Date of Award

Summer 8-14-2017

Degree Type

Thesis

Degree Name

Master of Public Health (MPH)

Department

Public Health

First Advisor

Gerado Chowell

Second Advisor

Linh Dinh

Abstract

Introduction. Yellow fever is an infectious disease endemic in Africa and South America, and it is a vaccine-preventable disease. The outbreak of yellow fever in Angola, 2015-2016 posed a risk to the global health. In the early phase of the outbreak, mathematical modeling can be a useful tool to predicting how the outbreak might spread and what factors affect the course of the outbreak so that health policy makers and health organizations can react effectively. In this project, two mathematical models, generalized-growth model and generalized Richard model, were applied retrospectively to predict the case rises in the early phase and to forecast the short-term trend of the outbreak.

Data and method. The data was extracted from the WHO weekly outbreak situation reports, for the period from Dec 04, 2015 until Sep 29, 2016. The analysis was focused on the data of 3 provinces including the capital Luanda, Huambo, and Benguela. The generalized-growth model (GGM) was applied for fitting the model into the early phase of the outbreak. GGM and the generalized Richards model (GRM) was applied for short-term forecasting the trend of the outbreak’s peak. Simulations to project the uncertainty of the epidemic trend was conducted with MATLAB R2017a (The Mathworks, Inc.).

Results. Fitting GGM to the data of 3 provinces shows the patterns of case rises in the early phase widely vary from above-constant to sub-exponentially growth. Short-term forecasting with GGM becomes less capable after the epidemic reach its peak. GRM is more advantageous for the forecasting period containing the epidemic peak. However, it can be less effective when the outbreak is partly controlled or there are some interventions on the epidemic transmission.

Conclusion. GGM and GRM can be the useful tool to characterize the dynamic growth of an infectious outbreak in the early phase and produce a short-term forecast of dynamic patterns that shape the guidance and action for the outbreak control and interventions.

DOI

https://doi.org/10.57709/11246831

Download

Share

COinS