ANZ J. Surg.

2003;

73

: 567–571

ORIGINAL ARTICLE

ORIGINAL ARTICLE

PATTERN OF EQUESTRIAN INJURIES PRESENTING TO A SYDNEY TEACHING HOSPITAL

J

AMES

L

IM

,* V

IKRAM

P

UTTASWAMY

,* M

ARCELLO

G

IZZI

,

†

L

INDA

C

HRISTIE

,

‡

W

ILLIAM

C

ROKER

†§

AND

P

HILIP

C

ROWE

*

§

*Department of Surgery and

†

Medicine, Prince of Wales Hospital,

‡

Sydney Children’s Hospital and

§

University of New South Wales, Randwick, New South Wales, Australia

Background

:

Equestrian activities are associated with a high rate of injury. Attempts to reduce the incidence and severity of injuryrequire accurate characterization of risk factors and pattern of injury. The present study was performed to analyse the injuries seen atan Australian centre where a large number of equestrian injuries present.

Methods

:

Data were collected prospectively over a 3 year period on all equestrian injuries presenting to the Prince of WalesHospital complex. These data were compared against and combined with retrospectively collected data over the preceding 3 years atthe same centre.

Results

:

Two hundred and twenty-one injured equestrians presented (181 consented) in the prospective period of the study, and208 presented in the retrospective period. Overall, 81% of riders were wearing a helmet at the time of their injury. Helmet use wasassociated with a significantly lower admission rate (27%

vs

55%;

P

< 0.0001, from combined data). Recreational equestrians had ahigher admission rate than professional equestrians, and had a significantly higher head and spine injury rate than the professionalgroup. Rate of helmet use increased from 72% in the retrospective group to 91% in the more recent prospective group, and totaladmissions decreased from 43% to 14%.

Conclusion

:

Significant and serious injuries are associated with equestrian activities, with the higher risk group being recreationalequestrians, and riders not using a helmet. The pattern of injury favours head and spine in recreational and non-helmeted riders, andextremities for professional and helmeted riders. Helmet use is still not universal among riders, although an increase in its use may becontributing to an overall reduction in admission rate. Facial and spinal injuries still occurred in helmeted patients.

Key words: equestrian, helmet, horse riding, injury pattern, paediatric, professional, recreational.

Abbreviations

: AJC, Australian Jockey Club; PCA, Pony Club Association; NEISS, National Electronic Injury SurveillanceSystem.

INTRODUCTION

Horse riding is a popular sport in Australia but it carries a sig-nificant risk of injury. The incidence of injury has been esti-mated at 1 per 350 riding hours

1

to 1 per 1000 riding hours.

2

Ofthese injured patients 15–27% will have an injury severeenough to require an admission.

3–5

Equestrians have been esti-mated to be at higher risk of serious injury than participants ofmany other sports,

6

the incidence of serious injury per hour ofriding being greater than during motorcycle and car racing.

7,8

One South Australian study has estimated the death rate attrib-utable to equestrian activities to be 1 per 1 million populationper year in Australia.

9

Many studies have attempted to characterize the pattern ofinjury and risk factors for serious equestrian injuries, in order toprevent or reduce the risk of such injuries. Almost all of theavailable literature comes from European and American data,with most of the Australian figures coming from South Austral-ian data.

5,9

The purpose of the present study was to analyse

equestrian injuries presenting to the Randwick Hospitals com-plex, which is in close proximity to a number of centres ofequestrian activities.

METHOD

Two data sets were examined in the present study. One set wascollected prospectively over a 3 year period from March 1994to March 1997. All patients who presented to the emergencydepartments of the Prince of Wales and the Prince of WalesChildren’s Hospitals with a horse-related injury over this periodwere invited to participate. Written consent was obtained fromthe patient or relatives to provide information on their injury andparticipate in a post-admission follow-up survey. At presenta-tion, information was collected about the rider, the horse, the cir-cumstances of the injury and any safety equipment used. Afterdischarge, final details of body parts injured, treatments receivedand the calculated trauma severity scores (modified trauma score,injury severity score, Glasgow trauma score) were recorded by aclinical research scientist. Finally a survey form was mailed outor a telephone conversation with the research scientist was madeto record level of functional recovery at 1 and 6 months after theinjury.

The second data set was collected retrospectively over a 3 yearperiod from 1991 to 1993. The same information as indicatedwas collected with the exception of post-discharge recovery data.The data was analysed using

χ

2

test and Student’s

t

-test.

J. Lim

FRACS;

V. Puttaswamy

FRACS;

M. Gizzi

BSc;

L. Christie

RN;

W. Croker

FRACP;

P. Crowe

FRACS.

Correspondence: Professor P. J. Crowe, Department of Surgery, Prince ofWales Hospital, Randwick, NSW 2031, Australia.Email: p.crowe@unsw.edu.au

Accepted for publication 8 April 2003.

568 LIM

ET AL

.

The patients were grouped into professionals if they wereworking or being employed to work on the horse at the time ofinjury, which included jockey, trainer, track worker, stablehand, or horse handler. Recreationals were pleasure or trailriders, or handlers of their own or a hired horse. Paediatricpatients were aged <16 years, adults were

≥

16 years. Injuredbody parts were grouped into head or spine (including face andneck); trunk (chest, clavicle, abdomen and pelvis); and extremi-ties (shoulder, hip, upper and lower limbs).

RESULTS

A total of 429 patients sustained injuries during the studyperiod. There were 221 equestrian injuries during the 3 years ofthe prospective study (Table 1). During this prospective phase,181 patients consented to the study, of whom 44 (24%) werepaediatric. Of the 137 adults, 95 (69%) were involved in a pro-fessional equestrian activity. Ninety-one per cent of 141 riders,where helmet status was known, were helmeted at the time ofinjury. A total of 14% of the 181 patients required an admission,and there were no deaths in this group.

Of the 208 patients in the retrospective group, 75% wereadults, of whom the majority (76%) were professional. Seventy-two per cent of 158 riders, where helmet status was known,were helmeted at the time of injury. Forty-three per cent of the208 patients required an admission, and there were two deaths,both women.

Professional versus recreational

When data from both the prospective and the retrospectivestudies were pooled, there were 219 equestrians involved in pro-fessional activities and 117 involved in recreational activities atthe time of their injury (Table 2). Male equestrians outnumberedfemale equestrians in the professional group, whereas femaleequestrians outnumbered male equestrians in the recreationalgroup. Professionals were more likely to wear a helmet and lesslikely to be admitted to hospital compared to recreational riders.One professional, a 31-year-old female helmeted rider, died fromsevere brain injury following a fall, and one recreational rider,also a helmeted 31-year-old woman, died from intracerebralhaemorrhage and severe head injuries following a fall.

Injuries sustained by professionals were more likely to beextremity injuries, compared with recreational riders, who had amore significant rate of head and spine injury (Fig. 1).

Disability, defined as the inability to return to previous workor perform previous house duties, either partially or totally,was assessed at 1 and 6 months following injury in the pro-spective group of patients. At 1 month, 13% of professionaland 19% of recreational equestrians were still disabled, and at6 months 1% of professional and 2% of recreational equestri-ans were still disabled.

Table 2.

Comparison of professional

vs

recreational injuries

Professional Recreational Retrospective Prospective Combined Retrospective Prospective Combined

Total (

n

) 118 101 219 37 80 117M:F 1.3:1.0 1.0:3.0Helmeted riders:

n

(%) 69 (84) 69 (96) 138 (89)* 9 (29) 60 (87) 69 (69)*Admissions (%) 31 11 22* 100 19 44*Deaths 1 0 1 1 0 1

*

P

< 0.0001 for combined group.

Table 1.

Descriptive data from prospective and retrospective groups

Prospective group Retrospective group

Total (

n

) 221 208

Consented (

n

) 181 –Average age

(range; years)25 (0.2–76) 26 (3–46)

M:F 0.74:1.0 (77:104) 0.79:1.0 (91:117)Paediatric:adult 44:137 53:155Professional:

recreational (adults)

95:42 118:37

Riders 144 179Helmeted riders

% (

n

)91 (129 of 141;

3 unknown)72 (114 of 158; 21 unknown)

Admissions % (

n

) 14 (26 of 181) 43 (90 of 208)Deaths 0 2

Fig. 1.

Pattern of injury in (

�

) professional (

n

= 213)

vs

(

�

) recrea-tional (

n

= 79) equestrians (*

P

< 0.01).

EQUESTRIAN INJURIES 569

Paediatric versus adult

Young female equestrians outnumbered male equestrians in thepaediatric group by almost 3:1, whereas adults had a near equalproportion of male to female (Table 3). In the prospectivegroup, which was the more recent data set, helmet use waslower among paediatric riders (86%) than in adult riders (92%),and admission rates were higher (18%

vs

13%), although thedifferences were not statistically significant. There were nopaediatric deaths.

The spectrum of body parts injured showed that extremitiesare the most common body parts injured, followed by head orspine, then trunkal regions (Fig. 2). Children had a higher pro-

portion of head or spine injuries than adults (

P

= 0.07), andadults had a slightly higher incidence of extremity injuries thanchildren (

P

= 0.1).

Helmets

Pooled data revealed that overall, 81% of riders were wearing ahelmet at the time of their injury (Table 4). There was a signifi-cantly higher proportion of riders requiring an admission if theywere not wearing a helmet at the time of injury, although thisdifference was not demonstrated in the prospective group alone.

Patients who sustained a head injury and had been wearing ahelmet, had a higher rate of concomitant facial injury (35%),

Fig. 2.

Pattern of injury in (

�

) paediatric (

n

= 97)

vs

(

�

) adult(

n

= 292) equestrians.

Table 3.

Comparison of paediatric

vs

adult injuries

Paediatric Adult Retrospective Prospective Combined Retrospective Prospective Combined

Total (

n

) 53 44 97 155 137 292M:F 1.0:2.8 1.0:3.0 1.0:2.9 1.0:1.0 0.9:1.0 0.96:1.0Helmeted riders

n

(%) 36 (97) 32 (86) 68 (92) 78 (70) 97 (92) 175 (81)Admissions (%) 28 18 24 48 13 32Deaths 0 0 0 2 0 0

Fig. 3.

Pattern of injury in (

�

) helmeted (

n

= 243)

vs

(

�

) non-helmeted (

n

= 56) equestrians (*

P

< 0.01).

Table 4.

Comparison of helmeted and non-helmeted injuries

Helmeted Non-helmeted Retrospective Prospective Combined Retrospective Prospective Combined

Total riders (

n

) 114 of 158 129 of 141 243 of 299 44 of 158 12 of 141 56 of 299Proportion of riders (%) 70 91 81 23 9 19Admissions (%) 39 16 27* 66 17 55*Head injured

n

(%) 35 (31) 37 (29)** 72 (30) 10 (22) 7 (58)** 17 (30)Proportion of head injured

involving face

n

(%) 25 (35) 2 (12)Neck or spine injured

n

(%) 25 (22) 23 (18) 48 (20) 9 (20) 1 (8) 10 (18)Deaths 2 0 2 0 0 0

*

P

< 0.0001 for overall group; **

P

< 0.05 for prospective group.

570 LIM

ET AL

.

compared to non-helmeted patients (12%). This suggests thathelmet use did not protect against facial injuries.

In addition, a significant number of neck and spinal injuriesoccurred in helmeted patients (Fig. 3). Thirteen of 15 neck soft-tissue injuries, 10 of 19 fractures of the vertebral column, andboth the spinal cord injuries (one with a Brown Sequard syn-drome associated with a C spine fracture, and the other a para-plegia associated with a thoracic spine fracture), occurred inhelmeted patients.

DISCUSSION

The Prince of Wales Hospital campus is located in the vicinityof a number of popular equestrian centres. The Royal RandwickRacecourse, home of the Australian Jockey Club (AJC), is oneof the largest and busiest training and horse racing facility inAustralia. It is located 1 km from the hospital. Centennial Park,where equestrian activities of a mainly recreational nature takeplace, is within 2 km of the hospital. The Royal AgriculturalSociety showground site where the Pony Club Association ofAustralia (PCA) has been based is within 3 km of the hospital.The hospital is thus well placed to receive the full spectrum ofhorse-related injuries.

Equestrian activities encompass a wide range of both profes-sional and recreational activities, and are enjoyed by a wide agerange. Horse racing, harness racing, dressage, trail riding, trackwork training, pony club riding and simple pleasure riding arecommon riding pursuits, but opportunities for injury also arisein many non-riding activities such as training, grooming,handling, saddling, shoeing and feeding the horse.

Horse riding has several potential risk factors. The horse canweigh 500 kg, move at speeds of 65 km per h, elevate the riderto 3 m above ground, and have the potential to act unpredictablyand independently of the rider at any time.

6,10

The injury rate is understandably high. In America, theCenter for Disease Control estimated the risk at 18.7 injuries per100 000 equestrian-related events

7

or between 1 in 350 and 1 in1000 riding hours.

1,2

In Canada, one study calculated the seriousinjury rate requiring admission to be 0.49/1000 h of riding,which was comparable to calculated total injury rates (includingadmitted and non-admitted patients) for motorcycle riding(0.14/1000 h) and skiing (0.6/1000 h).

10

The potential forserious injury associated with riding or handling horses hasbeen expressed by an old Arab proverb, ‘the grave yawns forthe horseman’.

9

Despite a long history of horse riding, it is only since themid-20th century that serious measures have been taken toprevent injury. Standards for helmets in motor racing weredeveloped after the end of World War II, first in Britain then theUSA. The first standards for equestrian helmets were publishedin Britain in 1963.

6

The injured population in the present study was relativelyyoung, a finding similar to other studies. Twenty-five per centof patients were under 16 years of age, identical to CanadianMinistry of Health data.

10

In the USA from National ElectronicInjury Surveillance System (NEISS) data, 55% of injuredpatients were <25 years of age.

11

Injured female patients out-numbered male patients, which likely reflects the higher propor-tion of female equestrians who participate in equestrianactivities. Female preponderance was universal in studies fromthe USA, UK and Canada.

3,4,10,11

There was a large proportionof professionals relative to recreational adult equestrians in the

present series, reflecting the proximity of this hospital to theracecourse. All paediatric patients were recreational equestrians,apart from four 15-year-olds who were involved in track work.

The overall admission rate was 30% when data from the twoseries were pooled. This figure is similar to that published in thestate of Victoria of 27%

5

and comparable to American andBritish figures of 15–27%.

3,4

A paediatric admission rate of24% compared with 32% for adults is comparable to Victorianfigures of 34% and 20%, respectively.

12

Recreational equestri-ans were significantly more likely to require an admission thanprofessional equestrians, suggesting a greater risk of seriousinjury for recreational equestrians. Non-helmeted riders werealso significantly more likely to require an admission than hel-meted riders.

Professional equestrians on the whole were more likely to bemale, to be wearing a helmet while riding, to injure an extrem-ity, and to be less likely to require an admission. Recreationalequestrians were more likely to be female, less likely to bewearing a helmet, more likely to have a head or spine injury,and more likely to require an admission. This pattern of mainlylimb injuries in jockeys and face/head injuries in amateur ridershas previously been reported in a British study.

13

One-third of the present injured patients was under 16 yearsof age. The significance of injured young equestrians has beenhighlighted previously by NEISS data showing that more thanhalf (54%) of all equestrian injuries occurred in personsyounger than 25, with head injuries causing most (57%) of thedeaths.

11

The use of helmets, however, in children under15 years has been reported to be associated with less frequenthead injury and lower modified injury severity score (MISS).

14

More experienced riders are likely to be better at breaking theirfall with an arm or leg and possibly protecting their heads thanless experienced younger riders.

There is a possible problem in data collection in the retro-spective group because in the subgroup of recreational adultequestrians, 37 of 37 required admission (when compared to 15of 80 in the prospective group). This raises the possibility ofeither a confounding factor favouring admission of an injuredrecreational equestrian during the retrospective study period,where every adult recreational equestrian was admitted, or,more likely, there was a deficiency in flagging or recording aninjury as equestrian if the injury was trivial and not professionalor work related, during the period of the retrospective study.Thus the total number of patients sustaining a recreationalhorse-related injury in the retrospective part of the study may bean underestimate. In addition, there were 40 patients who didnot consent to the study but we are unable to say if this groupwould have altered results had they consented. Finally, longerfollow up of disability would have been useful to see if thosewho returned to work were able to stay at work at the same levelor had to later stop work.

Overall, 81% of riders were wearing a helmet at the time ofinjury. Although it is of concern to see that 19% of riders werenot wearing a helmet when injured, it is also encouraging to seethat this number decreased from 28% in 1990–1993 to 9% in1994–1997. This may reflect an increasing compliance with theAJC and PCA requirements for the use of safety equipment whileriding. Interestingly, the admission rate also fell over this period,from 43% to 14%, and although it is unlikely that the increasedrate of helmet use is solely responsible, it is likely to have con-tributed. It is noteworthy however, that a significant proportion offacial and spinal injuries did occur in helmeted patients, suggest-

EQUESTRIAN INJURIES 571

ing that helmets, although useful for the head and skull, did notafford protection against spinal or facial injuries.

In summary, the present study demonstrates that significantand serious injuries can be associated with equestrian activities,with the higher-risk group being recreational riders, particularlyriders not using a helmet. Although helmet use is associatedwith a reduced risk of admission, it is still associated with facialand spinal injuries. Paediatric patients did not necessarily suffermore severe injuries.

CONCLUSION

Equestrian activities are justifiably considered a dangeroussport. Recreational equestrians and riders not wearing a helmetare likely to suffer a more serious injury. The pattern of bodyparts injured appears to favour head and spine in recreationaland non-helmeted riders, compared with extremities for profes-sionals and helmeted riders. There are still a significant numberof non-helmeted riders despite AJC and PCA recommendations,although this number appeared to be decreasing over time.

REFERENCES

1. Gierup J, Larsson M, Lennquist S. Incidence and nature ofhorse-riding injuries.

Acta Chir. Scand.

1976;

142

: 57–61.2. Danielsson LG, Westlin NE. Riding accidents.

Acta Orthop.Scand.

1973;

44

: 597–603.3. Hobbs GD, Yealy DM, Rivas J. Equestrian injuries: A five year

review.

J. Emerg. Med.

1994;

12

: 143–5.

4. Lloyd RG. Riding and other equestrian injuries: Considerableseverity.

Br. J. Sports Med.

1987;

21

: 22–4.5. Williams F, Ashby K. Horse related injuries. Hazard, 1995,

June, No. 23. Victorian Injury Surveillance System. MonashUniversity Accident Research Centre. http: //www.general.monash.edu.au/muarc/hazard/haz23.pdf

6. Watt GM, Finch CF. Preventing equestrian injuries: Locking thestable door. Sports Med. 1996; 22: 187–97.

7. Centres for Disease Control. Injuries associated with horsebackriding: United States 1987–1988. Morbidity and Mortality WeeklyReport 1990; 39: 329–32.

8. Firth JL. Equestrian injuries. In: Schneider RC, Kennedy JC,Plant ML (eds) Sports Injuries: Mechanism, Prevention, andTreatment. Baltimore: Williams & Wilkins, 1985; 431–9.

9. Pounder DJ. The grave yawns for the horseman: Equestrian deathsin South Australia 1973–1983. Med. J. Aust. 1984; 141: 632–5.

10. Sorli JM. Equestrian injuries: A five year review of hospital admis-sions in British Columbia, Canada. Injury Prev. 2000; 6: 59–61.

11. Bixby-Hammett DM. Paediatric equestrian injuries. Paediatrics1992; 89: 1173–6.

12. Finch CF, Ozanne-Smith J, Williams F. The Feasibility ofImproved Data Collection Methodologies for Sports Injury,Report no. 69. Melbourne: Monash University AccidentResearch Centre, 1995.

13. Whitlock MR, Whitlock J, Johnston B. Equestrian injuries. Acomparison of professional and amateur injuries in Berkshire.Br. J. Sports Med. 1987; 21: 25–6.

14. Bond GR, Christoff RA, Rodgers BM. Paediatric equestrianinjuries: Assessing the impact of helmet use. Paediatrics 1995;95: 487–9.