‘Is This Sport Safe Enough?’ Breaking Down the Latest Eventing Safety Statistics

In addition to being a cardiologist, Rob Stevenson has worn many hats in the sport of eventing: rider, coach, parent, event organizer, team selector, and, for the past five years, National Safety Officer for Canadian Eventing. In response to recent posts addressing issues of risk management in eventing, he contributes this timely article highlighting the latest available statistics on eventing safety in the context of the discussion. There is a mighty international effort of ongoing study in this area but, as Rob notes, "there needs to be an enhanced effort to communicate the findings of the various research efforts from around the world to the very heart of our sport." We thank Rob for his help in disseminating this valuable information; it is a must-read for anyone seeking to better understand and contribute to the conversations taking place.

Rob Stevenson and Risky Business II representing Canada in the 1992 Barcelona Olympic Games. Photo by Elizabeth Furth.

Rob Stevenson and Risky Business II representing Canada in the 1992 Barcelona Olympic Games. Photo by Elizabeth Furth.

Following the recent EN post by Dr. Joanna Newton, “A Way Forward: How Eventing Can Use Established Strategies to Improve Safety,” I thought I might provide an update from the perspective of the National Safety Officer for Canadian Eventing.

Declaration of conflicts: none. I am not an FEI official. I receive no funding from any sources with a vested interest in eventing safety. I am in my fifth year as a safety officer, whereby I serve as the Canadian delegate to attend the annual FEI Risk Management Seminar. In my role nationally, I attend to the dissemination of the annual safety statistics for the sport and serve as a contact person in times of need. I did work for two years as part of the USEA cardiopulmonary study, focused mainly on the collection of data which might predict sudden cardiovascular collapse in event horses.

Like many that will be reading this post, the sport of eventing has been a major part of my life. I have ridden at the upper levels, I have coached for more than 20 years, I organize events, I have served in the role as a team selector. And more recently, I have officially become the father of an event rider.

Those who have heard me speak know the profound effect that loss of life in this sport has on me personally. Though in my work as a cardiologist I deal with life and death every day, thankfully I can say that every life remains precious to me. Thus it is completely understandable to me just how hard it is for our greater community to live on with the three deaths that we have seen in international events in 2016.

The overarching dilemma is this: without risk, there is no sport. And secondly, no sport can ever be “safe.” We endeavor to maintain a sport that is “safe enough,” an obsession with an acceptable level of risk. These are key points in this discussion. I have struggled personally to understand just how to know what could be “safe enough”? I have struggled with the notion that a mistake in this sport should not result in significant injury, or loss of life — whether equine or human.

Yet I have realized this view of mine to be somewhat naive, as mistakes while driving, skiing, flying, cycling, swimming, climbing and many other activities can have fatal results. In my view this does not lessen the magnitude or the implications of athlete and horse injuries in our sport, it simply reminds us that in choosing to leave the safety of our homes, we take risks. We choose to do so, as the alternative would be a very limited life indeed.

Through work with the FEI, I have learned of the tremendous international effort made towards the safety to the horse, the rider and the sport. This is work done by the FEI and by the many national eventing organizations that report annually on safety statistics to the FEI.

We ultimately have a responsibility to protect our horses: they do not choose to participate in this sport. We have a responsibility to riders of all levels to ensure a reasonable test for the level of competition. 

Lastly, we have a responsibility to the sport itself, to ensure that the sport continues to grow and evolve and ultimately to protect it as an Olympic sport. In the language of the FEI, with the horse and rider we always want to see a “good picture.” This impression should not just be a grimacing acceptance by eventing insiders, but rather a positive impression to all those who might be watching the sport live, on television or on the internet.

Safety has been an obsession of this sport for many years. Deaths in the sport as a result of cross country falls led to the first FEI symposium on safety in 2000. The work done at that symposium essentially became the foundation for work that has been done by the FEI, and various national sporting organizations since that time. Cross country analysis with the collection of fall statistics has been compiled for more than 10 years. This culmination of data has provided an objective measure by which we can consider safety in the sport.

I consider the fall data as a point measure on the continuum of performance and risk. We want good riding. That is not to say that good riders don’t sometimes make a mistake, just that the data would suggest that they make fewer mistakes (see note on rider categorization later in this document). With good riding we hope to avoid bad riding, thus avoid a greater chance of near misses. With fewer near misses, we’ll have fewer falls, specifically falls of horse and rider.

With fewer falls of horse and rider, we’ll anticipate fewer serious injuries. I will highlight several key points from three very important eventing safety documents that have been released in the last six months. As I have said, I am not an FEI official. I attended the presentation of these reports and later read the reports with a very specific question in mind: “Is this sport safe enough?” And then considered in my role as safety officer: “What can we do to make this sport safer?”

If I have any complaint about these documents, I would say that they have been insufficiently disseminated. They should have been mandatory reading for everyone in the sport, not just for those with the FEI Eventing Risk Management site as their homepage! I hope this overview will pique interest and promote dialogue. Without knowledge of the work currently ongoing in the sport — and this is just a sample — one can feel helplessly adrift in this conversation on safety.


The first of these to consider is the 2016 FEI Eventing Risk Management Programme Statistics. This document captures 10 years of data from FEI competitions. There has been tremendous growth in the sport internationally with a near doubling of FEI events and starters within this period.

When considering all falls, there has been a downward trend in the past 10 years. In the most recent year of data available (2015), 5.60% of all starters fell. As one might expect, as the level of difficulty increases, there is an increasing percentage of fallers at the 1*, 2*, 3* and 4* levels cited as 4.85%, 5.75%, 6.96% and 10.47% respectively. These results are in keeping with targets established for the sport based on historical data.

Of the 5.60% of fallers in 2015, most of these were unseated rider falls (4.20%), with a lesser number of horse falls (1.41%). Of the total horse falls (1.41%), most were non-rotational (1.22%) with a lesser number of rotational falls (0.19%). This represents an approximate 58% relative risk reduction in the likelihood of a rotational fall since 2005.

Clearly, not all falls result in injury. Continuing in this document, we see that 93.7% of all falls have no injury. Of those that suffer injury approximately 3.5% will have slight injury and 2.8% will have serious injury. Serious injury is generally defined by injury requiring hospitalization.

The total number of serious rider injuries in FEI competitions in 2015 was 32: 13 from unseated riders, 12 from non-rotational horse falls and six from rotational horse falls. Yet the emphasis appropriately remains on reducing/preventing rotational horse falls, as the likelihood of serious injury from a rotational horse fall is 10-fold greater than if simply unseated (1.52% versus 15.79%).

The final statistics in the document describe a serious injury in one of every 520 starters and a fatal injury in one of 17,317 starters. As we must consider both the numerators and the denominators, it is worth noting that there were 20,351 starters at FEI events in 2015. Thus the risk of a fatal injury is very low; however, the risk is not zero.


The second document for consideration is based on the scientific analysis of that same data set. For the decade that the FEI Eventing fall data has been collected, one could follow trends and outcomes, but the data set could not predict association or cause for bad outcomes.

However, given that there had been a great deal of data collected through event fall reports, there was great potential for scientific analysis to potentially determine causation. This is the work of Professor Dominique Maes at the University of Brussels. Dr. Maes presented her preliminary analysis of the 2015 data with interesting findings.

She mined a data-set collected on over 600 data points on each of over 20,000 starters. For the purpose of this report, she then highlighted trends related to falls. She confirmed that falls increase as the level of competition increases. There was a much greater risk of falls — horse falls in particular — at championships at all star levels.

She examined falls at types of fences. Fence profiles described as “rounds,” “brushes” and “square spreads” accounted for the greatest number of falls. However, the normalization to account for the frequency of each jump type has yet to be applied.

As might be expected, her data analysis did predict that serious injury to horse or rider was associated with a rotational fall.

Fences associated with water did have a greater number of falls. Younger horses were found to be more likely associated with falls. Interestingly, excellence in dressage was found to correlate with clean cross country outcomes just as bad dressage was found to correlate with elimination on the cross country. Given the unique feature of our sport where women and men compete equally, she has also considered gender comparisons (final results pending!).

In summary with Professor Maes’ work, we see the first real analysis of the data set amassed by the FEI over the past decade. This type of analysis can lead to further consideration of safety. Rather than simply asking “Is the sport safe enough?,” this analysis may well predict the means by which we can make the sport safer.


In 2015, the FEI Eventing discipline chose to proceed with an external evaluation/audit of the sport. It was felt that it was time for an “outside” view on the sport in an effort to improve the sport as a whole, and specifically to consider the issues of safety and risk management. This is the third document that I will consider, “Analysis of Risk Factors for Horse Falls in the Cross Country Test of FEI Eventing.”

This audit was performed by Charles Barnett with the assistance of Drs. Murray, Huws and Singer from the Universities of Bristol and Liverpool in the UK. The study focused on data in the five-year period of 2010-2014.

Some of the results were expected: Horse falls increase with the star level and decrease with greater rider categorization. In consideration of fence types, square spreads and corners increase the risk of horse falls. Interestingly, frangible fences were also found to be associated with an increase in the risk of horse falls.* Fences associated with water or downhill approach also increased the risk of a horse fall. Certain venues and course designers were also associated with an increased incidence of horse falls.

Their key recommendations were grouped into factors of competition, horse/rider, fence, and data collection. The emphasis was upon the qualifications, rider categorization, specific qualifications for younger horses, rider education, monitoring the faller partnerships, raising awareness of high risk type fences, further analysis of frangible fences and more descriptive data collection. Thus an external review of the sport highlighted a number of areas of increased risk that were already known, and shared further insight into where we might refocus our attention.

(*This finding reflects a number of considerations, most notably the type of obstacle where frangible technology is currently employed. This finding highlights two very important points: statistical analysis does not always support conventional wisdom, and secondly that more study of this emerging technology is needed.)

These three studies could be considered more of an epidemiologic assessment of the health and safety of the sport of eventing. These studies considered outcomes and safety data from the last decade of the sport. In summary, we can conclude that the sport continues to operate in what has been determined to be an acceptable level of safety.

Secondly, the sport has seen a significant decline in the occurrence of rotational falls. We can glean that there is increased risk of falls at the higher star levels. We recognize that the higher the rider’s categorization, the lower their risk of a fall. Younger horses are associated with an increased risk of falls. We can identify certain jump types that increase the risk of falls. There is a signal that better dressage scores are linked to better cross country outcomes. And finally that a very low risk of a serious injury or death does not equate to zero risk.

In speaking with Australian course designer Wayne Copping, I once heard him reflect that a tremendous responsibility falls back on the course designer. Having attended a number of course build-design courses by fellow-Canadian, Jay Hambly, I have often been reminded of the scrutiny and pressure associated with the designer’s everyday work.

I have often considered Jay’s work in comparison to the work done by my cardiac surgeon colleagues: everyone expects them to do a perfect job every time. Just as the surgeon is expected to somehow reverse the ravages of 60 years of hard-living, so too are these designers expected to account for all permutations of horse-rider combinations that will challenge the four or five cross country tracks that they have provided for any given weekend.

I recognize that there is a rigorous training system in place in most of the leading eventing nations for these designers and builders; however, I cannot reflect upon the safety considerations of this sport without acknowledging the incredible work and the pressure that these people must face in their everyday work.

With respect to the work of course designer/builders, we see the ongoing research in the field of deformable/frangible mechanisms for fences. There are a number of mechanisms which are approved for use in the sport. At this point, it would seem that we have standards to which these devices can be designed and deployed, and I believe the next step will be to confirm that they do reduce serious injury (see note above where frangible fences are associated with increased horse falls).

There are clearly many considerations in the use of this technology, citing cost, correct use, scoring and actual outcomes to name a few. Further investigation is needed and studies are ongoing.

The USEA cardiopulmonary research study continues with ongoing investigation of safety-related issues. This group chaired by Dr. Catherine Kohn is looking specifically at factors that could affect sudden cardiovascular collapse in event horses.

I think it is important to note that the work done by this group is truly an extension of the groundbreaking work done by Dr. Kohn and colleagues that studied the effects of heat and humidity stress on horses in the early 1990s before the Atlanta Olympics. The knowledge gained at that time changed the way in which event horses were prepared for and managed in the hot and humid environments often found at major games and championships. This veterinary legacy is seen in the way that event horses are managed, transported and cared for in the competitive environment today.

In the same way that the management and care of event horses has evolved in 20 years, so has the delivery of advanced medical care at events. There are likely few people in the world that have had the experience of Rusty Lowe in the provision of medical care at equestrian events. Advanced medical care has been standardized to reflect the unique needs of the sport. This is to ensure a very coordinated communication system and access to all areas of the cross country courses for an appropriate medical/paramedical team. These teams are trained to meet the unique demands of this sport, all the while hoping that their expertise will not be needed.

Within the USEA, safety and risk-management considerations touch all aspects of the sport: governance, rules, officials, coaching, high performance, legal and fund-raising. The USEA task force on cross country safety serves as a clearing-house for all aspects of safety consideration. Fortunately, many of the dominant eventing nations have similar working groups to champion aspects of eventing safety.

As I refer back to the recent posting by Dr. Newton, I have tried to provide suitable background to address the very valuable points that she had raised. She proposed a framework considering the medical-research strategies that might be applicable to eventing: namely clinical trials, quality improvement studies, root cause analysis and data collection.

As I read through her piece, I realized that much of what she was proposing was in fact already occurring in various jurisdictions. There are indeed clinical trials studying equine health, frangible jump technology, protective vests, and helmets. Quality improvement initiatives will continue. As the epidemiologic data analysis continues, it may be that a greater emphasis will be placed on qualifications and categorizations though we recognize that any such consideration generally receives some push-back from competitors.

It may be that at some point certain types of fences will be phased out. However, at this time, I don’t believe there is sufficient data to stipulate the removal of any fence type; however, I am sure that the square spread type fences must receive the most careful scrutiny from the course builders.

The root-cause analysis is certainly worthy of consideration. Every horse that is lost in eventing competition does undergo a post-mortem evaluation to further determine the cause of death. In-depth investigations are conducted in each instance where there is loss of human life. Dr. Newton makes a very valuable suggestion that careful analysis should follow any rotational-type horse fall even if no significant injury to horse or rider ensues. I think this is a very valuable recommendation, and it may be that it is performed in other countries.

Lastly on the data collection and analysis, I do feel that we have more data and more analysis ongoing than ever before. I trust that this information will continue to shape the sport into the future.

In conclusion, there is no doubt that 2016 has been a disturbing year to date for the sport of eventing. The loss of life very appropriately prompts us to ask, ‘What more can be done?” or “What can I do?”

As a physician, a safety officer and as an eventing parent, I believe that I can conclude that there is a comprehensive structure in place that is managing risk in this sport. It is a complex undertaking that tries to manage athletes that are drawn to a risk sport. If there was no sport of eventing, what would event riders do? To this point, we have been well within the margins of acceptable risk as determined by the sport itself. That said, I have witnessed an ever increasing commitment to all aspects of safety in this sport in the past five years; suffice to say that I don’t think that anyone is satisfied even with the very low level of risk that exists in this sport today.

If someone were to ask me how their child could be safer in this sport, I would answer with a single word: coaching. We have to believe that for our developing riders that the emphasis must be on good riding.

I think two key areas to support nationally would be further investment in officials and specifically course designer/builders, as well as to support ongoing USEA studies in frangible jump technology and the USEA cardiopulmonary study.

Internationally, I feel that there needs to be an enhanced effort to communicate the findings of the various research efforts from around the world to the very heart of our sport. In order for people to become more involved in this issue, they need to know where we are on safety and risk management. In order to see the continued growth of this sport, it needs to be clear that eventing has a clear sense of its identity, not just knowledge of where the sport has been, but where it is headed as well.

I have included a number of links throughout this article. I trust that these will provide for further reading for those interested in this subject.