A non-rotational horse and rider fall at the 2006 World Equestrian Games. © Evalyn Bemis
A 'flip' is no joke
Between 1997 and the time of writing (December 2008), at least 37 eventing riders have died as a result of injuries incurred while competing in the cross-country phase of eventing at national or international level or at Pony Club. These 37 fatal falls have been at all levels of the sport, from domestic one-day events up to Regional Championships level, and they have occurred in most of the recognised eventing countries around the world, with concentrations in the United Kingdom (14) and the United States (8). Most recently, a young girl died after a rotational fall at a Pony Club event in England in October 2008; an American rider who had been severely injured in a fall in 2006 died of complications from her injuries in September 2008; an Irish eventing rider died in August 2008, probably as a result of an eventing fall two weeks earlier.
A telling point is that at least 25 of these 37 deaths have resulted from a somersaulting (rotational) fall of the horse, with 11 of the 16 deaths in 2007 and 2008 being reported as having resulted from a rotational horse fall.
Stay up and survive
In 1999 five eventing riders were killed in England within four months of each other, four of them as a result of a rotational horse fall, provoking a major crisis in the sport. The governing body for equestrian sport, the Fédération Equestre International (FEI), responded promptly to establish a high-level investigative committee to look at any common causes, and to put in place both immediate and long-term changes to reduce the risks for riders and horses. The committee articulated what everyone in the sport already knew, that reducing the chances of horses falling is the best single way to reduce risk for the riders:
A fundamental conclusion which pervades every detailed recommendation is that everything should be done to prevent horses falling: this single objective should greatly reduce the chances of riders being seriously injured, as well as significantly improving the safety of competing horses (authors' emphasis).
International Eventing Safety Committee (2000)
A frangible fence.
As it happened, this particular cluster of rider deaths in 1999 also had other significant repercussions for the sport, which has since seen its Olympic status seriously under threat and its basic structure at the three-day event level profoundly altered. The International Olympic Committee, under pressure from non-equestrian member countries who see all forms of equestrian sport as elitist and exclusive, put pressure on the FEI to improve its safety record and also to adapt the format of the sport to make it more television-friendly, under threat of expulsion from the Olympic family. Only last-minute concessions from the FEI secured eventing's continuation as an Olympic sport, and the roads and tracks and the steeplechase components, which some would say were as much an intrinsic part of the sport as the exciting cross-country and the often nail-biting showjumping finishes, have been phased out finally and forever. Some diehard supporters of these two phases indeed claim that removing them from the mix increases the risk to horses, as they provided a safe way for horses to warm up for the most testing phase, the cross-country.
After the horror year of 1999, there were several more rider deaths in eventing, but probably neither more nor fewer than would reasonably be expected in a sport in which the vast majority of rider injuries are minor and insignificant, but in which the possibility of catastrophic results always exists.
However, the last three years have seen a terrible return to dark times. There have been 18 documented rider deaths in cross-country since the beginning of 2006, including one very high-profile accident at the Asian Games, recorded on news video and still freely available on the Internet. There were nine deaths in 2007 alone, with several reports alluding to another two (unconfirmed) deaths in that year. Already in 2008 riders five riders have died in competition, including one at Pony Club, two have died as a result of a previous fall, and several high-profile US riders have been critically injured.
There have also been an alarming number of horse deaths in eventing in recent years. However, accessing information about these is not easy, as these deaths were often not reported in the past. Certainly in the last few years horse deaths in competition have been increasingly visible in news reports, but it is not possible to say whether the rate of horse deaths has increased over time. Regardless of this, however, at least 19 eventing horses, many of them top-level performers, have died since the beginning of 2007, most of them in the US (Horsetalk.co.nz, 2008c).
The Quiet Man and Sarah Hansel fall at fence
13 at the Kentucky three-day-event in April
2008. The Quiet Man was later euthanised.
© Tom Eblen/Lexington Herald-Leader
It appears that despite all the positive responses to the 1999 disaster, including rule changes governing rider and horse elimination after a fall, improvements in fence and course design, improved levels of reporting by officials, and recent very high levels of equestrian community engagement in the topic, horses are still catapulting over cross-country fences and landing on their fallen riders.
In Australia, we recently completed a five-year surveillance program which collected information on rider and horse injuries resulting from falls in cross-country (O'Brien and Cripps, 2008). We identified 57 rotational horse falls out of a known 271 horse falls during the study period (2002 to 2006 inclusive), a rate of about one in five. This is certainly an under-representation of the true figures both for rotational falls and for horse falls in general, because the current scoring system hides all horse falls behind the notation 'Eliminated', and if we were not given Fall Report Forms from an event, we could not identify horse falls from any other source.
This was a major concern for our research. Until recently, the only rider falls visible in results were those in which riders continued after their fall, visible because their penalty score ended in a '5' (65 for a fall, 85 for a stop and a fall, 105 for two separate stops and a fall, 125 for two stops at the same fence, as well as a fall). If a rider had a fall and then decided to retire, if they had two falls and were eliminated, or if they had a fall and were eliminated for another reason, this information also disappeared. Since riders can be eliminated for any number of reasons, ranging from missing a fence to having more than the allowed number of refusals over the course, as well as for a fall of horse or rider, simply lumping them all together as 'Eliminated' prevents any analysis of either the number or the rate of horse and rider falls.
This situation has now been compounded by new rules which were introduced by the FEI in time for the Beijing Olympic Games in August 2008 (and introduced in Australia on September 22, 2008). These new rules mean that a rider who has a fall associated with attempting to jump a fence is automatically eliminated, thus also disappearing from the results. From now on, all falls of riders as well as those of horses have effectively been obscured from view in official results. Only fence-by-fence analysis, which is generally not accessible publicly, will reveal the true figures about rider and horse falls, and so it will appear that no riders or horses ever fall in eventing. Of course, this could not be further from the truth.
If preventing horse falls is identified as the single best way to reduce risks to riders, how are governing bodies to know whether they have succeeded in this goal if they cannot measure the size of the problem?
Knowing the score
We have twice proposed to the FEI and the Australian national body, the Equestrian Federation of Australia (EFA) that a few simple additions to the current scoring system would allow information on horse and rider falls to be visible and accessible. These proposed new codes are detailed below:
All the incidents described in this chart are currently invisible from results, four of them disappearing behind 'E', the other not being recorded at all. All would be visible with the introduction of these new scoring codes, thus providing important information about all rider and horse falls, and forming the basis of accurate and comprehensive data collection. In 2006 Eventing New South Wales did start using such a scoring system at most events in that State, and in late 2007 the National Eventing Committee, which governs the conduct of the sport at the national level, did move to adopt this revised scoring system on trial. Unfortunately this 'trial' did not seem to progress very far in 2008. The proposal attracted mild interest from the FEI, but there has not yet been any official change to their scoring requirements.
A horse and rider take a tumble at the 2006 World
Equestrian Games. © Evalyn Bemis
Measuring the risk
Eventing has long been considered a risky sport. One study in particular (Paix 1999) determined that the human fatality rate in eventing was greater than that in motorcycle racing, based on hours of participation. Another recent magazine article (Newton 2007) claims that the number of deaths in eventing (not specified) is greater than those in football, boxing and motor racing combined, although the author offers no explanation of how this assertion is arrived at.
Our national surveillance study (O'Brien & Cripps 2008) is the first to collect national data over time at all events in Australia conducted by the EFA. It covers five years of data at 445 events, involving more than 58,000 individual starts in the cross-country phase of the competition, and nearly 1.5 million jumping efforts. The study identified 1732 rider falls, and a follow-up questionnaire was sent to almost all these riders, with over 50% return rate. The study identified 271 horse falls, 57 of these being rotational horse falls. Because of difficulties in obtaining complete data on rider and horse falls, it is certain that there were many more rider and horse falls than we were able to identify.
The time of greatest risk for a rider and horse in eventing is when the combination is actually jumping, so it is important to measure risk in a more refined way than just dividing the number of competitors by the number of falls, for example. The scope and breadth of the national data collected in this project allowed us to make an estimate of risk based on a rider's exposure to the greatest risk, that of jumping a fixed fence at speed, atop a 500kg creature with the capacity to make its own decisions.
The study shows that the rider fall rate in eventing is about 12 falls for every 10,000 jumping efforts. In practice this means that at an average weekend one-day event with around 200 competitors, riders will jump about 5000 fences between them (at an average of 25 per round). It is likely that about six riders will fall off, and about one or two of them will be injured, these injuries most likely being minor. The majority of riders who fall off will not be injured, and indeed many riders who returned a questionnaire in this study challenged the definition of a fall, commenting that since they were not hurt, they did not consider that they had fallen at all.
A British rider falls during a three-day-event.
© Jan Milne
At the international level of the sport, the reporting requirements are now stringent: all details of rider and horse falls are documented, collated and analysed, and indeed the FEI uses the number and rate of rotational horse falls as a Key Performance Indicator (KPI) for international competitions. It is interesting, however, that the FEI aggregates figures for riders who are seriously injured with those whose injuries are fatal, thus effectively hiding the actual number of fatalities (Fédération Equestre International 2008b).
British Eventing (BE) has been monitoring risks in eventing since 2002, collecting data on rider injuries and deaths, numbers of rider and horse falls, the type of fence involved in the fall, and the type of horse fall (rotational or not). BE also uses a KPI system to assess each and every eventing course, with an informal inquiry being held if an event returns a KPI score above a certain level. BE has recently announced that the risk of serious or fatal injury in eventing in Britain has decreased from 0.24% in 2002-2003 to 0.07% in 2007-2008. BE's Safety Officer is quoted in Horse and Hound in November 2008 as saying:
We've got to keep the number of rotational falls moving down. I don't think we can ever eliminate the risk, but what we need to do is manage the risk as much as we can.
Horse and Hound (November 2008)
On the other hand, the report from an FEI Eventing Safety Forum organised by the FEI in January 2008 (Fédération Equestre International 2008a) indicates that the rate of rotational horse falls at international level competitions overall increased from 21% of all horse falls in 2006 to 26% of all such falls in 2007, with about 30% of riders involved in a rotational fall being seriously or fatally injured.
In Australia, no falls data have been collected since our five-year study was completed at the end of 2006, nor is there any indication that a uniform transparent scoring system will be adopted which would allow ready access to information about all rider and horse falls.
The future for eventing
One might also question the appropriateness of the recent 'challenge' event conducted in conjunction with the Sydney Three Day Event, in which a group of Asian riders on borrowed horses competed in a CCI*. These riders are coached by local accredited coaches for a short time before the event, and ride appropriately qualified horses. However, the riders themselves need no such qualifications, and as one commentator remarked:
It would be fair to say that the challenge started from day one as many of our top coaches began to teach these keen Asian riders, many of whom have never left the safety of an arena and didn't know how to gallop or jump.
The Eventer (2008)
Rather than supporting gimmicky variations on the sport, the governing bodies of eventing should be concentrating on doing everything possible to minimise those risks in the sport which are amenable to outside influence.
A horse and rider come to grief at Burghley.
© 2008 Adam Wynne
These actions to minimise risk actions should encompass rule and format changes, changes in fence construction and design, improvements in rider and coach education and accreditation, improvements in the training and accreditation of eventing horses, an overhaul of course designers' training and accreditation, and closer liaison between the various levels of the sport. It is reassuring to see that the FEI has now also turned its attention to the role that riders themselves can play, along with officials and coaches, in continuing to make the prime goal that of preventing horses from falling (Fédération Equestre International (2008c).
However, any suite of risk assessment and risk reduction strategies must inevitably include comprehensive data collection and analysis, so that future changes are coherent, coordinated and evidence-based.
Eventing is a sport in which chance events and bad luck will continue to play a significant part, despite all the best efforts of participants, rule-makers, and organisers. But perhaps it is time to articulate what everyone involved in the sport already knows but may not dare to speak aloud: if you go eventing, there is a chance that you will die.
FEI (2008a), Eventing Safety Forum, Report and Conclusions; Date accessed: January 2008
FEI (2008b), FEI Eventing safety statistics 2007; Date accessed: August 2008
FEI (2008c), Eventing Safety Program; Date accessed: December 2008
Horsetalk (2008a), Organisers saddened at death of Call Again Cavalier; Date accessed: November 2008
Horsetalk (2008b), British clean sweep in Express Eventing; Date accessed: November 2008
Horsetalk (2008c), Eventing accidents and safety news; Date accessed: December 2008
International Eventing Safety Committee (2000), The International Eventing Safety Committee Report (the Hartington Report), FEI; Date accessed: 2000
Newton, R. (2007), Horse sense: Three-day eventing is the ultimate test of horse and rider, FXEXpress Publications Inc; Date accessed: 13 November 2007.
O'Brien, D. and Cripps, R.A. (2008), Safety for Horses and Riders in Eventing - the SHARE database, Rural Industries Research and Development Corporation, UF-11A, Kingston, ACT
Paix, B. R. (1999), Rider injury rates and emergency medical services at equestrian events, British Journal of Sports Medicine, 33 (1):46-48
Roome, P. (2008), Risk of serious or fatal injury while eventing in Britain is decreasing, Horse and Hound, November 3 2008; Date accessed: November 4 2008.
'Speak up about the upper level horses that have died in competition'; Date accessed: November 2008.