Large-scale vector-borne disease agent-based model, with application to Chikungunya in Colombia

This document presents the development of a large-scale agent-based model to represent vector-borne disease transmission dynamics. Specifically the model represents the transmission of chikungunya in Colombia. Due to their similarities, the model can also be applied to simulate dengue epidemics. The aim of this model is to contribute to the knowledge of chikungunya, to reproduce realistic epidemics, and to quantify the impact of vector control programs to halt the spread of the disease. Chikungunya is a disease transmitted by the \textit{aedes} mosquitoes, particularly \aegypti and \textit{Aedes albopictus}. Chikungunya symptoms are similar to dengue, but it is characterized by acute join-pain that can last for years. In Colombia, the \aegypti is found in larger proportions than the \textit{Aedes albopictus}. The \aegypti mainly obtains its food from humans, hence it is often considered a residential mosquito. The model proposed in this thesis represents humans and mosquitoes. Humans are represented by agents whose health status can be classified in a S-E-I-R structure (Susceptible, Exposed, Infectious, Recovered). Whereas mosquitoes are represented by a homogeneous meta population model with the S-E-I compartments. In the model, the virus transmission occurs in specific locations such as households, workplaces, or schools. In each location, the number of mosquitoes are computed based on temperature and the human density. Mosquitoes and humans can transmit the infection to each other with specific probabilities determined in the model. Transmission occurs when an infectious agent visits a place with susceptible vectors, or when a susceptible agent visits a place with infectious mosquitoes. These visits are determined by each agent's activities that are assigned in a synthetic population, these activities include: household visits, school attendance, work attendance, and travel. A synthetic population was developed to represent a realistic population of Colombia. The synthetic population represents the population of the 1122 municipalities and 33 departments of the country. Additionally, the synthetic population reproduces daily activities for each individual based on the census data. Human mobility was also represented in the model implementing a calibrated gravity model to represent air travel. The model's parameters were calibrated to represent chikungunya dynamics reported in the Riohacha, Guajira. Some of the parameter values were obtained from the literature while others were adjusted using an optimization algorithm. This calibrated model was used to estimate the impact of vector control strategies in the city of Santa Marta, Magdalena. The control parameters in the model were modified to determine improvements to design optimal vector control strategies. Lastly, the model was simulated in a national-scale to evaluate the burden of the chikungunya with and without vector control strategies.

• Autores: Camargo España, Guido Felipe
• Tipo de recurso: Doctoral thesis
• Fecha de publicación: 2015
• Institución: Universidad Nacional de Colombia
• Repositorio: Universidad Nacional de Colombia
• Idioma: spa
• OAI Identifier: oai:repositorio.unal.edu.co:unal/55902
• Acceso en línea: https://repositorio.unal.edu.co/handle/unal/55902; http://bdigital.unal.edu.co/51416/
• Palabra clave: 61 Ciencias médicas; Medicina / Medicine and health; 62 Ingeniería y operaciones afines / Engineering; Chikungunya; Modelos Basados en Agentes; Epidemiologia
• Rights: openAccess
• License: Atribución-NoComercial 4.0 Internacional