Aaron Reeves

Research Associate
Animal Population Health Institute
College of Veterinary Medicine & Biomedical Sciences
Colorado State University
Fort Collins, Colorado 80523
Telephone: 970-494-7591
Email: Aaron -dot- Reeves -at- colostate -dot- edu

Background & training

I am a research associate and Ph.D. student in the Animal Population Health Institute (APHI) at Colorado State University. While I am a biologist by training, since 1996 I have spent a considerable portion of my time designing and developing computer programs to address problems in the biological or biomedical sciences. I have written and published a variety of applications used by cytogeneticists, phylogeneticists, developmental biologists, and (occasionally) other programmers.

I have a Bachelor of Science degree in Biology from the College of William and Mary (1994), and a Master of Science degree in Botany from Colorado State University (1999).

Model building at APHI

I moved to the Animal Population Health Institute in June of 2004. Upon joining APHI, I was immediately impressed by the synthesis of biological knowledge with the application of quantitative and analytical approaches for the purpose of seeking answers to practical questions. It is this fusion of qualities that attracts me to epidemiological research, particularly in light of my background and general interests.

Model building "forces those concerned with building the simulation model to examine the system objectively... The enlightenment that the process provides is often surprising."¹

Most of my time is now spent on the North American Animal Disease Spread Model (NAADSM). NAADSM is a computer program that uses Monte Carlo methods to simulate the spread and control of animal diseases in a population of livestock herds.

Disease modeling in veterinary epidemiology is an area of research that continues to garner considerable attention from researchers and governmental institutions. With continuing improvements in computer technology, we are able to reasonably model ever more realistic and detailed scenarios. The use of models like NAADSM to study epidemiological phenomena has several key advantages:

Models allow us to investigate disease dynamics in the absence of an actual outbreak.
We can used models to assess disease behavior under a variety of conditions.
Models may be used to compare the efficacy of different disease control strategies.
Model construction is quite inexpensive relative to the costs associated with an actual epidemic.

Doctoral research

"During the modeling process, areas where our knowledge... is fundamentally lacking are often identified, so that field research can be directed more effectively."²

Most animal disease modeling efforts, including the NAADSM project, emphasize the study of foreign diseases and their effects on a completely naive population. A major objective is determine the optimal control measures for eradication of the disease and restoration of disease-free status in the country or region of interest. Foot-and-mouth disease (FMD), for example, is one of the prototypical animal diseases simulated by computer models because of its highly contagious nature and large potential economic impact. There has not been an outbreak of FMD in the US since the 1920's.

In many parts of the world, however, FMD is endemic: outbreaks may appear regularly or sporadically in a population that has some natural immunity to the disease. Control measures designed to emphasize disease eradication may not be appropriate in such a situation, particularly where resources available for control are limited. It might be preferable to examine strategies to use existing resources to minimize the impact of disease: such strategies may, or may not, be the same as those that might be employed for total eradication.

My doctoral research in the Department of Clinical Sciences at CSU (which is currently in its early stages) involves the creation and application of computer models similar to NAADSM, but which are suitable for the modeling of endemic diseases, beginning with FMD. Among the additional factors that must be considered are the differences in disease spread where natural prevalence is greater than zero, and where natural immunity to disease is present as a result of previous exposure.

References

¹ Dent and Blackie, 1979. Systems simulation in agriculture. Applied Science Publishers, London.

² Taylor, N. 2003. Review of the use of models in informing disease control policy development and adjustment. Available online at http://www.defra.gov.uk/science/documents/publications/2003/UseOfModelsInDiseaseControlPolicy.pdf.