A new HISA computer model to identify at-risk horses is being used by Churchill Downs officials to supplement the existing tracking safety network, say representatives of the Racing Integrity and Safety Authority. (HISA) and the Horse Racing Integrity and Welfare Unit (HIWU) announced at a press conference Tuesday the horse safety protocols in place for this year's GI Kentucky Derby and GI Kentucky Oaks.
“Basically, this is a model that looks at 44 potential risk factors that could put horses at risk of injury,” said HISA executive director Lisa Lazarus. Lisa Lazarus, CEO of HISA, added that the model assigns a “numerical” value to the horses analyzed.
Lazarus said HISA is currently beta testing the system with regulatory veterinarians at about 10 tracks around the country.
“But it is being offered to Churchill Downs Derby chiefs to be one of the tools that they can use with their sophisticated veterinary team and practices to help determine whether which horses at the Derby or on Derby Day are allowed to participate or not. there is a risk of injury,” she said.
Although the entire system is still in beta testing, “I hope by the end of this quarter it will be available to any managing veterinarian,” the HISA representative explained the basics in a follow-up conversation. according to.
“For now, it is capturing things that we already know are relevant,” the HISA representative added. “It's the 'what if' that's exciting, because we've never been able to think about using data in this way before.”
Experts have long identified specific factors that put racehorses at greater risk of injury, including fatal injuries. Some factors have long been understood including the racehorse's sex, the type and distance of the race, and the number of previous non-fatal injuries.
The 44 factors that went into the computer model included the veterinarians' listing history, whether the horse changed hands, any drops in class and the time between races, Lazarus said.
According to HISA representatives, the model also considers possible associations with the horse's veterinary treatment history.
“One of the things HISA can do that a state commission can't necessarily do at this scale is to look for linkages to treatment records,” the HISA representative said, explaining I wonder why even in fairly closed racing communities like California, such procedures are difficult to implement. done for logistical reasons.
“We can look at the association between this type of treatment in the past and poor outcomes,” the representative said.
HISA representatives said that, in terms of how the system works, a daily report is generated about each horse participating in the race. This report assigns a number from 0 to 44 depending on the number of individual risk factors flagged by the system.
“It has the name of the horse, it has a summary number and then it has a detailed breakdown of the 44 risk factors,” the HISA representative explained.
When asked whether a certain threshold number would trigger a particular set of tracking protocols, Lazarus said that regulators have discretion in how they model screening and safety measures. their entire existence.
“It's not a tool designed to say this horse has to be scratched. It's a tool designed to help veterinarians perform veterinary examinations as efficiently as possible, Lazarus said.
“These are horses that you may want to spend extra time with in your rounds,” the HISA representative explained further. “Or, if you're in a jurisdiction that does post-entry screening and then goes to see the horses, like in California, you'll probably add an extra horse to your list to see before race day. ”
Lazarus said the model is still being tweaked to ensure that horses at higher risk of injury aren't accidentally flagged by the system. “The reason why we haven't rolled it out everywhere is because we're still improving it and making sure it's optimal,” she said.
HISA representatives agree: “There is always a tension between false positives and false negatives.”
“There are horses with predictive value here [that] Come out and don't just run reliably, get out of the rankings and win,” the HISA representative added. “But the harm of being wrong in saying ‘this horse needs to be looked at more closely’ is much less harmful than missing out on a horse.”
When asked about the model's accuracy, HISA representatives said HISA had brought the percentage of “poor results” down to the low single digits. HISA representatives further explained the poor results, including horses being scratched on race day and other injury records. “It's not just deaths.”
HISA representatives said that currently, riders do not know the information generated by the model, and left open the possibility of such a situation occurring. “It's just a policy discussion that we haven't had yet.”
