Key words osteoarthritis - osteoporosis - anxiety - depression - athlete - chronic disease
Introduction
The beneficial effects of exercise on chronic disease have been well documented, with
significantly reduced mortality due to cardiovascular disease, hypertension, diabetes
and cancer [1 ]
[2 ]. Regular exercise has also been shown to have positive effects on mental health
in the general population [3 ]
[4 ]; however, it is known that professional athletes may be subject to unique stresses,
such as when injured or underperforming, which may increase their predisposition to
mental health problems [5 ].
The prevalence of osteoarthritis (OA) in the general population ranges from 12.3–21.6%
with risk factors including older age, female sex, obesity, genetic predisposition,
dietary factors and joint-specific factors including injury, malalignment and abnormal
loading [6 ]
[7 ]. In some former professional sports participants, the prevalence of musculoskeletal
problems is higher compared to the general population, particularly OA of the lower
limb [8 ]
[9 ]
[10 ]. Sport injuries contribute to the development of OA and mental health problems in
retired elite athletes [8 ]
[11 ].
Horse-racing is a popular sport in the UK. There are currently over 400 licensed professional
jockeys, over 350 amateur jockeys and over 5,000 stable staff. McCrory et al. reported
a fall rate of 0.4% per ride for flat jockeys, 6.8% for jump jockeys and an injury
incidence per fall of 39.8% for flat jockeys and 18.0% for jump jockeys in Britain
between 1992 and 2001 [12 ]. A more recent Irish paper reported falls and injuries rates in Irish jump and flat
racing between 2011 and 2015 [13 ]. Fall rates were of 3.8 and 49.5 falls per 1 000 rides, and 352.8 and 203.8 injuries
per 1 000 falls for flat and jump jockeys, respectively.
It has been shown that exercise is important in obtaining peak bone mineral density
and reducing the risk of osteoporosis [14 ]; however, in weight-restricted sports there is risk of presenting poor bone health
[15 ]. There is one published paper investigating musculoskeletal health in 28 retired
Irish jockeys, which did not identify an increase in osteoporosis compared to a reference
group [16 ]. Participation in elite-level cricket and rugby has been shown to have a beneficial
effect on cardiovascular disease [17 ]
[18 ]. There is only one paper referring to chronic disease in retired male jockeys and
this paper was primarily researching physiological and health markers in 28 retired
Irish jockeys [19 ]. In this cohort 14% reported having high blood pressure, 14% heart disease, 7% stroke,
11% impaired lung function and 4% kidney removal. To our knowledge, there have been
no studies investigating chronic conditions in retired professional jockeys.
This study was designed to fill this gap in epidemiological research on this unique
athlete population, aiming to improve knowledge on health status in retired professional
jockeys. Thus, the overarching aims of this study were: 1) to describe the prevalence
of chronic disease, including cardiovascular, musculoskeletal conditions and mental
health problems in a cohort of retired male professional jockeys, and 2) to compare
those with an age-matched sample from the general population.
Materials and Methods
Study Design
A standardised core questionnaire was designed and developed by the Arthritis Research
UK Centre for Sport, Exercise and Osteoarthritis to investigate the health of retired
elite athletes. The questionnaire was comprised of a core set of questions including
injury, medical and playing history as previously published [17 ]
[18 ]. This questionnaire was further adapted for horseracing with the involvement of
current and retired jockeys and stakeholder organisations from racing including the
British Horseracing Authority (BHA), The Racing Foundation, British Racing School
(BRS), Professional Jockeys Association (PJA), National Association of Racing Staff
(NARS), Racehorse Owners Association (ROA), Racing Welfare (RW) Injured Jockeys Fund
(IJF), Jockeys Education and Training Scheme (JETS), Northern Racing College (NRC)
and National Trainers Federation (NTF). Two further public participation (PPI) groups
were then undertaken. The first group comprised of 4 retired and current jockeys and
4 researchers. The jockeys were asked to complete the questionnaire, any concerns,
queries and difficulties were then discussed and recorded. The research team revised
the questionnaire taking into account the feedback. The second PPI group, comprising
of 4 retired and current jockeys and 3 researchers, was then asked to review the revised
questionnaire. Feedback was discussed and recorded and the questionnaire finalised.
The questionnaire was available online and in paper form. The paper version was distributed
via mailouts from the PJA, BHA and JETS and IJF almoners. The online version was emailed
to any retired jockeys contacting the research team as a result of social medial notifications,
adverts on racecourses, adverts in industry newsletter and personal communications.
Data were collected and managed using REDCap (Research Electronic Data Capture) platform
[20 ], hosted at the University of Oxford. REDCap is a secure web-based application designed
to support data capture for research studies. Paper questionnaires were manually entered,
whilst REDCap’s survey tool was used for electronic questionnaire completion.
Ethics approval for The Retired Jockey Study was granted from the University of Oxford
(R4403/RE002). The Study meets the ethical standards of the International Journal
of Sports Medicine [21 ]. Data were anonymised and informed consent was obtained from all participants.
Participants
The Jockey Study
One overarching database of all retired professional jockeys does not exist within
the racing industry; however, the PJA estimates that at least 95% of all GB licensed
jockeys will have been members of the PJA during their racing career. The PJA ‘badge
holders’ are retired professional jockeys who have been GB licensed professional jump
jockeys for at least 10 seasons, or flat jockeys for at least 15 seasons. There were
230 PJA badge holders at the time of distribution of the questionnaires in 2016. This
database was used as the primary data source, and was supplemented by contacting retired
professional jockeys using additional databases, as described in the study design,
held by BHA, JETS and the IJF.
Comparison cohort
The English Longitudinal Study of Ageing (ELSA) was used as a reference population
representative of the general population [22 ]. The ELSA study is a prospective study of community-dwelling older people, which
included individuals who were living within the household at the time of the Health
Survey for England (HSE) interview. Wave 1 of ELSA was used in this study. Data were
collected in 2002–2003 and a core sample of 11 391 women and men aged 50 and over
were recruited. Variables from Wave 0 of ELSA were used when they were not collected
at Wave 1. Details of the measurement protocol can be found at http://www.ifs.org.uk/elsa
Male participants aged 50–89 years were included in the reference population ([Fig. 1 ]). Every retired professional male jockey was matched individually by age with five
male participants from the reference population (1:5).
Fig. 1 Flowchart of the study.
Outcome: Chronic conditions and mental disease
Outcomes included seven self-reported, GP-diagnosed chronic conditions; heart problems,
stroke, hypertension, diabetes, asthma, osteoporosis and OA and two mental health
problems; lifetime depression and anxiety. Details regarding the harmonisation process
of each health outcome have been described in detail previously [17 ]
[18 ] (more details on measurements are available from the authors on request).
Covariates
Body mass index (BMI) (weight in kilograms divided by height in metres squared) was
calculated from clinically measured weight and height in the reference population
and from self-reported measurements in the retired professional jockeys. Smoking status
was self-reported in both studies and categorised as ‘current smoker’ vs. ‘non-smoker’
(including ex-smokers).
Statistical analysis
All analyses were conducted using Stata Statistical Software: Release 15 (StataCorp,
College Station, Texas). Characteristics for jockeys and the reference population
were assessed using mean and standard deviation (SD) for continuous variables, and
relative and absolute values for categorical variables. Conditional logistic regression
was used to estimate the differences between jockeys and the reference population
adjusting for BMI and smoking status. Prevalence odds ratio (OR) and their 95% confidence
intervals (CI) were calculated, using the maximum number of individuals available
for each health outcome. The level of significance was set at p<0.05 for all statistical
analyses.
Results
Of the 1 464 questionnaires distributed online and by post as described in the study
design, 260 were returned ([Fig. 1 ]). After excluding duplicates, incomplete questionnaires and those completed by non-professional
or female jockeys, a total of 209 questionnaires were available for analysis. The
only response rate that could be calculated was for questionnaires distributed to
PJA badge holders, as this was the only up-to-date accurate database. The response
rate for this sub-group was 53% (122/230). A response rate for jockeys recruited from
organisations other than the PJA could not be derived, as the number of jockeys who
physically received the invitation to participate in the study cannot be determined
due to inaccurate contact details, duplicate contacts and deceased jockeys.
There were 209 male professional jockeys, who returned completed questionnaires. Of
these, 135 jockeys aged 50+were included in the age-matched analysis alongside 675
participants from the reference population ([Fig. 1 ]).
Demographic characteristics of all eligible retired professional jockeys, jockeys
aged 50+, and reference population are shown in [Table 1 ]. The average age of all eligible jockeys was 56.1±14.6 years (range 22–88 years)
and 93.3% identified their ethnicity as white. They had retired from racing for 30±11
years, and 71% had 11 seasons or more of riding at a professional level. Thirty-one
percent had a career in flat racing, 62% in jump racing and 7% in both jump and flat
racing. Ethnicity, BMI, smoking status and racing history between all eligible retired
professional jockeys and those aged 50+were similar ([Table 1 ]). The retired professional jockey population had statistically lower BMI than the
reference population ([Table 1 ]). The percentage of white ethnicity and current smokers did not differ between groups.
Table 1 Characteristics of the retired, professional jockeys and reference population.
Participant characteristics
All jockeys (n=209)
Jockeys age 50+(n=135)
Reference population (n=675)
p-value*
Age (years), mean (SD)
56.1 (14.6)
64.7 (9.9)
64.7 (9.9)
Ethnicity, n (%)
0.138
White
195 (93.3)
122 (90.4)
650 (96.3)
Mixed and Other
9 (4.3)
8 (5.9)
23 (3.4)
Missing
5 (2.4)
5 (3.7)
2 (0.3)
BMI (kg/m2 ), mean (SD)
24.5 (2.9)
25.0 (3.0)
27.4 (3.8)
<0.001
BMI (kg/m2 ), n (%)
<0.001
Under/Normal weight
125 (59.8)
73 (54.1)
157 (23.3)
Overweight/Obese
72 (34.5)
57 (42.2)
451 (66.8)
Missing
12 (5.7)
5 (3.7)
67 (9.9)
Smoking status, n (%)
0.113
Non-smoker
176 (84.2)
116 (85.9)
546 (80.9)
Current smoker
28 (13.4)
16 (11.9)
118 (17.5)
Missing
5 (2.4)
3 (2.2)
11 (1.6)
Racing history
Type of race
Flat
65 (31.1)
44 (32.6)
-
Jump
130 (62.2)
81 (60.0)
-
Both
14 (6.7)
10 (7.4)
-
Years since retirement from riding, mean (SD)
30.0 (11.3)
33.4 (10.2)
-
* Retired professional jockeys age 50+compared with the age-matched reference population;
BMI=body mass index; IQR=interquartile range.
In the group of all eligible jockeys, the most prevalent health outcome was OA (40.7%),
followed by hypertension (27.8%), asthma (14.8%) and depression (12.9%). The prevalence
of the remaining health outcomes was 12% or lower ([Fig. 2 ]).
Fig. 2 Prevalence of chronic diseases and mental health problems of all eligible retired
professional jockeys.
The prevalence of osteoporosis and OA were significantly higher in retired professional
jockeys aged 50+than in the reference population (7.4 vs. 1.6% and 46.7 vs. 14.5%,
respectively; p-value ≤0.001). Retired professional jockeys aged 50+also showed a
significantly greater lifetime prevalence of anxiety and depression compared with
the reference population (11.1, vs. 4.2% and 12.6 vs. 5.0%, respectively). No significant
differences were found for any other conditions analysed ([Table 2 ]).
Table 2 Prevalence and odds ratios (crude and adjusted) of chronic diseases and mental health
problems in retired professional jockeys and reference population.
Outcomes
Jockeys age 50+(n=135)
Reference population (n=675)
N
Crude OR (95% CI)
Adjusted OR* (95% CI)
Chronic condition
Heart problems, n (%)
23 (17.0)
159 (23.6)
729
0.68 (0.41–1.12)
0.81 (0.48–1.37)
Stroke, n (%)
7 (5.2)
30 (4.4)
728
1.35 (0.55–3.31)
1.20 (0.48–3.04)
Hypertension, n (%)
51 (37.8)
247 (36.6)
727
1.13 (0.75–1.69)
1.29 (0.84–1.97)
Diabetes, n (%)
7 (5.2)
57 (8.4)
728
0.62 (0.27–1.40)
1.07 (0.44–2.60)
Asthma, n (%)
20 (14.8)
70 (10.4)
727
1.54 (0.88–2.70)
1.72 (0.95–3.14)
Osteoporosis, n (%)
10 (7.4)
11 (1.6)
726
5.81 (2.19–15.45)
6.49 (2.05–20.50)
Osteoarthritis, n (%)
63 (46.7)
98 (14.5)
715
5.23 (3.39–8.08)
7.45 (4.55–12.21)
Mental health condition
Anxiety, n (%)
15 (11.1)
28 (4.2)
728
3.15 (1.55–6.39)
2.81 (1.34–5.89)
Depression, n (%)
17 (12.6)
34 (5.0)
727
2.79 (1.47–5.28)
2.60 (1.33–5.07)
*BMI and smoking status were included in the model; OR=odds ratio; CI=confidence interval;
BMI=Body Mass Index
After adjusting for BMI and smoking status, the OR for musculoskeletal and mental
health outcomes remained significant ([Table 2 ]). An increased odds of having osteoporosis (OR=6.5, 95%CI 2.1–20.5), OA (OR=7.5,
95%CI 4.6–12.2), anxiety (OR=2.8, 95%CI 1.3–5.9) and depression (OR=2.6, 95%CI 1.3–5.7)
was seen in the retired professional jockey population.
Discussion
This study is the first to report the prevalence of chronic diseases and mental health
problems in retired GB professional jockeys. Retired professional jockeys had a higher
prevalence of musculoskeletal conditions and mental health problems compared with
the reference population, which remained after adjusting for BMI and smoking.
This study has found that retired professional jockeys are at significantly increased
odds of having osteoporosis in retirement with a prevalence of 7.4% and OR of 6.5
(95%CI 2.1–20.5), after adjusting for BMI and smoking status. This result does not
reflect the results of a previous retired jockey study, which found no difference
in the bone mineral density (BMD) of 28 retired Irish jockeys compared to reference
ranges [16 ]. This may be due to differing retired jockey populations, as the previous study
had a narrow age range (50–70 years old); which is younger than this study population,
and osteoporosis is known to be an age-associated morbidity. Jockeys are at risk of
low energy consumption and poor weight management practices, which are thought to
impact their ability to reach peak BMD [23 ]
[24 ]
[25 ]. In current newly-licensed jockeys in GB, Jackson et al. found 76% of male flat
and 52% of male jump jockeys had low BMD (less than one SD below the average in the
age-sex-reference) [15 ]. This result provides supporting evidence that low BMD persists into retirement
(Table 1S ).
The prevalence of OA in retired professional jockeys over 50 years was 46.7% with
an adjusted OR of 7.5 (95%CI 4.6–12.2). Two previous studies also found a higher prevalence
of OA in retired professional rugby and cricket players compared to the general population,
with an increased risk of 4.0 and 3.6-fold, respectively [17 ]
[18 ]. These consistent findings are likely to be a consequence of increased injury rates
in former elite athletes. Injury has been reported as central to the development of
lower limb OA in sporting populations [6 ]
[26 ], and injury is known to be prevalent in jockeys, which may suggest post-traumatic
OA as the rationale for the higher prevalence reported in this study. Due to increased
availability of routine healthcare provision there may be increased reporting in these
populations in comparison with the general population (Table 2S ).
Our study also found a higher lifetime prevalence of self-reported, physician-diagnosed
anxiety and depression compared to the reference population. A recent study by Losty
et al. in 42 professional Irish jockeys has reported 57% exceeding the threshold for
depression and 21.4% for generalised anxiety disorder when completing an online validated
self-reported measurement tool [27 ]. There is great variability in the lifetime prevalence of depression and anxiety
in the general population due to diverse collection and reporting methods. King et
al., using a Composite International Diagnostic Interview of general practice attendees
in 6 European countries, reported previous 6-month anxiety prevalence to be 8.4%,
and major depression to be 12.7% [28 ]. Kessler et al. reported the lifetime prevalence of anxiety and depression disorders
as 28.8 and 16.6%, respectively [29 ] using a face-to-face household survey of English speaking household residents in
the USA. Compared to retired professional elite rugby and cricket cohorts, using the
same reporting method, the lifetime prevalence of anxiety and depression was 6% and
5% respectively in rugby [17 ] and 12.4 and 8.8% respectively in cricket [18 ]. Therefore, whilst retired jockeys had significantly higher rates of anxiety and
depression than the reference population, they appear to have similar rates to other
retired athlete populations (Table 3S ).
There are multiple possible contributing factors to the increased rates of anxiety
and depression. The transition out of professional sport when athletes retire is associated
with an increased risk of mental health problems including anxiety and depression
[30 ]. There is an association between repeated mild traumatic brain injury and depression,
which jockeys may have been exposed to during their riding career as a result of multiple
falls [31 ]. Jockeys have been found to report depression 46 times more frequently if they have
a current injury [27 ]. Racing is a weight-restricted sport and weight management practices may be associated
with a higher risk of mental health problems during a racing career [32 ]. Pain and depression are recognised to frequently co-exist, with studies reporting
increased rates of depression in patients with pain [33 ]. OA and career injuries may have increased pain in this population, in turn increasing
the prevalence of mental health outcomes.
No differences were found in the prevalence of cardiovascular disease in retired professional
jockeys compared to the reference population. Compared to previous studies of retired
jockeys our study demonstrates a higher prevalence of hypertension (37.8) compared
to the Cullen et al. study, with 14% of retired Irish jockeys reporting high blood
pressure, but this may be explained by the increased mean age in our study [19 ]. In other sports, Davies et al. [17 ] found a significantly reduced prevalence of diabetes in retired rugby players and
Jones et al. [18 ] found significantly reduced levels of heart problems in cricket players. This may
be an indication of different physiological demands during elite sports participation,
and the capacity of these demands to positively influence longer-term clinically relevant
health outcomes.
Strengths and limitations
To our knowledge, this study is the first to investigate the prevalence of chronic
diseases, such as cardiovascular and musculoskeletal diseases as well as lifetime
mental health problems, in a retired professional jockey population. Furthermore,
this study represents the largest, well-characterized cohort of retired professional
jockeys. There was a high response rate for the PJA badge holders making this generalizable
to more established jockeys with longer, more successful careers. However, this may
be less generalizable to jockeys licenced for less than 10 years jump racing or 15
years flat racing.
Some potential limitations of this study include the exclusion of retired female jockeys
so the findings may only be generalizable to retired professional male jockeys. There
is a potential selection bias from those jockeys with chronic health problems being
more likely to return a questionnaire regarding their health status. Alternatively,
there may be a selection bias from jockeys who want to report positive health outcomes
following a career in racing. As a cross-sectional study, it is not possible to investigate
the cause of the higher prevalence of musculoskeletal diseases and lifetime depression
and anxiety; therefore we cannot infer causality for these morbidities. Another limitation
of this study included the self-reported height and weight in the retired professional
jockey population compared to clinically measured BMI in the reference group. Previous
studies that examined the accuracy of self-reported height and weight compared to
measured values consistently reported that BMI tended to be underestimated [34 ]
[35 ]. A recent study based on a sample of Irish ex-jockeys aged 50–70 [16 ] found that the average of measured BMI was 26.7 kg/m2 , around 1.7 kg/m2 higher compared to our jockey population. Since the mean difference in BMI between
the Irish ex-jockeys population and our reference population was smaller and non-statistically
significant, we do believe that the impact of the self-reported BMI does not substantially
influence on our findings. Information on physical activity and on duration, intensity
or quantity of smoking was not available in both study populations. Those measures
are potential confounders, which may influence musculoskeletal and mental health problem
outcomes. Finally, whilst the data from ELSA Wave 1 is now 15 yrs old it is a comprehensive
and relevant database available for comparison. The diagnosis of OA, osteoporosis,
depression and anxiety has not changed significantly over the last 15 yrs so therefore
the prevalence data should remain comparable to the retired professional jockeys.
In addition, the survey questions used in ELSA were comparable to those used in our
study.
Conclusion
Reported musculoskeletal diseases and mental health problems were significantly more
prevalent in retired, male professional jockeys than in the reference population.
No differences in the prevalence of cardiovascular diseases were found. Additional
studies of female-only or mixed gender cohorts are needed to confirm the generalizability
of these findings for female jockeys. Further research of the risk factors for increased
musculoskeletal disease and mental health problems is required to identify potential
interventions for current and future professional jockeys.
Practical Implications
Osteoporosis is more common in retired professional jockeys than the general population.
Educating young jockeys and addressing modifiable risk factors throughout their careers
is important for their future bone health.
These results confirm that the high incidence of low bone mineral density in newly-licensed
jockeys is also present in the retired, professional jockey population.
Strategies to improve safety and prevent injuries at racecourses and in training yards
are required to reduce the risk of subsequent OA in both retired professional jockeys
and racing staff.
Mental health problems are prevalent in retired professional jockeys. Industry provision
of mental health support during jockeys’ careers, the transition into and during retirement
is important.
