Database unlocks disease trends

The federal government wants trend data on outbreaks of zoonotic disease and is looking to corporate veterinary practice to provide it.

While such information might eventually come from a variety of sources including national associations and teaching hospitals, Banfield, The Pet Hospital, maintains the nation's largest database tracking companion animal health. More than 12 million files updated by a weekly 75,000 new cases from 500 practice franchises landed Purdue University researchers a $1.2-million grant to outfit Banfield's database to monitor bioterrorism in 2003. Almost three years later, the government seeks alerts on emerging diseases as well as reports on drug safety.

The resulting project is titled the Purdue-Banfield National Companion Animal Surveillance Program. While the United States Department of Agriculture wants researchers to tally adverse vaccine reactions, the Centers for Disease Control and Prevention (CDC) backs exploring the database to identify disease patterns that include everything from West Nile virus to antimicrobial resistance within Salmonella cases.

The work is vital considering up to 70 percent of U.S. households include a dog, says Dr. Nina Marano, associate director for veterinary medicine and public health within CDC's National Center for Infectious Diseases.

"Much of this is early and developmental, but it certainly shows we need to partner in companion animal health to have some predictors to public health," Marano says. "The potential use is to look at something like tick-borne problems such as Ehrlichia or Rocky Mountain spotted fever. We have a lot of interest in exploring this database."

Banfield's data isn't the only information CDC covets. Teaching hospitals stockpile huge numbers of records, and with 3,000 member practices nationwide, the American Animal Hospital Association (AAHA) appears ripe for a centralized database. Although Marano confirms development talks are underway, AAHA Executive Director Dr. John Albers flatly denies direct discussions currently are taking place.

Data extraction

That won't slow Purdue University professor Dr. Larry Glickman, who keeps busy mining Banfield's catalog of information. With a small team of researchers, the epidemiologist uses what he describes as a Web portal to map out case facts that date back to 1993. At least three research papers employing the database to study protozoal parasites, leptospirosis and nematode parasites are scheduled for journal publications in the coming months. A fourth paper describing the database system is planned for release in an upcoming issue of the journal Vector-Borne and Zoonotic Diseases.

"We've been learning how to get at this data and explore it for about the last year-and-a-half, and it's mind-boggling," Glickman says. "There are so many projects and questions. For 25 years as an epidemiologist, I've been wishing for a database like this, and it's appeared."

Testing the information

What's failed to develop are numbers showing a decline in animal health following Hurricane Katrina's wrath along the Gulf Coast. Days after the storm, Glickman honed in on eight practices in Federal Emergency Management Agency-designated disaster zones. The impending hurricane impacted the number of patients seen at those practices up to six days prior to Katrina's Aug. 29 landfall, and population remained low for three weeks following the storm. Most surprising, Glickman says, is the data showed no upswing in incidence of respiratory, gastrointestinal and skin infections that were commonly reported in humans.

"The only thing I can suggest is that these pets were not in shelters, they were evacuated," he contends. Glickman plans to continue to monitor area pets for long-term health effects.

Future vision

With that kind of capability, it seems natural to market the system to state and federal health departments as well as government veterinarians, Glickman says. Current plans are underway to upgrade the database to allow for real-time access to federal agencies. For company insiders who are monitoring canine influenza, the approach already exists.

"I get a report every day of exactly where the hot spots are and how many pets are involved," says Dr. Hugh Lewis, a Banfield clinical pathologist charged with launching the database system. "We want to collect samples from individual dogs before they start treatment for influenza, and the only way to do that is in real-time. Our interest is not only learning from the accumulative database but also better understanding the disease of pets that do occur."