Avian Influenza: Modeling, Analysis and Implications for Control

University Of Florida, Gainesville FL

Investigators

Abstract

This interdisciplinary study integrates the efforts of mathematicians and biologists in developing models and addressing questions related to the complex ecology and evolution of avian influenza. Highly pathogenic avian influenza (HPAI) of the subtype H5N1, which now threatens to mutate and cause a major pandemic, evolves from the low pathogenic avian influenza (LPAI) while circulating in poultry. Understanding evolution of viral pathogens requires a detailed analysis of interactions between different genetic types of the virus in the context of host population structure. The first part of this project studies the interplay of HPAI and LPAI in wild birds and domestic birds, particularly questions related to the cross-immunity that LPAI potentially provides to HPAI. The principal investigator and her colleagues develop and analyze multi-strain age-structured models in a two-host-species system (corresponding to domestic bird stocks, and wild bird populations) to study competitive exclusion and persistence of HPAI and LPAI strains. Furthermore, mathematical techniques are developed that elucidate the dynamical behavior of the system. Preliminary results suggest that time-since-recovery structure coupled with cross-immunity may be responsible for more complex, oscillatory behavior of LPAI and HPAI when there is coexistence in birds. The second part of the project builds on these models to study the symbiotic effect of multiple control strategies applied to control highly pathogenic H5N1 influenza in poultry. Preliminary results suggest that strategies applied to poultry, such as vaccination and culling, should be significantly more efficient than strategies applied to humans, such as wearing protective gear, in reducing the prevalence of H5N1 among humans. Further efforts are made to understand the interplay of vaccination of poultry and culling as the two primary control measures applied today, in order to provide quantitative insight into alternative potential measures of disease control. This project produces models of highly pathogenic avian influenza (bird flu, caused by an H5N1 influenza virus) that are well-grounded in the complex biology of the virus and validated with available data. Such models are required both to understand the basic disease dynamics, and to develop and implement effective control measures. The models are designed to fit available data on the cumulative number of human H5N1 cases and to be able to perform reasonable projections of future cases. Furthermore, the threat of a world-wide human epidemic (pandemic) caused by H5N1 requires strict control measures for the virus. These control measures (especially culling, the killing of all birds in a poultry operation when H5N1 infection is detected) cause significant economic losses in the poultry industry in many parts of the world. The models and methods developed as a part of this project evaluate the effectiveness of alternative control measures in reducing the number of H5N1 human cases. Analysis of these mathematical models expands our ability to choose the best control measures. In addition, the project studies the emergence of highly pathogenic strains from low pathogenic strains and their circulation in wild birds and domestic birds. The evolution of highly pathogenic strains capable of infecting humans is a major pathway through which a pandemic strain can emerge. Understanding the circumstances that foster such evolution increases our preparedness in identifying and combating potential pandemic threats caused by avian influenza.