06/27/2010 11:00PM

Study inconclusive on injuries, surfaces


LEXINGTON, Ky. - An analysis of injury data for racehorses collected over a one-year period has concluded that there is not yet a statistically significant difference between catastrophic injury rates for horses starting on dirt or artificial surfaces, according to an epidemiologist who is studying the data.

Dr. Tim Parkin, a research fellow at the University of Glasgow who studied 12 months of injury reports submitted by officials at 73 U.S. racetracks, said that the analysis could not make a case for a difference in the catastrophic-injury rate because of a smaller set of data for starts on artificial surfaces compared with starts on dirt. According to the raw numbers, 2.14 horses per 1,000 horses who started on dirt during the study suffered a catastrophic breakdown, whereas 1.78 horses per 1,000 starts who started on artificial surface suffered a fatal injury, a difference of 0.36 horses per 1,000 starts.

Dr. Parkin used data collected from Nov. 1, 2008, to Oct. 31, 2009, for his analysis. The results of the analysis were presented during the Welfare and the Safety of the Racehorse Summit in the sales pavilion at Keeneland outside of Lexington on Monday.

Dr. Mary Scollay, the administrator of the database and the equine medical director for the Kentucky Horse Racing Commission, said that difference between the raw numbers and the inability of researchers to draw a statistically significant conclusion despite that difference would ensure that disagreements between supporters and critics of artificial surfaces would continue to linger until researchers are able to collect more data. Most artificial surfaces were installed at racetracks because of claims by supporters that they would reduce fatalities.

"It tells us that the debate goes on," Scollay said. "Opinions aren't going to be changed by what we presented. Let's be clear. It doesn't say that there is or is not a difference. We need more data."

The injury-reporting project was launched in 2008 as a means for the racing industry to collect data on the circumstances under which horses suffered fatal injuries. Administrators of the database have asked racing veterinarians to submit detailed reports about catastrophic injuries in order to build up a large store of information that could provide conclusions about whether certain factors, such as age, sex, or race distance, influence risk.

The initial analysis of the data did yield two statistically significant results, according to Parkin. The first was that female horses are 50 percent less likely to suffer a catastrophic injury than an intact male horse, and the second was that 2-year-old horses are 30 percent less likely to suffer a catastrophic breakdown than older horses.

According to the raw numbers, fillies suffered 1.84 fatal breakdowns per 1,000 starts and mares suffered catastrophic injuries at a 1.66 rate, compared with a rate of 3.16 for colts and 4.06 for horses.

Both Dr. Parkin and Dr. Scollay said that the data could not provide a basis for understanding why the catastrophic-injury rates differed so dramatically between male and female horses. However, Parkin also said that several studies had also purported to demonstrate a difference between the breakdown rates, but that those differences fell away when the studies were analyzed for multi-variable factors.

For example, Dr. Scollay said that the fatality rate for male horses may be higher because at-risk females are removed from the racing pool at a greater rate than colts because injured fillies are often retired to the breeding shed rather than rehabilitated and sent back to the track, as is often the case with male horses who have no future as stallions. Horses who have suffered injuries are at far greater risk for breakdown than horses who have not suffered injuries, according to other studies, so the male rate is being influenced by the larger number of at-risk horses within the population.

In other conclusions, the analysis did not find any statistically significant difference in the risk of fatality when looking at the distance of a race, the weight carried in a race, or the condition of the racing surface, either on turf or dirt, despite sometimes wide variations in the raw numbers.

Again, Parkin stressed that the data for one year in some categories - such as starts on yielding turf - remained too small to discern a difference within categories even if one actually exists. The administrators of the database plan to subject the data to epidemiological studies each year, in the hopes of further refining the numbers and pinning down the most significant risk factors.

"It's still potentially dangerous to be looking at this data after only one year," Parkin said.