12/09/2009 12:00AM

Scots take to fore in injury research


TUCSON, Ariz. - Equine-health researchers at the University of Glasgow in Scotland have implemented a system to identify horses at risk of a tendon injury while racing and training at tracks in Hong Kong, the lead researcher said on Wednesday during a presentation at the University of Arizona Symposium on Racing and Gaming here.

Working with detailed records maintained by the Hong Kong Jockey Club, the researchers collected data on tendon injuries from more than 100,000 starts in order to identify the factors that put horses most at risk of sustaining a tendon injury that would result in a horse's retirement or worse, according to Dr. Tim Parkin, a University of Glasgow scientist who is leading the research. In 2007, the researchers used those risk factors to implement a monitoring system that is now being used by Hong Kong racing authorities to track the horses deemed likely to suffer injuries, in an effort to potentially intervene in the horse's training before the horse breaks down.

"The whole point is to identify at-risk horses before they retire," Parkin said.

Though the researchers have not yet been able to determine whether the monitoring program has been successful in reducing the amount of tendon injuries, the project has enormous implications for the horse-racing industry in the United States, which has just passed a one-year anniversary in the collection of injury data from more than 80 racetracks. The goal of the U.S. program, like the Glasgow research, is to identify risk factors for injuries in the United States so that the racing industry can craft policies to minimize catastrophic breakdowns.

According to market research, catastrophic breakdowns are one of the leading reasons why horse racing remains a fringe sport. In 2008, the National Thoroughbred Racing Association attempted to address some of the sport's public-perception problems surrounding racehorse injuries by launching an accreditation program for tracks called the Safety and Integrity Alliance, concurrent with the effort to start collecting injury data.

Dr. Mary Scollay, the equine medical director for the Kentucky Horse Racing Commission and the administrator of the injury-reporting database, said on Wednesday that the racing industry will need to gather far more data in order to begin to identify the types of factors that the Scottish researchers have identified. The process is slow because a glut of data is needed to find statistically significant factors. Once those factors are found, analysts will almost certainly need additional time to identify significant causes and mark those causes for more research, Scollay said,

Parkin said that researchers believe that they will need to collect data on the monitoring program for at least another year before they will know whether their efforts to identify at-risk horses reduced injuries.

"You'd probably need three years to five years to get the data that could say whether you had a statistically significant impact on those rates of injuries," Parkin said in an interview after his presentation.

Parkin also detailed separate research to identify risk factors for steeplechase horses performing in the United Kingdom, using a common statistical method called multi-level modeling. The method allows researchers to identify the most probable causes of injuries among a host of different factors, and, furthermore, it allows researchers to identify factors within the most probable causes as targets for additional data analysis.

For example, Parkin said that researchers found that risk factors were tied most strongly to the horse itself and to the type of race being run, while genetics was identified as a low-risk factor, and the jockey was identified as a no-risk factor. In addition, within the data identifying the horse itself as the biggest risk factor, researchers were able to determine, somewhat logically, that the horses most at-risk were those who had suffered previous tendon injuries.

U.S. researchers do not have near that level of detail yet, but Scollay said that 83 tracks are currently participating in the injury-reporting project, and that those tracks represent approximately 84 percent of all flat races run in the United States, including Quarter Horse races. The project passed a full year of collecting data on Oct. 31, 2009.

Although Scollay said that the main purpose of the database is to help the racing industry identify the leading causes of injuries, she also said that collecting the data would allow racing to counteract efforts by opponents of racing to paint the industry in a bad light. Those groups are increasingly collecting their own data, using far less rigorous standards for accuracy, and publicizing their results on the Internet, where the line between fact and fiction is often blurry.

"These folks are motivated and they are working hard," Scollay said.

Scollay also cautioned the racing industry about jumping to conclusions based on the results of the data, especially at the local level, where results can be skewed significantly by a spate of injuries, or, on the flip side, by a long stretch of time without any breakdowns. She urged racetracks to hire epidemiologists and other specialists in analyzing data before putting in place any policies or procedures to address any apparent problems.

"The analysis of the data can be daunting, it can be confusing, and there is a critical need to get it right," Dr. Scollay said.