In an age of emerging infectious diseases, ecologists and evolutionary biologists have significant contributions to offer to medical research and practice. Containing the spread of antibiotic-resistance bacteria in hospitals, preventing avian flu from triggering a human pandemic, or even slowing the progression of HIV within an individual patient - each of these problems is literally an exercise in applied evolutionary biology.

In order to anticipate and to contain disease spread and disease evolution, we need to understand the underlying population biology and population genetics of both pathogen and host. Conversely, through the wealth of available data and the rapidity of the pathogen evolution, infectious disease biology offers to population biologists an opportunity to observe evolution taking place in "real time," and as such provides a rich set of study systems for biologists who are interested in the basic ecological and evolutionary principles.

In a major focus of our infectious disease efforts, we use mathematical models to study how novel infectious diseases such as H5N1 Avian Influenza (bird flu) or SARS emerge into human populations. We have already used the theory of branching processes to model the interplay of ecological factors and evolutionary change in disease emergence; the critical next step is to use models like these to help us detect and contain future epidemics. With Marc Lipsitch at the Harvard School of Public Health and Jacco Wallinga at the National Institute for Public Health and the Environment in the Netherlands, I am part of an NIH MIDAS team focused on avian influenza. Together we are developing new statistical methods for detecting outbreaks of novel diseases, for monitoring their spread in real time, and for rapidly selecting appropriate interventions in response.

I am also interested in the grave and growing problem posed by antibiotic resistant bacteria. Over the past few years, we have developed a set of mathematical models to explain and predict the evolution and spread of antibiotic resistance in hospitals and nursing homes. Most recently, we have used models derived from mathematical ecology to explore whether antimicrobial cycling, a much-touted ``crop-rotation'' strategy for controlling antibiotic resistance, would be effective. The models predicted that this strategy would not work, explained why, and suggested a better alternative: ensuring heterogeneous drug use within a hospital whenever possible. Now that the results from the initial round of clinical trials are coming out as we expected, our model helps clinicians see what went wrong, and offers alternative suggestions for future trials.