Article Text

Download PDFPDF

European consensus on epidemiological studies of injuries in the thoroughbred horse racing industry
  1. M Turner1,
  2. C W Fuller2,
  3. D Egan3,
  4. B Le Masson4,
  5. A McGoldrick3,
  6. A Spence5,
  7. P Wind6,
  8. P-M Gadot4
  1. 1British Horseracing Authority, London, UK
  2. 2FIFA Medical Assessment and Research Centre, Zurich, Switzerland
  3. 3The Turf Club, Dublin, Ireland
  4. 4France Galop, Paris, France
  5. 5Darley Stud Management Co Ltd, Newmarket, UK
  6. 6Direktorium für Vollblutzucht und Rennen, Köln, Germany
  1. Correspondence to Dr Michael Turner, British Horseracing Authority, 75 High Holborn, London WC1V 6LS, UK; mturner{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Reported data for injuries sustained by jockeys in thoroughbred horse racing vary greatly. While the range of abilities and types of racing may account for some variation, the variations result mainly from differences in the definitions and methodologies employed in the studies.

The British Horseracing Authority (BHA) facilitated a meeting of experts from the four leading horse racing countries in Europe. Using an approach adopted by the previous consensus groups addressing sports injuries, issues related to definitions, methodology and implementation were discussed and voted on by the group during a structured one-day meeting. Following this meeting, two members of the consensus group produced a draft document, based on the group discussions, which was circulated for review; two revisions were prepared before the final consensus statement was produced.

The definition of injury and the criteria for recording the severity and nature of injuries are proposed for use in future epidemiological studies of injuries sustained by riders and associated support staff in the thoroughbred horse racing industry. Suggestions are made for recording participant baseline information together with recommendations on how injuries sustained during racing and associated activities should be reported.

The definitions and methodology proposed for recording injuries, sustained during thoroughbred horse racing activities will lead to more uniform data being collected. The surveillance procedures presented here may also be applicable to other equine sports such as trotting, polo, show jumping and eventing.


Based on injuries reported to racecourse medical officers in France, Ireland and the UK over the period 1992–2001, jockeys have an incidence of 1–2 injuries per 1000 rides in flat racing and 6–12 injuries per 1000 rides in jump racing.1 Insurance claims from injured jockeys indicate lower incidences of 1 injury per 1000 rides in flat racing and 3 injuries per 1000 rides in jump racing2 but reports of this nature are known to under-estimate the total injury burden in a sport.3 Injuries to jockeys in North America4 ,5 and Australia6 have been reported but it is not possible to deduce the incidence of injury in these countries because the reports did not include exposure data. Although variations in the reported incidence of injury in Europe can be partially attributed to differences in racecourse conditions, styles of riding and the number of horses taking part in races, many of the studies involved retrospective analyses of data collected over several years by on-course medical officers so that differences will also be related to different methodologies. Jockeys, however, are only one of several stakeholder groups exposed to the risk of injury in the sport of horse racing: other occupational groups, such as stable and stud staff, veterinary surgeons and racecourse officials, are also at risk during the racing, transport and breeding of thoroughbred racehorses. Despite the risks associated with working with and breeding horses, there are a few studies reporting the incidence or nature of injuries sustained by non-riding employees in horse racing and most of the information that is available comes from retrospective surveys and insurance claims6 ,7 (and Filby M, Jackson C, Turner M. Only fools and horses? Accidents and injuries in the horseracing industry. J Occ Health 2011. (Submitted.))

A study by the Department of Health and Human Services in the USA concluded that the horse racing industry was a high-risk occupation.8 The study identified two important deficiencies in the industry's procedures for monitoring the risk of injury; namely, the lack of appropriate denominator data (exposure) to calculate the incidence of injury, and a standardised injury reporting system. The governing bodies of several sports have addressed similar problems by developing consensus methods for epidemiological studies.9,,11 This statement aims to continue the consensus approach by presenting definitions and procedures for epidemiological studies that are applicable to the specific requirements of people working with horses in the European thoroughbred horse racing industry.

Consensus procedure

Following discussions between the governing bodies of several equine sports in Europe, the British Horseracing Authority (BHA) facilitated the formation of a Horseracing Consensus Group with a remit to develop a consensus for the recording and reporting of injuries sustained by riders and non-riders in the European thoroughbred horse racing industry. Based on the successful outcomes from groups developing injury surveillance, consensus statements for football,9 rugby union10 and tennis,11 and a mixed methods consensus methodology that employed elements of the Nominal, Delphi, National Institutes for Health (NIH) and Glaser's State of the Art consensus approaches12,,14 was adopted as the most appropriate method to address the issues involved. In summary, the consensus process included the following stages:

  • Key stakeholders within European horse racing identified the need for a consensus statement on injury surveillance in horse racing (Glaser). Subsequently, a group of experts with experience of managing injuries in the horseracing industry was identified. (Delphi, Glaser, Nominal). A non-medical moderator with appropriate experience was appointed to facilitate the consensus meeting (Glaser, Nominal).

  • A formal, structured procedure for conducting the consensus meeting was circulated and agreed to before the meeting (Nominal). An agenda and the previously published consensus statements for football and tennis were circulated to all members of the group to provide the basis for discussions; participants were requested to identify topics they considered should be added to the agenda (Delphi).

  • During the consensus meeting, topics on the agenda were discussed in an open forum, including making suggested modifications to the proposed definitions and procedures (Nominal). The moderator ensured that all attendees were given the opportunity to present their views on each topic but the discussions were constrained within the agenda and the total time available. On conclusion of the discussion about each topic on the agenda, the moderator called for a vote on each proposal (NIH) with a minimum of six of the seven attendees (86%) agreeing on an item for its acceptance.

  • Following the consensus meeting, a draft statement was prepared by the two lead authors based on the discussions and voting outcomes recorded during the meeting. This document was circulated to all group members with a request for comments to be returned within a 3-week period. Comments returned were reviewed by the two lead authors and, where appropriate, incorporated into a subsequent revision of the statement. Where a suggested change was not incorporated, a written explanation was prepared and circulated to all members of the group. This process was repeated for two iterations (Delphi), after which all members of the group confirmed their agreement with this final statement.

The consensus meeting took place over one day in February 2011 at the BHA offices in London.


Thoroughbred horse racing, which includes racing in France known as ‘Galop racing’, differs from other sports because it involves a wide range of human-horse interactive activities, such as riding, training, providing shelter, healthcare, transport, race course management and breeding, each of which can result in people being injured. The conclusions and recommendations presented in this statement drew extensively on the proposals made for football, rugby union and tennis but they also reflect a number of specific issues facing the horse racing industry, which makes this the first consensus statement to include criteria for recording and reporting injuries sustained by non-athletes. The discussion is presented in a similar format in sequence to that used in previous consensus statements to provide a convenient format for comparisons.



An injury is defined as:

Any physical complaint sustained by a person that results from competitive riding, training or other recognised activity that brings a person into contact, or in close vicinity and with the potential for contact, with one or more thoroughbred racehorses, irrespective of the need for medical attention or time loss from horse racing activities.

Although direct contact between the horse and person is the main cause of injuries in the horse racing industry, it is not a requirement of the definition; however, an injury must result from direct interaction between a person and a horse for it to be included. For example, an injury sustained by a stable lad falling over while moving away from a rearing horse would be included. Injuries that result from activities in the equine industry that do not involve direct interaction between a person and a horse, such as office work, course maintenance, on-course betting and catering, should not be included.

Most surveillance studies will record injuries that require medical attention or result in time loss from equine-related activities. In this context, medical attention refers to an assessment of a person's injury by a qualified medical practitioner, such as a doctor or physiotherapist.

Recurrent injury

A recurrent injury is defined as:

An injury of the same type and at the same site as an index injury, and the one that occurs after a person's return to full participation in equine-related activities following the index injury.

Injuries such as contusions and lacerations should not be recorded as recurrent injuries, as they are unlikely to be related to a previous injury.

Severity of injury

Severity of an injury is defined as:

The number of days that have elapsed between the date of onset of the injury to the date of the person's return to full participation to their normal equine-related activities.

The assessment of severity (days lost) commences on the day following the injury if the person is unable to take part in their normal equine-related activities. The day on which an injury occurs and the day the person returns to equine-related activities do not count towards the severity of the injury. A ‘career-ending’ injury is one that leads to a person's retirement from their normal equine-related activities.

Classification of injuries

Whenever possible, a qualified medical practitioner should provide a written diagnosis of each injury or preferably use a sport-specific injury code, such as the Orchard15 or University of Calgary16 coding systems. In addition to recording whether an injury is an index or a recurrent injury, injuries should also be classified according to their mode of onset, body location and side and type.

Acute and gradual-onset injuries

Injuries can be classified according to the way in which they are present. An acute-onset injury refers to an injury resulting from a specific, identifiable event or when there is a sudden onset of (relatively severe) pain or disability. A gradual-onset injury refers to an injury that manifests itself over a period of time, or when there is a gradual increase in the intensity of pain or disability, without a single, identifiable event being responsible for the injury. Examples of acute-onset injuries include muscle tears and fractures, while gradual-onset injuries include patella tendinopathy and osteitis pubis.

Location of injury

The location of the injuries is categorised into two levels: Level 1 – four general body regions; Level 2 – substructures within the four general body regions; see table 1. In larger epidemiological studies, data should be recorded using detailed injury codes, which allow information to be collapsed into the generalised Level 1 and Level 2 groupings.

Table 1

Main groupings and subcategories used for classifying the location of injuries

Type of injury

Injuries should be categorised according to the tissue affected using one of levels of categorisation: Level 1 – five general tissue types; Level 2 – subtypes within the five general tissue types; see table 2. In larger epidemiological studies, data should be recorded using detailed injury codes, which allow information to be collapsed into the generalised Level 1 and Level 2 groupings.

Table 2

Main tissue groupings and subcategories used for classifying the type of injury


Exposures are grouped into one of three possible categories:

Race exposure

Race exposure is defined as:

A competitive ride under the jurisdiction of a Horseracing Authority recognised by the International Federation of Horseracing Authorities.

A competitive ride commences at the moment the jockey mounts the horse and ends at the moment the jockey dismounts from the horse. For race jockeys, the number of competitive rides should be recorded and categorised against the type of race and the type of jockey (see table 3).

Table 3

Classification of types of race exposure and categories of jockey

In general epidemiological studies of injuries, it is not necessary to record the length of races or the number of fences in each individual jump race. However, in some studies these may be important risk factors, in which case this more detailed information should be recorded for the purposes of the study.

Training exposure (schooling and riding work)

Training exposure is defined as:

Time spent riding a thoroughbred racehorse for the purpose of improving the fitness and/or ability of the jockey and/or the horse.

Training exposure should be recorded as hours mounted on the horse and categorised against the type of training and the type of jockey (see table 4).

Table 4

Classification of types of training (riding out) exposure and categories of jockey

Other exposures (non-riding activities)

All other forms of non-riding exposure are defined as:

The time spent working with one or more thoroughbred racehorses where there is actual, or the potential for, person – horse contact or interaction.

An individual's exposure should be recorded as the number of hours working with horses irrespective of the number of people working with the horses or the number of horses involved; for example, three people working with five horses for 2 h should be recorded as 6 h of exposure (3 × 2 h). Exposure should be categorised against the type of activity involved (see table 5). Social activities with horses that are not scheduled as part of a person's formal activities should not be included as exposure. For studies investigating the synergistic risks of working with several horses at the same time, it may be appropriate to differentiate between an individual's exposure to one or several horses.

Table 5

Classification of the types of non-riding exposure and subcategories of activity

Recording information

Studies should be approved by a recognised institutional ethics committee. As recommended in previous consensus statements,9,,11 a prospective, cohort design should be used with standardised paper-based or electronic data collection forms available. The provision of detailed guidance documents for studies enhances the accuracy and reliability with which report forms are completed and data are collected. It is essential that all injuries are recorded and transmitted in a secure format to preserve the confidentiality of personal information.

Participant baseline information form

Baseline information required in most surveillance studies should include a person's study reference number, age, gender, height, body mass, and normal working activity within the horseracing industry (see table 5). Jockeys should also record their riding standard (full, apprentice/conditional, amateur).

Injury report form

The report form should provide the person's study reference number, date of injury, the date of the person's return to full participation to their normal equine-related activities, whether the injury was sustained during racing, training, or other activity (see tables 35), information describing the circumstances leading to the injury including the area where the injury was sustained (see table 6), the cause of the injury (table 7) and the protective equipment being worn at the time of injury (helmet, safety vest, goggles, mouth guard, safety boots, other). The nature of the injury (acute or gradual-onset; body location, type and side; index or recurrence) should also be recorded on the form. A section should be provided to enable a specific pathology or injury classification code to be recorded; free-text sections may be required for some studies if additional study-specific information is required. Report forms should be completed as soon as possible after injuries are sustained but information should be updated as additional information about the injury management becomes available, such as the use of imaging or surgery. Wherever possible a physician, physiotherapist or equivalent medically qualified person should provide the medical data.

Table 6

Classification of the location where injury occurred

Table 7

Classification of the cause of injury

Exposure form

This form should record the date, type and number or duration of exposures as appropriate (see tables 35).

Reporting information

The cohort should be clearly defined at the beginning of the study with the number of participants and the participants' ages (mean, SD and range) and gender reported as a function of the horse-related activities (see table 5). Numbers of injuries and exposures should be reported separately for race, training and other activities, with the incidence of race injuries reported as the number of injuries/1000 rides; training injuries as injuries/1000 training-hours and other injuries as injuries/1000 activity-hours.

The overall severity of injuries should be reported separately for race, training and other activities as both the mean and the median number of days-lost; however, severity can also be grouped according to the period of time lost – namely, slight (0 days), minimal (1–3 days), mild (4–7 days), moderate (8–28 days), and severe (>28 days). Catastrophic injuries, such as paraplegia and fatalities, and career-ending injuries should be reported separately.

This paper provides a proposal for the definition of injuries that should be recorded in epidemiological studies in the thoroughbred horse racing industry, including criteria for recording the severity and nature of these injuries. Recommendations are made for recording individual participant's baseline information, and race, training and other exposures. The definitions and methodology proposed within this consensus statement should lead to more consistent and comparable injury data being collected in the European thoroughbred horse racing industry and this should in turn lead to the development of a standardised injury reporting form that could be used in the wider international setting of horse racing. The recommendations may also be appropriate for collecting injury data in other equine sports, such as trotting, polo, show jumping and eventing.


View Abstract


  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles