In 2003, the Ontario Racing Commission (ORC) made reporting of all racehorse fatalities in the province mandatory, becoming the first jurisdiction in North America to do so. Simply called the Death Registry, the resulting database provided post-mortem information on racehorses that died within 60 days of racing, or being entered or qualified to race, from 2003 onward.

In 2016, the ORC was folded into the Alcohol and Gaming Commission of Ontario (ACGO). Now racing’s provincial regulatory body, the AGCO continues to maintain a database of equine fatalities (all racehorse breeds) under its Equine Incidences in Ontario Racing (EIOR) program, part of its Equine Health Program. Dr. Peter Physick-Sheard, Professor Emeritus of University of Guelph’s Department of Population Medicine was allowed access to the Death Registry, from 2003-2015, to conduct an independent academic review.

“The reason behind having someone go through our data and present it was to understand the reasons and the risk factors so we could take that information and try to reduce the numbers or the [fatality] rates equally going forward,” says Dr. Adam Chambers, AGCO’s senior manager of veterinary services.

In October 2021, the journal Animals published “Factors Associated with Fatality in Ontario Thoroughbred Racehorses: 2003-2015” – the third report Physick-Sheard has written based on the dataset.

Ontario Racehorse Death Registry: Descriptive analysis and rates of mortality” was published in the Equine Veterinary Journal in 2018. That study revealed, of all the racing breeds in the province – Thoroughbreds, Standardbreds and Quarter Horses – that Thoroughbreds had the greatest exercise-associated fatality risk and the highest overall fatality. Then, in early 2021, “Factors Associated with Mortality in Ontario Standardbred Racing” was published in Animals.

Musculoskeletal Injury Main Complaint

In his most recent study based on the data, the records and details of fatalities in 236,386 race starts (433 fatalities) and 459,013 workouts (252 fatalities) were combined in multi-variable logistic regression analysis – or modelling – explains Physick-Sheard to Canadian Thoroughbred.

“The procedure tries to find out how much of the variation in who died or didn’t – i.e., how much of the information can be explained by each of the other factors you put in, such as sex and age. You’re basically shaking them up in a big bag and seeing which ones fall out in which order.”

Dr. Peter Physick-Sheard.

The “bag shaking” resulted in a fatal incident database of 695 fatalities. The horses ranged in age from two to 10 years, with 237 geldings, 166 stallions and 292 females. Ten horses died in early training without participating in a work-event, leaving 685 for event-based multi-variable analysis. The overall fatality rate was 2.61% – considered high when compared with published reports, the study notes. (However, case and general fatality information data can differ between reports, so comparison should be done with caution.)

The presenting problems leading to fatality were narrowed down to nine categories. Musculoskeletal injury was most associated with death at 66.62% (mainly fractures). There was also a high incidence of horses dying suddenly at 12.8%, followed by colic at 5.9%, then accidents, 5.61%, medical (i.e. laminitis, respiratory), neurological, iatrogenic (i.e. medication rection, septic arthritis), hemorrhage and unknown.

Liability was high for young horses early in the season, particularly intact males. Results suggested rapid accumulation of workload in animals in early preparation is damaging. Geldings seem to be able to benefit from accumulating work-events (both races and workouts) better than stallions or mares, even though they had more juvenile starts and therefore more risk.

Liability was high for young horses early in the season, particularly intact males…. intact males may be managed differently in terms of training distance and time and the impacts are masked by “sex-linked differences in an animal’s approach to work.”

Physick-Sheard suggests behaviour and temperament may be a more significant factor than physical preparation for intact horses early in the season and says that evidence in his paper and others, “strongly suggests – in the pattern of mortalities in young horses at the track for the first time – much of the contribution of injury had to do with their immaturity and their unfamiliarity with the track.”

While part of the equation is a lack of skeletal maturity, he explains, the other aspect could be “colts never see a filly until they get to the racetrack. It’s a new experience for them and probably has a major impact on fatalities. The data from this and the Standardbred study points to that,” says Physick-Sheard. Further, in the study he notes, intact males might be managed differently in terms of training distance and time and the impacts are masked by “sex-linked differences in an animal’s approach to work” such as greater robustness and a more aggressive attitude.

Workouts and Field Size a Factor

Horses undertaking repeated workouts (annual workouts per horse ranged from one to 29) had higher liability. As Physick-Sheard says, workouts are used for many different reasons, not the least of which is demonstrating to race stewards a horse is ready to compete. “But when a horse gets older, and it’s an average horse, you ask why would it be going to workouts so frequently? Well, one interpretation is, there’s a problem,” says the researcher. “Overall, there seems to be an association between undertaking frequent workouts and fatalities. As though that group was, in fact, a less-than-healthy group of horses.”

While race distance was not a significant liability factor – except for young horses in short-distance sprints – an increase in field size was. According to the study’s findings, horses in the middle of the pack may be relatively safe, with a larger field providing “more opportunity to settle into the pace.” However, early-finishing horses were at risk. This “may reveal an impact of the effort involved in getting to the front and staying there.” The greatest fatality was associated with races between nine to 12 furlongs and with an intermediate field size (eight to 11 horses).

While race distance was not a significant liability factor – except for young horses in short-distance sprints – an increase in field size was.

Jockey strategy used during a race “might influence probability of fatality, and in either direction,” he says in his research paper. A jockey’s role is to get a horse to the front to win. This requires making decisions from the moment the gates open as to where they’re going to position the horse in the field and how hard they’re going to push the animal – and when to ease off of a horse that is spent. Physick-Sheard has completed a more in-depth analysis beyond the scope of this study on the subject and alluded to further details about the ins-and-outs of the jockey’s job, especially when it comes to races with larger fields and the “protected” middle horses. He says to CT, “Let’s say, jockey strategy within a race is probably a more important factor than previously considered.”

Track surface was not as significant a factor as suspected. “Aside from the issue in turf racing, particularly in workouts,” says Physick-Sheard, referring to horses being at higher risk when switching from dirt/synthetic to turf. “We had to comment on [surface] because it’s such a big issue in the industry.” He did want to emphasize, however, that the major concern with track surface “tends to be injury,” not fatalities.

Fatality fell toward the end of a season and also for horses with long, successful careers, suggests the study. However, horses that do not show “this robustness and staying power” are of the greatest concern in reducing fatality. He concludes, if the associations identified in his study represent sources of stress – current or cumulative – identifying at-risk horses on this basis “could be as productive as targeting specific suspected causes that contribute to individual deaths.”

A Fatality is a Failure

And, as Physick-Sheard notes in the study itself, the probability of fatality fell over the study period probably due in part to increasing awareness of the AGCO’s Equine Health Program and the adoption of new procedures and regulations.

“It’s an ongoing, dynamic process,” he tells CT. “Their elaborations, their changes are not just based upon the report. They will be able to use the results of the study to focus on particular areas and introduce measures that sharpen their ability to identify problems.”

The commission’s Equine Incidences in Ontario Racing program records much more than its ORC death-registry predecessor. As well as fatalities and related post-mortems, it also details minor and serious accidents, performance and drug testing data and horse welfare information all with the goal of “reducing the numbers and the rate of fatalities,” says Dr. Adam Chambers.

Some of the AGCO’s efforts since it assumed regulatory control of racing six years ago include: 24-hour ban on all race day medications; extreme weather standards for racetracks and no penalties for weather scratches to reduce horse health and transportation risks; annual racetrack facility and track surface inspections.

The addition of computerized tomography scans in post-mortem examinations in 2015 “has been really helpful in changing the thinking of how we approach some of the catastrophic musculoskeletal injuries in horses.”

All racehorses are observed by official veterinarians before, during and after competition. Horses displaying symptoms of illness, lameness or injury cannot race for a minimum of seven days and may require a veterinary exam before racing again. Racetracks must notify AGCO if any horse experiences distress during morning exercise with a non-race day follow-up examination by an official veterinarian.

As well, there has been an increase in the number of safety reports to identify horses at an increased risk of injury and mortality. For example, official veterinarians now conduct examinations after training “when signs [of a problem] may be more accentuated, rather than on race days before the horse has exercised,” says Chambers. “Identification of mild injuries in horses before they suffer a catastrophic injury is the ultimate goal.”

Chambers also says the addition of computed tomography (CT) scans in post-mortem examinations in 2015 “has been really helpful in changing the thinking of how we approach some of the catastrophic musculoskeletal injuries in horses.” He says it’s removed the idea that these horses have inherently weak bones, instead, in many cases, they have pre-existing lesions, sometimes stress fractures.

“The AGCO – and before that the ORC – has encouraged veterinarians to broaden their understanding of the cause of injuries. With respect, we wish to challenge a traditional narrative that the cause of many of these fatalities was a ‘bad step,’” asserts Chambers.

“AGCO Veterinary Services provides training to all official Veterinarians and continues to meet with them monthly to continuously improve the regulatory oversight of horses on race day,” he adds, citing several experts who have come to Ontario over the years to speak to veterinarians on topics such as the identification and pathogenesis of stress fractures and other injuries, the use of an magnetic resonance imaging in a pre-race context, epidemiology of thoroughbred racehorse risk factors for death and advancements for joint therapies.

In Physick-Sheard’s study, after musculoskeletal injuries, the second-highest number of fatalities was horses that died suddenly. The AGCO, recognizing this as a major concern, in 2017, implemented a Sudden Death Protocol for horses that die within an hour of exercise.

“That has been an important aspect of trying to elucidate the causes of sudden death associated with exercise in racehorses,” says Chambers, explaining there are racehorses with exercise-induced pulmonary hemorrhage or problems with their aorta. “And then there’s a proportion of horses that don’t show too much in terms of significant finding on post-mortem and the Sudden Death Protocol really tries to look at the microscopic level at what’s going on at the heart, in part, but also other areas.”

Chambers also notes the AGCO is proactively starting to return findings of post-fatality inspections to all the involved parties – trainer, jockey, veterinarians, etc.

“Because basically a fatality is a failure,” he says. “And we’re trying to learn from that. So, we can have more of a frank discussion about what came to pass. I think we’re closing the loop by providing the people who are intimately involved with decisions of the horse’s final days, final months of care to give them as much feedback as we can.”

He concluded with, “All horse racing participants – the regulator, owners, trainers, and racetracks – should play an active part in the industry’s efforts to reduce the number of these incidents. Trainers can play a particularly important part, as they are the ones with intimate knowledge of the horses under their care.”