After Trials At Two More Racetracks, Here’s What We Know About What StrideSafe Is, And What It Isn’t by Natalie Voss|06.27.202306.27.2023|10:42pm10:43pm This spring, the racing industry faced renewed criticism of its safety record, after Churchill Downs saw an increased number of fatal injuries during its prominent spring meet. Insiders have been hopeful that a new technology called StrideSafe can one day reduce the number of injuries on America's racetracks. The system, which is made up of sensors carried in a horse's saddle pad that measure motion in all directions, was trialed in New York on most runners starting in 2021. The sensors are designed to detect and quantify changes in movement that could signal an impending injury. The system was engineered in Australia and has been used on the there and in Tasmania to alert trainers and veterinarians to stride changes that may be imperceptible to the human eye. In New York, the system was used in 6,626 starts. StrideSafe software ranks a performance similarly to a traffic light – red, amber, or green – with green showing no significant deviations from normal movement for that horse, red indicating significant change, and amber indicating that the horse should be monitored with caution. This year, horses at Rillito Park and Churchill Downs have been wearing the sensors, as researchers and regulators try to get an idea of how it can – and can't – be used to improve equine safety. We spoke with StrideSafe CEO Dr. David Lambert, and also tuned in to a presentation by Lambert and Churchill Downs equine medical director Dr. Will Farmer before Kentucky horsemen this week. Here's what we know so far about the system and its potential. —The system works better the more times the same horse wears it. StrideSafe is designed not just to compare a horse's way of going to what its software considers “normal” but, more importantly, to establish what an individual horse's “normal” way of moving looks like. It can give you a red, amber, or green rating for a horse when it's used on the horse for the first time, but each subsequent rating becomes more meaningful because the software can learn to look for changes for that individual. That means it's going to be most effective when it's used on the same circuit for a long time, or on multiple regional circuits over time. This hasn't yet happened outside of New York. The study run at Churchill's spring meeting will end when the Churchill meeting at Ellis concludes July 2. That study was funded by the Kentucky Equine Drug Research Council and included the collection of data from horses in races and breezes. The funding also extended to free diagnostic imaging for horses who received a red rating, in hopes of better defining the best way for horsemen and veterinarians to interpret those ratings. It's still uncertain whether funding will be found to continue readings at the Ellis meet. The StrideSafe system costs $35/horse per use. It was also used at Rillito Park's meeting earlier this year, with funding provided by owner Jim McIngvale. [Story Continues Below] —So far, the sensors are good at correctly identifying a horse most at risk of an impending musculoskeletal injury. At Churchill's spring meet, Lambert said seven of eight musculoskeletal breakdowns during racing had sensor outputs that were abnormal. (It's important to note that does not mean they had any physical changes or gait problems that would have been visible to the naked eye; all of them passed pre-race veterinary exams.) Those statistics – and the 19 of 20 fatal breakdowns flagged in New York – refer to the horse's rating in its last race, however. Data isn't uploaded anywhere in real-time, and in fact isn't collected and analyzed until the race card is over. For Lambert's purposes, it's useful to show that the sensors are not falsely rating very many performances as fine when they're not. Red and amber ratings also come up in plenty of performances where horses don't suffer fatal injury, and Lambert believes an alteration in a horse's movement may begin several performances ahead of a fatal breakdown. His hope is that red ratings can generate extra scrutiny early enough to avoid breakdowns. —Red and amber ratings can happen for reasons other than an injury. In the New York trial, Lambert found 17 percent of 6,616 starts had red ratings. He further split red ratings into multiple categories depending on severity of the horse's deviation from normal movement. Five percent of starts (or 331 starts) were in the highest-risk red category. These horses, he found, had a 300 times higher risk of fatality than horses with green runs. But the sensors in StrideSafe are measuring abnormal movement and do so without knowing context. That means that if a horse is checked or bumped during a race, or if they break from the gate awkwardly, the sensor will correctly realize that something is different and may give a red or amber rating. It's also possible a horse running on a new surface for the first time may move differently because they have difficulty with depth or traction, but haven't actually experienced damage that signals an oncoming fracture. StrideSafe researchers at Churchill watched race video of performances that generated red ratings prior to informing trainers of the red rating so they could try to give them as much context as possible. Lambert also encourages trainers to give feedback to StrideSafe about what they think may have caused an abnormal rating so he can better pinpoint what the sensors are picking up. That also means there are going to be red ratings which don't show any sign of impending injury on diagnostics, and red ratings that may switch to green in a horse's next start even without intervention. Lambert believes it's still worthwhile to screen those horses and hopefully prevent some fatal injuries. Support our journalismIf you appreciate our work, you can support us by subscribing to our Patreon stream. Learn more.Subscribe —There are also some types of injury or death that StrideSafe isn't likely to warn you about. In his presentation at Churchill this week, Lambert said that soft tissue injuries like ligament failure don't have the same pattern of forewarning that bone fractures do, based on human data, and he doesn't expect StrideSafe will see anything abnormal prior to the moment of injury. It also can't anticipate sudden deaths, which are commonly thought to be cardiac or pulmonary in nature and aren't preceded by much change in the horse's movement more than a few steps prior to a fall. Still, he's encouraged that the sensors can provide useful data. “Don't let's beat the whole idea to death just because [false greens] will happen,” he said. —For this reason, this is not going to become a regulatory tool. Lambert says there is not likely a world where a horse could be added to a veterinarian's list based on a red rating. In the Churchill trial, horse-specific results were not reported to the state racing commission. The future of StrideSafe is, to Lambert, as a tool in the toolbox of a trainer and their private veterinarian to let them know which horses may benefit from diagnostic scrutiny. —This system is also not going to tell a trainer what type of injury a horse has brewing. Over time, Lambert said, the system will be able to tell trainers whether it's noticing abnormalities in a front versus hind end, or left versus right side of the body, but that's it. “It's not a diagnostic tool,” said Lambert. “It's set up for screening.” Trainers of Churchill horses who received red ratings had the option of getting free diagnostic scans to correlate those ratings with injury risk. Three of those flagged horses received PET scans and one received a bone scan but those results aren't available to the public. Lambert said there are key points on certain bones that racetrack veterinarians already know are most likely to accumulate stress-related damage that can lead to a fracture, so this should give them a base to work from when they go hunting for any unseen problems. —StrideSafe was tested on horses recording timed workouts this spring. This would have marked the first time the system was used broadly on breezing horses. It does not take data from slower works or from the warm-ups before a horse enters the gate for a race because the horse is moving too slowly and if they're being ponied, that will impact their gait. Farmer said it has proven difficult to accumulate data on horses in the morning because training plans can often be subject to change and trainers don't have to notify anyone too far in advance that they're planning to work a given horse. He's hopeful that ultimately the sensors could warn of an impending injury based on workout data, before the horse is placed at risk in a higher-intensity race situation, but it's also hard to trace when horses who got red ratings in workouts may not have gone onto the next race as planned and been put on the sidelines for diagnostics instead. —Further details from Churchill aren't available to the public yet. Because the Churchill study was done with the hope it would result in a peer-reviewed publication, Lambert was unable to provide specific numbers on how many horses total were flagged as red, amber, and green, or the results of any follow-up examinations on red-flagged horses. Peer-reviewed journals often require that publishable work not have been shared prior to publication, which binds researchers to some silence as they go through the process of data analysis and journal review. —Practicing veterinarians will soon have a chance to learn about StrideSafe outputs. Lambert said he plans to hold a series of trainings for racetrack veterinarians to help them understand the statistics and charts the system generates, which could give them a better frame of reference for interpreting results and figuring out where to hunt for an injury.