Predicting Future Epidemics

By Pat Bailey

Veterinary epidemiologists Tim Carpenter and Mark Thurmond talk casually of California’s future outbreak of foot-and-mouth disease.

It’s not a matter of if, but when, they say. From one herd—likely on a dairy somewhere in the Central Valley—the viral disease will probably spread within just 21 days to at least 18 other herds. By the time the epidemic is brought under control many months later, it will have affected hundreds of herds and cost California several billion dollars in killed animals, closed farms, and lost revenue and jobs.

The story is all in the data now being meticulously gathered by the Center for Animal Disease Modeling and Surveillance in the UC Davis School of Veterinary Medicine. When processed through a computer-based model developed with information about the biology of the disease and the complexities of California’s animal agriculture, the numbers paint a picture of how an outbreak of foot-and-mouth disease will likely unfold here.

In 2001, the disease infected a mere 1 percent of the 120,000 farms in the United Kingdom but resulted in 7 million sheep and cattle being killed to stop the spread. By the time the epidemic ended, it had cost Britain roughly $15 billion. Now it is just a matter of time until international trade and travel bring the disease to our own shores, Carpenter said.

Foot-and-mouth disease is strictly an animal ailment, but because of the enormous economic implications and the potential for the disease to be introduced by terrorists, the federal Department of Homeland Security and the Armed Forces Medical Intelligence Center fund most of the work at the UC Davis animal disease modeling center.

Several years ago, the UC Davis center began collecting information for models involving domestic and global spread of foot-and-mouth disease. This included data from a detailed survey sent to thousands of livestock producers throughout the state, ranging from the largest dairy operation down to hobby farmers and 4-H youths, plus data for the past 50 years from all over the world.

The survey asked: How many animals are on your farm? Who comes on your farm? Where do trucks transporting milk or animals go when they leave your farm? Surveys have also been sent to veterinarians, truckers, hoof trimmers—anyone whose interaction with the farm might transmit the disease.

The data are being poured into a test model developed in 1998 for Kings, Tulare and Fresno counties. So far, the disease simulation has told the researchers what could be expected in terms of the length, magnitude and cost of a potential foot-and-mouth epidemic. And it has raised as many questions as it has answered—questions that can be directed back at the model.

In an outbreak, what would happen if the state decided to vaccinate livestock against foot-and-mouth to halt its spread? What if a surveillance program were instituted for earlier detection? How would a one-week jump on detecting the disease affect its impact?

The Davis center is starting to apply the model nationally, collecting data first in Iowa. Preliminary analyses of 50,000 relationships in the livestock industry suggest that a foot-and-mouth outbreak there would last about three months and affect a couple hundred farms. The federal government would be unable to provide enough veterinarians to euthanize the massive number of culled animals, Carpenter noted.

As the model is expanded to encompass the entire United States, it will be continually refined. Eventually, it will spin a numerical web reflecting data on 1 million farms, ranches and backyard sites.

Thurmond’s group has also developed several models for a global foot-and-mouth surveillance system, which will predict where and under what conditions foot-and-mouth disease is likely to occur and to project the risk of the disease entering the United States.

Other work by his group includes development of remote detection systems for the virus and a real-time Web portal system for the global transfer, mapping and display of foot-and-mouth information. In order to understand forces that can affect evolution of the foot-and-mouth virus, the group is also modeling the molecular changes in the genetically fickle virus, which is capable of generating many mutant strains that can evade the host animal’s immune system and complicate detection and vaccination.

In other work, the UC Davis center is developing models for avian influenza, West Nile virus and toxoplasmosis, a parasitic disease that is killing California sea otters.

Such intensive modeling efforts are aimed at identifying features of disease transmission and virus evolution that can be used to prevent and mitigate disease, not only for the U.S. but also for other countries that currently have the disease.

“Through modeling, we can take biology as we understand it in subjective terms . . . and convert it to something that is objective and has defined quantitative elements that can provide a predictable structure for the biology,” Thurmond said. “If we can’t predict an outcome, the obvious question is why? Thus, modeling helps us understand in quantifiable terms what we do not yet know about the disease.”

Related stories:

Introduction

Uniting Against a Possible Pandemic

The Top 10: Epidemic Hall of Infamy

A Big Look at Small Invaders

Ducking Bird Flu and other Contagious Diseases

Pat Bailey writes about the agricultural and veterinary sciences for UC Davis.