Avian influenzas are zoonoses, or pathogens borne by wildlife and livestock that

can also infect people. In recent decades, and especially since the emergence of highly pathogenic avian influenza (HPAI) H5N1 in 1996, these diseases have become a significant threat to animal and public health across the world. HPAI H5N1 has caused severe damage to poultry populations, killing, or prompting the culling of, millions of birds in Asia, Africa, and Europe. It has also infected hundreds of people, with a mortality rate of approximately 50%. This dissertation focuses on the ecological and socioeconomic drivers of avian influenza risk, particularly in China, the most populous country to be infected. Among the most significant ecological risk factors are landscapes that serve as “mixing zones” for wild waterfowl and poultry, such as rice paddy, and nearby lakes and wetlands that are important breeding and wintering habitats for wild birds. Poultry outbreaks often involve cross infections between wild and domesticated birds. At the international level, trade in live poultry can spread the disease, especially if the imports are from countries not party to trade agreements with well-developed biosecurity standards. However, these risks can be mitigated in a number of ways. Protected habitats, such as Ramsar wetlands, can segregate wild bird and poultry populations, thereby lowering the chance of interspecies transmission. The industrialization of poultry production, while not without ethical and public health problems, can also be risk-reducing by causing wild-domestic segregation and allowing for the more efficient application of surveillance, vaccination, and other biosecurity measures. Disease surveillance is effective at preventing the spread of avian influenza, including across international borders. Economic modernization in general, as reflected in rising per-capita GDP, appears to mitigate avian influenza risks at both the national and sub-national levels. Poultry vaccination has been effective in many cases, but is an incomplete solution because of the practical difficulties of sustained and widespread implementation. The other popular approach to avian influenza control is culling, which can be highly expensive and raise ethical concerns about large-scale animal slaughter. Therefore, it is more economically efficient, and may even be more ethical, to target the socio-ecological drivers of avian influenza risks, including by implementing the policies discussed here.

Mathematical epidemiology, one of the oldest and richest areas in mathematical biology, has significantly enhanced our understanding of how pathogens emerge, evolve, and spread. Classical epidemiological models, the standard for predicting and managing the spread of infectious disease, assume that contacts between susceptible and infectious individuals depend on their relative frequency in the population. The behavioral factors that underpin contact rates are not generally addressed. There is, however, an emerging a class of models that addresses the feedbacks between infectious disease dynamics and the behavioral decisions driving host contact. Referred to as “economic epidemiology” or “epidemiological economics,” the approach explores the determinants of decisions about the number and type of contacts made by individuals, using insights and methods from economics. We show how the approach has the potential both to improve predictions of the course of infectious disease, and to support development of novel approaches to infectious disease management.