The goal of epidemiology is to identify multi-factorial causes of disease in populations and apply this knowledge to minimize the prevalence or impact of deleterious outcomes. In most species, the primary application of epidemiology is usually related to the control of infectious disease. Although there are obvious examples of infectious disease epidemiology applied to athletic horse populations (Equine Influenza,1,2 Hendra Virus,3 West Nile Virus,4,5,6) most epidemiological work in this area has focused on injury prevention and improving horse performance. The vast majority of this work has been conducted in Thoroughbred racehorses and as such most of this chapter will focus on that particular breed. As far back as the 1960s and early 1970s, the equine veterinary community started describing racehorse injury and fatality at a population level: leg problems in racehorses,7 sore shins,8 third metacarpal (MCIII) and third metatarsal (MTIII) fractures,9,10,11 carpal fractures12,13 and equine fatalities during racing on British racecourses.14 Since then, many case definitions have been used in the major epidemiological studies of racecourse injury or fatality. Choice of case definition within each study may be based on wider issues such as local or regional priorities or specific jurisdiction requests, but in most instances, case definition has typically been based on the limitations of data resolution (i.e. the level of detail contained within race reports). In early studies, this resulted in fairly broad case definitions. However, data recording has improved over time (possibly as a result of increased awareness of the importance of these data and developments in computational power). As a result, more recent studies have been able to describe and analyze the risk of specific injury types. There are relatively few studies investigating the risk of orthopedic disease in the Thoroughbred.15,16 Most descriptive studies have focused on the risk of musculoskeletal injury or fatality of Thoroughbreds during racing or training. In response to increased public cognizance of animal welfare issues, many of these studies have been driven by the racing industry in an effort to publicly demonstrate the impact of interventions to improve safety and minimize injury and fatality. For example, the risks associated with jump racing in Victoria, Australia were formally reviewed in 1994, 1998 and in 2002 in response to public demand for improved safety at jump races. It is important that these studies correctly report the level of risk (with both numerator and denominator) before an intervention is introduced, so that the impact of an intervention can be assessed appropriately. A reduction in the annual number of fatalities at a particular racecourse should not be regarded as good evidence of a safer racecourse unless the annual number of race starts over the same time period is also reported. The risk of disease or injury is typically measured by prevalence, incidence risk, or incidence rate.17,18,19 Prevalence describes the number of instances of disease in a defined population at a specific point in time without any distinction between recent (acute) and long established (chronic) cases. Incidence risk (synonym, cumulative incidence) is a measure of the average risk of developing the disease during a specific time period and is a proportion (and therefore a value between zero and 100%).18 Mortality ‘rates’ are special measures of incidence used when the outcome is death.17,18 Studies of racecourse injury or fatality typically report ‘the number of events per 1000 starts’.20,21,22,23,24,25,26,27 Incidence rate uses the number of new cases that occur in a defined population of animals over a specified period at risk.17 This measure of risk is reported less frequently as it requires very detailed information on the time at risk. As such, it is best suited to studies of training injury where the number of events per 100 horse days (or months) in training (at risk) is reported.28,29,30,31,32 The risk of injury or fatality of Thoroughbred horses due to catastrophic musculoskeletal injury whilst racing varies within and between countries and also with race type. Since 1983, there have been multiple studies in North America, Great Britain and Australia that have described the risk of fatality due to catastrophic musculoskeletal injury in different race types: flat33,34,35,36,37 and jump.35,37 In New York, between 1983 and 1985, there were 2.1 fracture injuries per 1000 starts on dirt tracks and 1.1 per 1000 starts on turf tracks.33 In Kentucky, between 1992 and 1993, there were 1.4 catastrophic injuries per 1000 starts34 and in California 1.7 fatalities per 1000 starts has been reported.36 In the UK, between 1987 and 1993 there were 0.33 catastrophic fractures per 1000 flat racing starts, 1.4 catastrophic fractures per 1000 hurdle racing starts and 2.3 catastrophic fractures per 1000 steeplechase racing starts.35 In Australia, the risk of serious musculoskeletal injury (resulting in euthanasia or a failure to return to racing for at least six months) on city tracks in Victoria was 0.6 per 1000 starts in flat races, 6.3 per 100 starts in hurdle races and 14.3 per 1000 starts in steeplechase races.37 Typically, the risk of injury in these studies has been calculated from veterinary reports on racing injuries (numerator) and centrally collected racing data on the number of starts (denominator) from different racing jurisdictions.20,35 In some studies, the site of injury was identified in these veterinary reports. In those instances, the majority of musculoskeletal injuries were reported to have affected the bones of the distal forelimb. In North American studies, fractures of the proximal sesamoids were the most common injuries, followed by fractures of MCIII and carpal bones.25,34,38 In the UK, lateral condylar fractures of MCIII were the most common racing injuries.14,35,24 Differences between countries in the anatomic location of fractures may point to differences in the pathogenesis of different fracture types. Other studies have initiated or used on-going post-mortem examinations to ensure accurate injury classification.14,24,25,39,40 The cost of racecourse post-mortem examination programs often restricts their implementation, but unpublished data would suggest that in order to investigate accurate and precise case definitions, they are invaluable.41 In 1990, the California Horse Racing Board (CHRB) requested that the California Veterinary Diagnostic Laboratory System carry out a post-mortem examination on all horses that die at California racecourses under the jurisdiction of the CHRB. The risk of catastrophic musculoskeletal injury (of which 88% (69/78) were fractures) was 1.7 per 1000 starts on these tracks.26,36 Similar post mortem examination programs in the UK and Australia have subsequently highlighted the value of such examinations to enable the diagnosis of fracture at specific sites.24,39 For example, in the UK, lateral condylar fractures of MCIII were most common in national hunt racing (from 0.28 per 1000 starts in hurdle racing to 0.96 per 1000 starts in national hunt flat racing), whereas proximal phalangeal fractures were most common in turf flat racing (0.16 per 1000 starts) and biaxial sesamoid fractures were most common in all-weather flat racing (0.39 per 1000 starts).24 Continued collection of cases enabled the identification of risk factors for precise fracture definitions such as lateral condylar fracture of MCIII/MTIII.42 In Victoria, Australia, the most common catastrophic musculoskeletal injury in horses submitted for post mortem examination was fracture of the MCIII or MTIII.43 The second most common injury was the concurrent fracture of MCIII or MTIII and the proximal sesamoid bones and proximal phalanx. Although extremely costly and requiring a great deal of industry initiative and cooperation, further efforts to introduce routine post-mortem examinations in other jurisdictions should be encouraged. There are fewer reports describing the level of risk of soft tissue injury during racing and all of these studies focus on tendon injuries.20,22,44 The dearth of studies in this area of research is most likely due to lack of good quality data. Many soft tissue injuries do not require immediate euthanasia and may indeed only become clinically apparent once the horse has left the racecourse. As a result, these injuries may not be recorded by racecourse veterinarians and many of the estimates that are derived from racecourse reports are likely to be underestimates of the true level of risk. In Hong Kong, where detailed data on all racehorses in training and racing is recorded, Superficial Digital Flexor Tendon (SDFT) injury is the most common reason for horse retirements from racing (14%).44 This highlights the importance of long-term soft tissue injury in curtailing the racing career of Thoroughbreds. In the USA, the earliest report on the frequency of soft tissue injury comes from work on 24 race courses where 0.9 and 0.6 severe tendon injuries per 1000 starts on dirt and turf tracks, respectively were reported.45 These figures are comparable with later reports in the UK where the majority of races are on turf. The risk of injury to the suspensory ligament (SL) and superficial (SDFT) and deep digital flexor tendons (DDFT) together in the UK varied from 0.78 per 1000 starts in flat racing up to 9.12 per 1000 starts in steeplechase racing.20 The majority of these injuries were likely to be associated with SDFT strain. In comparison, using data from 2000 to 2005 on the same UK racecourses, 0.55 SDFT strains per 1000 starts in flat racing, 6.6 per 1000 starts in hurdle racing and 8.0 per 1000 starts in steeplechase racing, were reported.22 In the UK, the incidence rate of non-traumatic fractures per total days in training in flat racing horses was estimated at 1.15 per 100 horse months,32 whereas in national hunt horses, the incidence rate was estimated as 1.5 per 100 horse months.46 In the same national hunt population, 2.1 tendon and ligament injuries per 100 horse months were reported.46 In New Zealand, incidence rates of 0.13 SDFT cases per 1,000 horse training days at risk and 0.12 suspensory apparatus (SA) cases per 1000 horse training days at risk were calculated.29 In that study, SA included both soft tissue (suspensory ligament and distal sesamoidean ligament) and proximal sesamoid bone injuries. In Japan, forelimb superficial digital flexor tendonitis has been reported as being significantly more common than SL desmitis in training and racing Thoroughbred racehorses.47 The prevalence of forelimb SDFT tendonitis and SL desmitis was 11.1% (1130 cases) and 3.61% (370 cases) of the population, respectively. Much research has already focused on the identification of risk factors for musculoskeletal injury in the Thoroughbred horse. Case definitions that have predominated include fatality, musculoskeletal injury and fracture. Recent studies have also included tendon injury.29,48 The majority of these risk factor studies have focused on flat racing.23,26,34,38,49,50 In the UK,51,52 Australia37,53 and the USA,54 risk factors for musculoskeletal injury or fatality in jump racing have been identified. Horse-level risk factors are likely to make an important contribution to the variation in the likelihood of injury or fatality. Some studies have attempted to investigate training regimens.28,48,51 A history of previous injury has also been identified as a risk factor. However, few studies have investigated this explicitly. Reardon and colleagues52 did identify an association between racecourse SDFT injury and previous racecourse SDFT injury, but this study did not include complete medical histories as data from training was not available. Horse breed is also likely to be one of the most important factors with respect to prevalence of different injury types (although it may simply be a partial proxy measure of the use to which that horse is put, which in turn determines how it is trained). However, risk factors like horse breed or gender36,50,55 are not readily modifiable and unlikely to be causal in their association with the outcome. The value of such findings is that they may provide insight into the pathogenesis of a particular injury type. Turf and ‘all weather’ tracks are different with respect to the risk of injury or fatality. In the UK24 and the USA,25 biaxial sesamoid fracture is the most common type of catastrophic fracture on all weather and dirt surfaces, respectively. In the UK, where the majority of racing is on turf, lateral condylar fractures of the third metacarpus are the most common type of fracture in national hunt racing and proximal phalangeal fractures predominate in turf flat racing.24
Epidemiology of orthopedic diseases in athletic horses
Epidemiology
The racehorse
Risk of orthopedic disease, injury, or fatality in Thoroughbred racehorses
The level of risk during racing
Risk of fracture (catastrophic or non-catastrophic fractures)
Risk of soft tissue injury
The level of risk during training
Risk factor studies
Intrinsic risk factors
Extrinsic risk factors
Racing surface
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
