Epidemiology of race day injury in young professional jockeys in Great Britain from 2007 to 2018: a retrospective cohort study

Abstract

Introduction:
There are limited injury data in professional horse racing, particularly by sex.

Objectives:
To describe injury incidence, characteristics and falls in male and female, flat and jump jockeys in Great Britain.

Design and setting:
Retrospective cohort study of professional jockeys in Britain.

Participants:
245 jockeys licensed between 2007 and 2017.

Primary and secondary outcome measures:
The primary outcome measure was injury on a race day. Injury incidence (per 1000 rides; per 1000 falls) was derived. Incidence-rate ratios (IRR) were calculated to compare incidence between flat and jump racing, male and female jockeys, and male flat and male jump jockeys for: (i) injury incidence, (ii) fall incidence and (iii) injuries per fall.

Results:
234 British professional jockeys were included. Jockeys were on average 19.5±2.0 years old at licence date, 79.9% male and 58.1% flat. The time of follow-up (racing in the study) was 3.7 (SD=2.3) years. There were 278 injuries, occurring in-race (81.7%), in the stalls (10.8%) or parade ring (6.1%). After one injury was removed to preserve anonymity, 57.2% were soft tissue injuries, 25.3% fractures and 10.5% concussion. There were 1634 falls, with 92% in male jump racing. The injury incidence was higher in jump racing (5.1 vs 1.0/1000 jockey rides). The falls incidence was 1.8/1000 rides in flat and 46.2/1000 rides in jump racing (IRR 0.04, 95% CI 0.03 to 0.04). There were over five times higher injuries/1000 falls in flat than jump racing (IRR 5.56, 95% CI 4.05 to 7.53). Male flat jockeys fell less than female flat (IRR 0.57, 95% CI 0.35 to 0.97).

Conclusion:
Most injuries occurred in-race and were soft tissue injuries. Jump jockeys fell more often than flat, and female flat jockeys fell more often than male flat. Flat jockeys injured more frequently when falling. No sex differences were seen for injuries per fall.

Full text

1
DaviesM, etal. BMJ Open 2021;11:e044075. doi:10.1136/bmjopen-2020-044075
Open access
Epidemiology of race day injury in
young professional jockeys in Great
Britain from 2007 to 2018: a retrospective
cohort study
Madeleine Davies ,1,2 Kate A Jackson,3,4 Anna Louise Mackinnon,3,5
Alison Turner,6 Kerry Kuznik,3,7 Jerry Hill,7 Julia L Newton,3,6
Maria Sanchez Santos 8
To cite: DaviesM, JacksonKA,
MackinnonAL, etal.
Epidemiology of race day
injury in young professional
jockeys in Great Britain from
2007 to 2018: a retrospective
cohort study. BMJ Open
2021;11:e044075. doi:10.1136/
bmjopen-2020-044075
►Prepublication history for
this paper is available online.
To view these les, please visit
the journal online (http:// dx. doi.
org/ 10. 1136/ bmjopen- 2020-
044075).
Received 21 August 2020
Accepted 21 July 2021
For numbered afliations see
end of article.
Correspondence to
Dr Madeleine Davies;
m. davies@ bath. ac. uk
Original research
© Author(s) (or their
employer(s)) 2021. Re- use
permitted under CC BY- NC. No
commercial re- use. See rights
and permissions. Published by
BMJ.
ABSTRACT
Introduction There are limited injury data in professional
horse racing, particularly by sex.
Objectives To describe injury incidence, characteristics
and falls in male and female, at and jump jockeys in
Great Britain.
Design and setting Retrospective cohort study of
professional jockeys in Britain.
Participants 245 jockeys licensed between 2007 and
2017.
Primary and secondary outcome measures The
primary outcome measure was injury on a race day. Injury
incidence (per 1000 rides; per 1000 falls) was derived.
Incidence- rate ratios (IRR) were calculated to compare
incidence between at and jump racing, male and female
jockeys, and male at and male jump jockeys for: (i) injury
incidence, (ii) fall incidence and (iii) injuries per fall.
Results 234 British professional jockeys were included.
Jockeys were on average 19.5±2.0 years old at licence
date, 79.9% male and 58.1% at. The time of follow-
up (racing in the study) was 3.7 (SD=2.3) years. There
were 278 injuries, occurring in- race (81.7%), in the
stalls (10.8%) or parade ring (6.1%). After one injury was
removed to preserve anonymity, 57.2% were soft tissue
injuries, 25.3% fractures and 10.5% concussion. There
were 1634 falls, with 92% in male jump racing. The injury
incidence was higher in jump racing (5.1 vs 1.0/1000
jockey rides). The falls incidence was 1.8/1000 rides in at
and 46.2/1000 rides in jump racing (IRR 0.04, 95% CI 0.03
to 0.04). There were over ve times higher injuries/1000
falls in at than jump racing (IRR 5.56, 95% CI 4.05 to
7.53). Male at jockeys fell less than female at (IRR 0.57,
95% CI 0.35 to 0.97).
Conclusion Most injuries occurred in- race and were soft
tissue injuries. Jump jockeys fell more often than at,
and female at jockeys fell more often than male at.
Flat jockeys injured more frequently when falling. No sex
differences were seen for injuries per fall.
INTRODUCTION
Professional horse- racing in Great Britain
(GB) is an industry worth more than
£4 billion annually and employing over 20 000
staff.1 Employment as a professional jockey
carries a risk of injury that can be severe
and career ending, with associated financial
consequences for the jockey and industry.2
Therefore, understanding the epidemi-
ology of injury in jockeys provides important
information to jockeys, physiotherapists and
medical staff and the broader racing industry,
to inform strategies to reduce injury risk,
ensure optimal readiness for participation
and support the planning of medical and
financial support for injured jockeys.
In GB, professional jockeys are insured with
the Professional Riders Insurance Scheme
(PRIS), and professional horseracing is cate-
gorised into flat and National Hunt (jump)
racing, Nearly 50% of PRIS claims are for frac-
tures,2 while jump racing injuries are respon-
sible for the majority of insurance claims.
There have been efforts to describe injury
epidemiology and risk factors in horseracing,
most recently in countries such as the USA,
Strengths and limitations of this study
►This study is the rst to report in- race injury rates
in male and female, jump and at jockeys across
Great Britain.
►These data are representative of all British jockeys
licensed between July 2007 and November 2017.
►Jockeys observed from the point of licensing may
be younger, tter and more recently fall- trained than
all professional jockeys irrespective of license date,
and therefore these injury data may not be repre-
sentative of all current professional jockeys.
►As there are multiple races per day in horseracing,
injuries could not be matched with certainty to a
specic race and were therefore matched by race
day.
►As each injury was matched by day, we are unable
to differentiate between concomitant injuries in one
race or in two separate races on one day.

2DaviesM, etal. BMJ Open 2021;11:e044075. doi:10.1136/bmjopen-2020-044075
Open access
New Zealand and Australia.3–13 A recent study in Ireland
by O’Connor et al found that there had been a significant
increase in injuries per 1000 falls and falls per 1000 rides
in jump jockeys when comparing injury incidence prior
to 2006, to those from 2011 to 2015.14 The most recent
GB jockey injury data were published over 15 years ago15
and may not reflect changes to the sport’s organisation
and novel athlete initiatives which have been developed
in this time. Mandatory residential courses at the point
of jockey licensing in GB have been developed (7- day
course from 2000 onwards, 14- day residential course
from January 2017) to: embed fitness assessments, diet
and nutrition training, sports science awareness, rule and
regulation information and riding work. This programme
of training also includes falls training and simulated
assessments at the point of licensing. To date, there has
been a limited research focus on recently licensed profes-
sional jockeys. A formal licensing process and residential
course may, however, provide intervention opportunities
as jockeys are obtaining their professional license to better
prepare them for a professional career in racing. Detailed,
current epidemiological injury data are required to iden-
tify prominent injury types and characteristics which can
become targeted prevention opportunities.
The aim of this study is to describe injury incidence
over an 11- year period in a sample of young professional
male and female, flat and jump jockeys in British horse
racing. Secondary aims were to describe injury type, loca-
tion, severity (time loss) and performance (falls and wins)
in GB horseracing.
METHODS
Study design, setting and participants
Using a retrospective cohort study design, a sample
of 245 professional jockeys were selected. Eligibility
was defined as a jockey who was licensed as a profes-
sional jockey for the first time between July 2007 and
November 2017.16 Eligibility (licensing) was achieved
after attendance at a licensing course at either the
British Racing School or National Horseracing College
(formerly Northern Racing College) or via the British
Horseracing Authority. The cohort was therefore
jockeys beginning their professional career, and data
were extracted from that point onwards.
Variables and data sources
Since 1992, it has been compulsory for all incidents
sustained on a licensed racetrack in GB to be reported
to the Racecourse Medical Official and recorded on
one central database. Performance data are routinely
collected in the industry. All injuries and performance
data (rides, falls and wins) for eligible jockeys from
license period until 1 July 2018 were extracted.
Injury data for each incident included: race location
(GB and non- GB), body part, incident type, surface
type (jump/flat), resolution (cleared to return) date,
whether the injury is acute or recurrent and the outcome
of the incident (health service provision required). Only
injuries that were reported by the Medical Officer on
a race day in GB (ie, not in Europe or internationally)
were included (n=278). Severe injury was defined by
time loss duration (from racing), with 28 days or more
between the injury date and resolution date considered
as severe.17 In cases where a single incident resulted
in more than one injury, each body part or injury type
was counted as a separate injury. Body site of injury was
grouped according the Orchard Sports Injury Classifi-
cation System.18 Injuries during the study period were
excluded from analysis if they occurred outside of GB,
in races prior to professional licensing or as an amateur
jockey. In- race injury incidence was estimated using only
those injuries occurring specifically during a race expo-
sure and not those occurring in the parade ring, stalls or
between the parade ring and the start of the race.
Performance data available in Wetherby’s perfor-
mance records include race date, finishing position,
surface type and data on non- completion of the race.
Falls were derived from Wetherby’s race non- completion
data, using the categories ‘fell’, ‘slipped up’, ‘brought
down’ and ‘unseated rider’. The Strengthening the
Reporting of Observational Studies in Epidemiology
(STROBE) guidelines were used to report methods
and findings.19 This study was performed in accordance
with the 1964 Declaration of Helsinki ethical standards.
Patient and public involvement
Patient and public involvement was undertaken for this
study. A forum was organised with former jockeys, trainers
and project stakeholders to establish the overarching
grant’s aims and work packages, of which this article is
one deliverable. Two former professional jockeys, and a
lay representative, resided on the study steering group
throughout. Therefore, there was involvement of these
groups in developing research questions and outcome
measurements, study design and processes, and informing
the study’s recruitment strategy and reporting of findings.
Statistical methods
All analyses were conducted using Stata V.15 statistical
software (Stata, College Station, Texas, USA). Descrip-
tive statistics (mean (SD) for continuous variables and
number (percentage) for categorical variables were
used to describe demographic, performance (wins) and
injury data. The injury incidence (number of injuries
per 1000 rides and falls) and number of falls and wins
per 1000 rides during racing and their 95% CI were
calculated. All results are presented separately for male
and female and for flat and jump jockeys. Incidence-
rate ratios (IRR) and 95% CI were calculated for flat
and jump, male and female flat jockeys and male flat
compared with male jump jockeys.
RESULTS
Professional jockey cohort characteristics
Two hundred and forty- five professional jockeys (90 968
rides), who received a British professional license between

3
DaviesM, etal. BMJ Open 2021;11:e044075. doi:10.1136/bmjopen-2020-044075
Open access
July 2007 and November 2017, were eligible for inclusion
in the study. Professional jockeys with at least one GB
professional race (figure 1) were retained (n=234; 90 711
rides),with an average time of follow- up (professionally
racing in the study) of 3.7 (SD=2.3) years. Jockeys had
a mean age at license date of 19.5 years (±2.0), and 47
(20.1%) were female (table 1). Jump jockeys comprised
41.9% of the sample; and 33.1% of flat jockeys and 2.0%
of jump jockeys were female.
Injuries
Two hundred and seventy- eight race- related injuries
were recorded over the study period, 175 of which were
in jump racing (62.9%) and 103 (37.1%) of which were
in flat racing. One female jump injury was recorded in
the study period; therefore, this injury has been removed
from type and severity (table 2), to maintain confidenti-
ality. However, this injury was retained for incidence anal-
ysis, as performance data is not restricted and is freely
available.
Soft tissue was the most common injury type (57.4%),
followed by fractures (25.3%), concussions (10.5%),
lacerations (4.7%) and dislocations (1.8%). One injury
(0.4%) was recorded as unclassified. Over a third of
injuries were severe in time loss duration (35.0 %), and
the median time loss from all severe injuries was 60 days
(IQR: 40–128). A higher proportion of injuries were
severe in duration for jump racing (40.8%) than in flat
racing (25.2%). Soft tissue injuries were the most prev-
alent injury type across male flat, female flat and male
jump racing. For male flat and male jump racing, the next
prominent injury type was fracture (table 2). However, for
female flat racing, concussion was the second most prev-
alent injury, followed by lacerations (table 2), while no
female flat jockeys sustained a fracture during the study
period. The greatest time loss injuries were fractures
(n=70), with a median time loss of 47.5 days, followed
by dislocations (n=5) with a median time loss of 47 days.
The majority of fractures (73.7%) were severe in time loss
duration. The least severe injuries for time loss were lacer-
ations (3 days), and soft tissue injuries (4 days).
Figure 1 Participant ow from injury data and performance data, to cohort of recently licensed professional jockeys with at
least one ride during the study period.
Table 1 Demographic characteristics of the study sample
Characteristic, n (%) N=234
Sex, n (%)
Male 187 (79.9)
Female 47 (20.1)
Type of surface, n (%)
Flat 136 (58.1)
Jump 98 (41.9)

4DaviesM, etal. BMJ Open 2021;11:e044075. doi:10.1136/bmjopen-2020-044075
Open access
The proportion of soft tissue injuries was lower in- race
than in stalls and/or parade ring (52.0% vs 81.3%), and
higher proportion of concussion and fractures occurred
in- race compared with in the stalls/parade ring (concus-
sion: 12.3% vs 2.1%; Fracture: 28.6% vs 10.4%).
Injury incidence, falls and wins per 1000 rides
Of the number of rides in the performance dataset over
the study period, 57 429 (63.3%) were in flat racing. There
were 1634 falls in the 90 711 races, of which 1539 (94.2%)
were in jump racing (table 3). The overall in- race injury
incidence in professional horse racing was 2.5 (95% CI
2.2 to 2.9) injuries per 1000 jockey rides, 1.0 (95% CI 0.8
to 1.3) in flat racing and 5.1 (95% CI 4.3 to 5.9) per 1000
rides in jump racing, representing an IRR of 0.20 (95% CI
0.14 to 0.27) for flat versus jump racing. The injury inci-
dence was significantly lower for male flat jockeys than
for male jump jockeys (IRR 0.18, 95% CI 0.13 to 0.26).
There were no statistically significant differences in injury
incidence between male flat (0.94 injuries per 1000 rides)
and female flat jockeys (1.42 injuries per 1000 rides (IRR
0.66, 95% CI 0.34 to 1.37)).
Falls were significantly more common in jump than flat
racing, with an incidence of 46.24 falls per 1000 rides in
jump racing and 1.77 falls per 1000 rides in flat racing,
representing an IRR of 0.04 (95% CI, 0.03 to 0.04) in flat
versus jump racing (table 3). There was a sex difference
in falls per 1000 rides, with female flat jockeys signifi-
cantly more likely to fall (2.61 falls per 1000 rides), than
male flat jockeys (1.55 falls per 1000 rides), representing
an IRR of 0.66 (95% CI 0.44 to 10.10) in male flat versus
female flat.
The overall injury rate per 1000 falls was 138.9; however,
this was significantly higher in flat racing (610.5/1000
falls) than for jump racing (109.8/1000 falls), repre-
senting an IRR of 5.56 (95% CI 4.05 to 7.53) in flat
versus jump racing (table 3). There were no statistically
significant sex differences in injury rate per fall between
male and female flat jockeys (IRR 1.16, 95% CI 0.60 to
2.40).
Overall, there were 97.9 wins per 1000 rides, with 95.0
wins per 1000 rides in flat racing and 102.8 wins per 1000
rides in jump racing. There was a higher rate of wins in
flat male than flat female jockeys (IRR 1.15, 95% CI 1.07
to 1.25) (table 3).
DISCUSSION
This study is the first in GB to report in- race injury rates
nationally by sex, in a defined sample of young, profes-
sional, flat and jump jockeys. Injury types were broadly
similar between flat and jump jockeys; however, more
injuries were classified as severe in jump than in flat
racing. Injury incidence was higher in jump racing than
flat racing; however, injury rate per fall was significantly
higher in flat racing than jump racing. There were sex
differences for fall incidence, pattern of injuries and
wins per 1000 rides. Strengths of this study include the
use of national data from official data sources for up to
11 years of professional racing (depending on license
date), and these data being representative of young
professional jockeys in GB for this period. This study
is the first to our knowledge which has excluded none
in- race injuries (parade ring, stalls, postrace) from inci-
dence estimates and calculated in- race injury incidence
for male and female jockeys. A significant limitation in
sport epidemiology research, including in racing, is the
paucity of data for female athletes. The presentation of
data stratified by sex here is a strength of this analysis.
Further strengths of this study include the duration of
data collection, standardisation of data collection and the
level of routine injury reporting on race days in GB, as
reporting and recording these injuries are stipulated in
racing regulations.
Table 2 Type and severity of injuries occurring in at, jump, male and female, young professional jockeys between 2007 and
2018 (n=278 injuries; n=277 once excluding female jump jockey)
Flat (n=103) Jump (n=174)
Total
Time- loss* (days)
(Median, (IQR))Male (n=86) Female (n=17) All Male
Type of injuries, n (%)
Fracture 15 (17.4) 0 (0.0) 15 (14.6) 55 (31.6) 70 (25.3) 47.5 (25–87)
Soft tissue 56 (65.1) 11 (64.7) 67 (65.1) 92 (52.9) 159 (57.4) 4 (1–13)
Concussion 9 (10.5) 3 (17.7) 12 (11.7) 17 (9.8) 29 (10.5) 24 (14–38)
Dislocation 1 (1.2) 0 (0.0) 1 (0.1) 4 (2.3) 5 (1.8) 47 (41–78)
Lacerations 5 (5.8) 2 (11.8) 7 (0.7) 6 (3.5) 13 (4.7) 3 (2–16)
Unclassied 0 (0.0) 1 (5.9%) 1 (0.1%) 0 (0.0) 1 (0.4) 235 (235–235)
Severe injuries, n (%)
28 or more days- loss 21 (24.4) 5 (29.4) 26 (25.2) 71 (40.8) 97 (35.0) 60 (40–128)
Results expressed as number (%).
*3 injuries did not resolve during the study period.

5
DaviesM, etal. BMJ Open 2021;11:e044075. doi:10.1136/bmjopen-2020-044075
Open access
There are limitations regarding the generalisability of
these results to the whole jockey population. This study
is representative of professional jockeys from the point
of licensing in GB. However, this younger (more recently
falls trained and observed from the point of licensing)
cohort of jockeys may not be representative of all current
(older) male and female, professional flat and jump
jockeys. There is the potential for younger jockeys to be
fitter and potentially more recently fall- trained and there-
fore demonstrate less falls and injuries per fall. However,
there may also be a survival effect, whereby jockeys
sustaining professional careers may be those who are less
likely to fall and have accumulated varied ‘successful’
(less injurious) falls experiences or are less likely to
injure when falling. Furthermore, these associations may
be confounded by their equine athlete pairings. Appren-
tice (newly licensed flat) and conditional (newly licensed
jump) jockeys may ride less reliable horses, who could be
at a higher risk of their own or jockey injuries. Recently,
licensed jockeys may also make more errors in judgement
resulting in horse falls or ride in lower grade races with
more falls around them. Previous research has found
experienced horses to have a lower falls incidence20 and
horse age and race class to be suggestive of increased like-
lihood of equine catastrophic musculoskeletal injury.21
Unfortunately, we have limited horse data (ie, behaviour,
age of horse, training, race load, horse injury) to support
discussion of horse experience and reliability.
Injuries could not be matched with certainty to a
specific race; therefore, we matched injury and perfor-
mance data by day. Each injury was analysed in isolation,
including any concomitant injuries in one race, or race
day. We were able to match all injury dates to a race date,
which is reassuring for data completeness in both data-
sets. The detailed level of data management is consid-
ered to be a strength of this analysis compared with other
previous epidemiological studies. While this is the best
data source available in GB for in- race injury capture, it
may not include jockeys who sought care for these inju-
ries in the days postrace and any jockeys who did not
present with injury.
The most common injuries were similar to those
described previously,15 with soft tissue injuries, fractures
and concussion most common, and fractures the greatest
time loss injuries. For female flat jockeys, concussion
was the second most prevalent injury, followed by lacer-
ations, with no fractures. However, for male flat jockeys,
the second most common injury was fracture followed by
concussion. Previous research has suggested the concus-
sion rate to be broadly higher for flat than jump jockeys,
although sex was not considered in this work.15 Periods
of relative energy deficiency22 have been discussed in
weight- restricted sports and are associated with changes
in bone density, which might also suggest sex differences
in fracture risk. Females may be more readily able to make
weight categories, particularly as apprentice jockeys, in
comparison with males where it has been suggested that
making weight with minimal disruption to well- being may
Table 3 Exposure, number and incidence of in- race injuries and wins for at and jump, male and female recently licensed professional jockeys
Flat (n=136) Jump (n=98)
Overall
IRR (95% CI)
Male Female All Male Female All Flat vs jump
Flat male vs
at female
Flat male vs
jump male
Race rides 48 986 8443 57 429 32 765 517 33 282 90 711 – – –
Injuries 46 12 58 168 1 169 227
Falls 73 22 95 1502 37 1539 1634 – – –
Wins 4746 709 5455 3365 57 3422 8877 – – –
Injuries per 1000
rides (95% CI)
0.9
(0.7 to 1.3)
1.4
(0.7 to 2.5)
1.0
(0.8 to 1.3)
5.1
(4.4 to 6.0)
1.9
(0.1 to 10.8)
5.1
(4.3 to 5.9)
2.5
(2.2 to 2.9)
0.20
(0.14 to 0.27)
0.66
(0.34 to 1.37)
0.18
(0.13 to 0.26)
Injuries per 1000
falls (95% CI)
– – 610.5
(463.6 to 789.2)
– – 109.8
(93.9 to 127.7)
138.9
(121.4 to 158.2)
5.56
(4.05 to 7.53)
1.16
(0.60 to 2.40)
5.63
(3.97 to 7.85)
Falls per 1000
rides (95% CI)
1.5
(1.2 to 1.9)
2.6
(1.6 to 3.9)
1.7
(1.3 to 2.0)
45.8
(43.6 to 48.2)
71.6
(50.4 to 98.6)
46.2
(44.0 to 48.6)
18.0
(17.2 to 18.9)
0.04
(0.03 to 0.04)
0.57
(0.35 to 0.97)
0.03
(0.03 to 0.04)
Wins per 1000
rides (95% CI)
96.9
(94.1 to 99.7)
84.0
(77.9 to 90.4)
95.0
(92.5 to 97.5)
102.7
(99.3 to 106.2)
110.3
(83.5 to 142.8)
102.8
(99.4 to 106.3)
97.9
(95.8 to 99.9)
0.92
(0.89 to 0.96)
1.15
(1.07 to 1.25)
0.94
(0.90 to 0.99)
Emboldened text denotes statistical signicance.
IRR, incidence rate ratio.

6DaviesM, etal. BMJ Open 2021;11:e044075. doi:10.1136/bmjopen-2020-044075
Open access
not be feasible.23 Previous research in this cohort has iden-
tified a very low bone density (defined as a z- score <−2.0),
in 29% of male, flat jockeys but not in female jockeys.16
Future research in this area could expand on this rela-
tionship with fracture risk, particularly given the nature
of 50% of PRIS claims being for fracture,2 and therefore
the financial implications of fractures for the industry, as
well as financially and functionally for jockeys.
Jump jockeys having a higher injury incidence than flat
jockeys, and falling significantly more than flat jockeys,
but flat jockeys being more likely to be injured when
falling are consistent with previous studies.15 18 Our focus
on true ‘in- race’ injuries could have contributed to a more
conservative estimate of injury incidence. However, race
day injuries in the stalls or parade ring are important and
have therefore been included in table 2, though excluded
from incidence analyses. In other sports, warm- up injuries
are considered separately to match injuries as these do
not involve match exposure.24 It has been recommended
in international consensus statements for reporting injury
and illness in sporting environments that training and
competition injury data should be reported separately,
as there is often markedly different injury risk between
these activities.25 For these reasons, we have focused on
in- race injuries as those where jockeys are competitively
exposed in competition with each other.
The higher injury incidence in jump racing is aligned
with the majority of PRIS insurance claims being in jump
racing26 and may be contributed to by the challenges asso-
ciated with jumping obstacles, becoming unbalanced and
unseated or brought down by other horses and jockeys
during these activities. These additional factors may also
be associated with the higher severity of injury seen in
jump racing. Given an awareness of upcoming obstacles,
jump jockeys may be able to foresee and prepare for a
fall if approaching an obstacle incorrectly or if another
horse has already fallen in front of them. Navigating
these obstacles also leads to slower racing speeds and a
subsequent slower speed of falling in comparison with
flat racing. These combined factors may contribute to
reducing injury risk per fall. Conversely, flat jockeys
generally travel at greater speeds, potentially with more
runners per race, which may mean that their injuries are
resultant of the fall itself and postfall circumstances. The
higher injury rate per fall in flat jockeys may be associ-
ated with the number of competitors, the force of ground
impact at speed, likelihood of trampling when fallen and
reduced reaction time to employ fall techniques. The
‘going’ (racing surface) may additionally be affected by
seasonality, with more jump racing in winter months, and
flat racing in the summer, with a harder contact surface
for any falls.
Female flat jockey fall rates (2.6 (1.6–3.9)/1000 rides)
were higher than male (1.5 (1.2–1.9)/1000 rides), which
is consistent with previous research.27 In male flat and
jump jockeys, fall rates were lower in our study than
previous studies, with rates of 1.5 falls/1000 rides and
45.8 falls/1000 rides, respectively. GB data from 2002
found 4.2 falls/1000 rides and 68 falls/1000 rides, respec-
tively, for male flat and jump jockeys,15 and an Australian
study estimated 4.2 falls/1000 rides and 52.6/1000 rides,
respectively.28 More recent data from Ireland reported
3.8 falls/1000 rides and 49.5 falls/1000 rides for flat and
jump jockeys, respectively.29 It has been suggested that fall
rates have decreased over recent years particularly in flat
(4% from 2005/2006 to 2017/2018) and also in hurdle
racing (2% from 2005/2006 to 2017/2018).20 Possible
explanations for this discrepancy may be differences in
the study population or methodological considerations.
Turner et al15 did not define a fall, and there is the poten-
tial for recorded falls to be at any point on a race day or
more strictly between the start of a race and finishing post.
O’Connor et al define a fall as ‘the rider being dislodged
from the horse, regardless of outcome’ which is similar
to our study. Using Wetherby fall data from all licensed
UK racecourses, we included all unseated in a race cate-
gory (‘fell’, ‘slipped up’, ‘brought down’ and ‘unseated
rider’).
For young professional jockeys, recent interventions
such as mandatory licensing courses, including falls
training and simulated riding assessment, may have
contributed to a reduced falling incidence over recent
years, compared with prior studies. There have addi-
tionally been changes in horse racing organisation and
practices, including jockey coaching from 2006, jockey
coaches accreditation from 2011, improved turf manage-
ment, improved personal protective equipment such as
helmets and body protectors, the introduction of safety
officers from 1985, improved fence and course design.
While these will not all impact fall rate, they may influ-
ence the injury rate per fall and safety in the horseracing
environment, in comparison with previously published
research.
CONCLUSION
Young professional jump jockeys in GB have a higher
in- race injury incidence and fall incidence than flat
jockeys; however, flat jockeys had a far greater injury rate
per fall. Flat and jump jockeys presented with similar
injury types, but more injuries were classified as severe
in jump than in flat racing. There were sex differences
for fall incidence and injury type, which warrants further
study. Future research in this population could be more
mechanistic in nature and focus on the higher injury
incidence and severity in jump racing and mitigating the
higher injury rate per fall in flat racing.
Author afliations
1Institute for Sport, Exercise and Health, University College London, London, UK
2Department of Health, University of Bath, Bath, UK
3Centre for Sport, Exercise and Osteoarthritis Research Verses Arthritis, University
of Oxford, Nufeld Department of Orthopaedics Rheumatology and Musculoskeletal
Sciences, Oxford, UK
4Sporting Knee Injury Prevention Programme (SKIPP), Headington, UK
5Injured Jockeys Fund, Newmarket, UK
6NIHR Oxford Biomedical Research Centre, Oxford, UK

7
DaviesM, etal. BMJ Open 2021;11:e044075. doi:10.1136/bmjopen-2020-044075
Open access
7British Horseracing Authority Ltd, London, UK
8Nufeld Department of Orthopaedics Rheumatology and Musculoskeletal Sciences,
University of Oxford, Oxford, UK
Acknowledgements We would like to acknowledge the project funders and
our project steering group for their sustained support throughout this grant. We
would also like to acknowledge the athlete representatives, patient and public
representatives and participants who have helped us to shape this project and
directed us towards addressing valuable questions for jockeys and the industry.
We would like to also acknowledge Dr Betsy Georgiou and Mr Aodhagan Conlon for
proof- reading our nal manuscript.
Contributors JLN, ALM, AT, KK, KAJ and JH contributed to the study conception
and design and prepared and obtained project funding. AT, KAJ, KK and ALM
obtained necessary data permissions. All authors provided project oversight, and
KK and JH provided support regarding horse- racing queries. MSS, KAJ and MD
prepared the draft versions and the nal manuscript. ALM and KK collected and
cleaned data, and MSS and MD prepared data for, and undertook, analyses. All
authors reviewed and approved draft and nal manuscripts.
Funding This research was supported by the British Horseracing Authority, the
Professional Jockeys Association (PJA) and the Injured Jockey Fund (IJF). This
work was also supported by the Racing Foundation (grant number HFR00920), the
British Association of Sport and Exercise Medicine (grant number HFR00840) and
the Centre for Sport, Exercise and Osteoarthritis Research Versus Arthritis (Grant
reference 21595).
Competing interests MD has received project funding from the Rugby Football
Union, Rugby Football Union’s Injured Players Foundation, British Medical
Association and British Horseracing Association. Since submitting this manuscript,
MD has become a sessional employee of the British Horseracing Authority,
supporting the COVID-19 response in racing. KK has no competing interests. After
completing her academic contract on this study, KK has commenced employment
in the Medical Department of the British Horseracing Authority. JH is employed
by the British Horseracing Authority. MSS is supported by the NIHR Biomedical
Research Centre at the University of Oxford.
Patient and public involvement Patients and/or the public were involved in the
design, or conduct, or reporting, or dissemination plans of this research. Refer to
the Methods section for further details.
Patient consent for publication Not required.
Ethics approval This study has received ethical approval from the University of
Oxford Central University Research Ethics Committee (MS- IDREC- CX-2014-147).
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement No data are available. No novel datasets were
generated during the current study. Anonymised injury data analysed in this study
were made available by the named collaborators for the purpose of this analysis
and ownership is retained by those collaborating organisations.
Open access This is an open access article distributed in accordance with the
Creative Commons Attribution Non Commercial (CC BY- NC 4.0) license, which
permits others to distribute, remix, adapt, build upon this work non- commercially,
and license their derivative works on different terms, provided the original work is
properly cited, appropriate credit is given, any changes made indicated, and the use
is non- commercial. See: http:// creativecommons. org/ licenses/ by- nc/ 4. 0/.
ORCID iDs
MadeleineDavies http:// orcid. org/ 0000- 0002- 6047- 7340
MariaSanchez Santos http:// orcid. org/ 0000- 0003- 1908- 8623
REFERENCES
1 . Returning safely – British racing goes back to work – key points
British Horseracing Authority; 2020. http:// media. britishhorseracing.
com/ bha/ covid19/ Resumption_ key_ points. pdf
2 Turner M, Balendra G, McCrory P. Payments to injured professional
jockeys in British horse racing (1996-2006). Br J Sports Med
2008;42:763–6.
3 Curry BA, Hitchens PL, Otahal P, etal. Workplace injuries in
thoroughbred racing: an analysis of insurance payments and
injuries amongst jockeys in Australia from 2002 to 2010. Animals
2015;5:897–909.
4 Curry BA, Hitchens PL, Otahal P, etal. Australian insurance costs of
jockeys injured in a race- day fall. Occup Med 2016;66:222–9.
5 Rueda MAF, Halley WL, Gilchrist MD. Fall and injury incidence
rates of jockeys while racing in Ireland, France and Britain. Injury
2010;41:533–9.
6 Hitchens PL, Hill AE, Stover SM. Jockey falls, injuries, and fatalities
associated with thoroughbred and quarter horse racing in California,
2007-2011. Orthop J Sports Med 2013;1:232596711349262.
7 Press JM, Davis PD, Wiesner SL, etal. The National jockey injury
study: an analysis of injuries to professional horse- racing jockeys.
Clin J Sport Med 1995;5:236–40.
8 McCrory Petal. An analysis of injuries resulting from professional
horse racing in France during 1991-2001: a comparison with injuries
resulting from professional horse racing in Great Britain during 1992-
2001. Br J Sports Med 2006;40:614–8.
9 Waller AE, Daniels JL, Weaver NL, etal. Jockey injuries in the United
States. JAMA 2000;283:1326.
10 Wylie CE, McManus P, McDonald C, etal. Thoroughbred fatality and
associated jockey falls and injuries in races in New South Wales
and the Australian Capital Territory, Australia: 2009-2014. Vet J
2017;227:1–7.
11 Yim VWT, Yeung JHH, Mak PSK, etal. Five year analysis of Jockey
Club horse- related injuries presenting to a trauma centre in Hong
Kong. Injury 2007;38:98–103.
12 Bolwell C, Rogers C, Gee E. Descriptive epidemiology of race-
day jockey falls and injuries in New Zealand. Comp Exerc Physiol
2014;10:49–55.
13 Balendra G, Turner M, McCrory P. Career- ending injuries to
professional jockeys in British horse racing (1991-2005). Br J Sports
Med 2008;42.
14 O’Connor S, Warrington G, McGoldrick A. Epidemiology of injury due
to race- day jockey falls in professional at and jump horse racing in
Ireland, 2011 -2015. J Athl Train 2017.
15 Turner M, McCrory P, Halley W. Injuries in professional horse racing
in Great britain and the Republic of Ireland during 1992-2000. Br J
Sports Med 2002;36:403–9.
16 Jackson KA, Sanchez- Santos MT, MacKinnon AL, etal. Bone density
and body composition in newly licenced professional jockeys.
Osteoporos Int 2017;28:2675–82.
17 Bahr R, Clarsen B, Derman W. International Olympic Committee
consensus statement : methods for recording and reporting of
epidemiological data on injury and illness in sport 2020 (including
STROBE Extension for Sport Injury and Illness Surveillance
(STROBE- SIIS)). Br J Sports Med 2020;8:1–18.
18 Rae K, Orchard J. The orchard sports injury classication system
(OSICS) version 10. Clin J Sport Med 2007;17:201–4.
19 Von Elm E, Altman DG, Egger M. The strengthening the reporting
of observational studies in epidemiology (STROBE) statement:
guidelines for reporting observational studies. PLoS Med 2007.
20 Legg KA, Cochrane DJ, Bolwell CF, etal. Incidence and risk factors
for race- day jockey falls over fourteen years. J Sci Med Sport
2020;23:1154–60.
21 Hitchens PL, Morrice- West AV, Stevenson MA, etal. Meta- Analysis of
risk factors for racehorse catastrophic musculoskeletal injury in at
racing. Vet J 2019;245:29–40.
22 Mountjoy M, Sundgot- Borgen J, Burke L, etal. The IOC consensus
statement: beyond the Female Athlete Triad--Relative Energy
Deciency in Sport (RED- S). Br J Sports Med 2014;48:491–7.
23 Wilson G, Hill J, Martin D, etal. Gb apprentice jockeys do not have
the body composition to make current minimum race weights: is it
time to change the weights or change the jockeys? Int J Sport Nutr
Exerc Metab 2020:101–4.
24 Fuller CW, Molloy MG, Bagate C, etal. Consensus statement on
injury denitions and data collection procedures for studies of injuries
in rugby Union. Br J Sports Med 2007;41:328–31.
25 Bahr R, Clarsen B, Derman W, etal. International Olympic Committee
consensus statement: methods for recording and reporting of
epidemiological data on injury and illness in sport 2020 (including
STROBE extension for sport injury and illness surveillance (STROBE-
SIIS)). Br J Sports Med 2020;54:372–89.
26 Turner M, Balendra G, McCrory P. Payments to injured professional
jockeys in British horse racing (1996-2006). Br J Sports Med
2008;42:763–6.
27 Hitchens P, Blizzard L, Jones G. Are physiological attributes of
jockeys predictors of falls? A pilot study. BMJ Open 2014;1:1–7.
28 Hitchens PL, Blizzard CL, Jones G, etal. The incidence of race- day
jockey falls in Australia, 2002-2006. Med J Aust 2009;190:83–6.
29 O’Connor S, Warrington G, McGoldrick A, etal. Race day
concussion incidence in Irish professional at and jump horse racing
from 2011 to 2016. J Sci Med Sport 2017;20:20–1.