while maximizing the efficiency of local public health authorities.

Keywords: Epidemiology; Feeding preferences; Mathematical modelling; Mosquito community; Passer domesticus; Vector-borne diseases.