Can pets be used as sentinels to provide early warning of terrorist attacks? Researchers at Purdue University think so and are in the process of developing such a system with Banfield Pet Hospitals, a chain of 314 hospitals across the United States. The system, Veterinary Medical Data-Surveillance of Syndromes or VMD-SOS, will gather health data from these hospitals, analyze them for adverse spontaneous health events, conduct on-site evaluations when necessary, and alert human health agencies when a suspicious health problem occurs.
"We are developing analytical techniques that, when used in a timely way, could signal a terrorist attack," says Dr. Larry Glickman, a professor of epidemiology and environmental medicine at the Purdue University School of Veterinary Medicine. "This approach is intended to complement, not replace, human medical record-based surveillance systems currently under development and give practicing veterinarians a key role in the war on terrorism."
The idea that pets can help identify problems that may threaten human health is certainly nothing new. In the early 1950s, residents of a village near Minamata Bay in Japan began noticing that cats, dogs, and other domestic animals were suffering from violent convulsions and mysteriously dying. A few years later, people began suffering symptoms including paralysis and impaired vision and speech. An investigation revealed that a neighboring chemical plant was polluting the bay water and fish with mercury, and consumption of the fish was causing the problems.
"There are many such examples of where animals sent signals but nobody paid any attention to them," Dr. Glickman said.
Throughout the past 30 years, animals have also been recognized as effective sentinels for many human diseases of biological origin such as rabies, plague, and West Nile infection. More recently, animals have also been used as "detectors" of biological or chemical weapons. In 1995, Japanese police carried canaries in cages while searching a suspected cult headquarters for the presence of sarin gas.
Several characteristics contribute to the ability of pets to serve as sentinels. They are less mobile than humans and constantly explore their environment by smelling and licking objects. They are smaller than humans and, therefore, may be more sensitive to a fixed dose of toxin or infectious agent. And, generally, pets have a shorter latency for most toxic events.
Purdue researchers chose to work with Banfield Pet Hospitals because it is a large chain; its 314 hospitals are located in 44 states, and about 60,000 dogs and cats are seen each week. That number is also likely to grow.
Most important, however, Purdue chose Banfield because medical information from all clinics is stored in a central computer. Trying to enlist a slew of independent clinics for the project would have been difficult, because many hospitals have different systems for storing data. "Here we have gone to 314 clinics who already have all the data standardized and computerized and are willing to share it," Dr. Glickman said. "That is the unique part of this."
The researchers will systematically mine health data collected from two million dogs and cats that visit these hospitals annually. On a daily basis, the hospitals will share their electronic records with Purdue epidemiologists who are developing computer-based algorithms that will rapidly detect problems caused by biological and chemical agents.
"(Our database) contains much clinical data, clinical signs, test results, etc., which makes it possible to screen the database for groups of clinical signs that reflect certain organ system dysfunctions," said Dr. Hugh Lewis, Banfield Pet Hospitals' senior vice president of practice development. "This could be more revealing than screening for various diagnoses because in bioterrorist threat situations, the (specific) diagnosis may not be apparent." At least it may not be apparent without time-consuming tests, Dr. Glickman said.
Researchers will create models of any clinical signs and abnormal test results, mapping them to specific geographic locations and time periods. When scientists detect an unusual cluster of adverse health events, they will notify veterinary hospitals, and veterinarians will collect additional information and samples from the pets involved. Purdue investigators might also conduct on-site visits.
If researchers confirm that an unusual health problem has occurred, they will contact the national Centers for Disease Control and Prevention and the Department of Homeland Security.
Although the VMD-SOS will initially focus on searching for health events that could indicate terrorist attacks, the system can also be used to track infectious animal and zoonotic diseases, and noninfectious diseases such as cancer. In fact, initially this was the focus of the project.
"When we started this whole project, our intent was to have the first nationwide system to be able to do surveillance of diseases affecting dogs and cats, look at how they are distributed in the United States, look if they are changing over time, see if we can spot epizootics that are occurring, and alert veterinarians around the country," Dr. Glickman said. "Then, following 9/11, we said 'hey, we have a system that we think can help in terms of homeland security. Why don't we switch our efforts for the meantime and focus this surveillance on acts of bioterrorism.'"
Dr. Lewis says VMD-SOS could help identify iatrogenic disease situations as well as endemic or zoonotic diseases. "Given such a heads-up warning of a disease hot spot allows one to investigate without delay and determine which explanation holds," Dr. Lewis said. "Having access to large numbers of pets with or without the syndrome also provides an opportunity to follow the level of infection, as well as monitor its spread."
The Humane Society Charitable Trust of Indianapolis provided the small grant that allowed the project to begin. Now, Purdue University and Banfield Pet Hospitals have sent a grant proposal to the National Institutes of Health asking for funding of roughly $1.25 million to continue the development of the VMD-SOS. They expect to hear this coming July.
To cover their bases, they have also sent proposals to the CDC and the Department of Homeland Security. Researchers hope one or a combination of the three agencies will come through with funds to help continue their work.
Only time will tell whether the project will be a success but, if it is, it will help satisfy a plea from a 1991 National Academy of Sciences study. That study called for animal sentinel systems to be coordinated on regional and national scales.