Journal of Science and Medicine in Sport (2006) 9, 498—505

ORIGINAL PAPER

Sport- or leisure-related injury hospitaladmissions: Do we need to get moreout of being struck?

Rebecca Mitchell ∗, Andrew Hayen

NSW Injury Risk Management Research Centre, University of New South Wales,Sydney, NSW 2052, Australia

Received 11 January 2006; received in revised form 28 April 2006; accepted 1 May 2006

KEYWORDSSport or leisure injuries;Hospitalised injury;Injury classification;Struck by/struck against

Summary The usefulness of New South Wales (NSW) hospitalisation data for theidentification of prevention measures for sport- or leisure-related injury hospital-isations for one common injury mechanism, struck by/struck against injuries, isillustrated.

Sport- or leisure-related hospitalisations were identified during 1999—2000 to2003—2004 from the NSW hospitalisation data using activity and place of occur-rence information. Struck by/struck against injury hospitalisations were identifiedusing the International Classification of Disease, 10th Revision, Australian Modified(ICD-10-AM) codes W20—W23 and W50—W52.

Information regarding the number of hospitalisations for basic demographicdescriptors (such as age and sex), the type of injury experienced, the injury mech-anism, the activity, and the place of occurrence of the injury event are availablefrom NSW hospitalisation data.

Additional information than what is currently available would be required for theidentification of targeted injury prevention strategies for sport- or leisure-relatedstruck by/struck against injuries leading to hospitalisation. Assessing the feasibilityof collecting information regarding the object or agent of injury, the phase of activity

at the time of the injury, the collection of narrative text and the date of injury are allrecommended. These recommendations have national and international implicationsas ICD-10 is widely used to classify hospitalised morbidity data.© 2006 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

∗ Corresponding author. Tel.: +61 2 9385 4837;fax: +61 2 9385 6040.

E-mail address: r.mitchell@unsw.edu.au (R. Mitchell).

B

Io

1440-2440/$ — see front matter © 2006 Sports Medicine Australia. Published bdoi:10.1016/j.jsams.2006.05.005

ackground

n Australia, information regarding the magnitudef sport- or leisure-related injuries can be obtained

y Elsevier Ltd. All rights reserved.

G

fsessm

o1Hot(pttotiaoteeau(b

tvicosioobaad

N

Tsapthdfwt

ebci

cmnvihraf

Is

Sw2r

•

•

•

•

csetbiwca

C

D

etting more out of being struck

rom a number of sources, including hospital admis-ions, sports medicine clinics, general practition-rs, insurance records, surveys, and local sports-pecific data collections.1 Many of these dataources use different taxonomies to collect infor-ation on sport- or leisure-related injuries.Internationally, the International Classification

f Disease, 9th (ICD-9)2 or 10th Revision (ICD-0),3 is often used to classify hospital admissions.owever, detailed information regarding the typef sport or leisure activity is not available fromhese classification systems. In New South WalesNSW) and throughout Australia, injury-related hos-ital admissions are classified using the Interna-ional Classification of Disease, 10th Revision, Aus-ralian Modified (ICD-10-AM).4—6 Within chapter XIXf the ICD-10-AM on Injury, poisoning and cer-ain other consequences of external causes certainnjury types are identified by three character codes,nd a fourth and fifth character identified the placef occurrence and the activity of the individual athe time of the incident, respectively, in the firstdition4 and in supplementary variables in laterditions.5—7 Since 1998, sport- and leisure-relatedctivities have been able to be recorded, but notntil more recent editions have particular sportse.g. soccer, cricket, Australian rules) been able toe reported.

Unfortunately, information obtained from hospi-alisation data is not always ideal for injury pre-ention purposes since it often lacks specificity ors absent.8 Evaluation of data quality from hospitalollections has been published elsewhere,9 but usef hospitalised data to identify injury preventiontrategies has not been examined in NSW follow-ng the implementation of ICD-10-AM. Using onef the most common injury mechanisms for sport-r leisure-related hospitalisation in NSW, strucky/struck against injuries, (A. Hayen unpublishednalysis) an illustration of the type of informationvailable at a state level upon which to base theevelopment of prevention strategies is provided.

SW inpatient statistics collection (ISC)

he NSW Inpatient Statistics Collection is a cen-us of all services for admitted patients to publicnd private hospitals, private day procedures, andublic psychiatric hospitals in NSW.10 Hospitalisa-ion data include information on episodes of care inospital, which end with the discharge, transfer, or

eath of the patient, or when the service categoryor the admitted patient changes. Also includedithin the ISC are hospitalisations of NSW residents

hat occurred in another state or territory. How-

ctii

499

ver, these data were not available for 2003—2004,ut were estimated to include 233 or 3% of allases of hospitalisation for struck by/struck againstnjuries based on the average of the past 4 years.

Hospitalisations relating to transfers or statisti-al discharges were excluded using the separationode variable in order to attempt to partly elimi-

ate ‘multiple counts’, which occur when an indi-idual has more than one episode of care for a givennjury. These exclusions refer to transfers betweenospitals or changes in the on-going clinical careequirements (e.g. from acute to rehabilitation) forpatient during the one period of stay in a single

acility, respectively.10

dentification of sport- or leisure-relatedtruck by/struck against injuries

port- or leisure-related injuries of NSW residentsere identified in the NSW ISC for the 1999—2000 to003—2004 financial years using the following crite-ia:

the hospitalisation was for a patient who was aresident of NSW;a principal diagnosis was in the ICD-10-AM rangeS00-T98; andan activity code in the fifth character subdivi-sion in ICD-10-AM version 1 was equal to 0 or 1 orthe activity code in ICD-10-AM versions 2 and 3indicated the activity of the injured person wassport- or leisure-related (i.e., U50—U72); ora location code in the fourth character subdivi-sion in ICD-10-AM version 1 was equal to 3 or thelocation code in ICD-10-AM versions 2 and 3 wasequal to ‘sports and athletic areas’ (i.e., Y92.3).

Struck by/struck against injuries refer to theontact made between one person and another per-on(s) or object(s) as the result of an unintentionalvent.3 For example, struck by injuries can refer tohe type of unintentional contact that may occuretween players participating in a sporting activ-ty. Struck by/struck against injury hospitalisationsere identified using the ICD-10-AM external causeodes in the range W20—W23 for struck by injuriesnd W50—W52 for struck against injuries.

hanges to ICD-10-AM

uring the period under study there were three

hanges to the coding frames used by ICD-10-AMo classify both the activity and location of thenjurious incident, largely in the form of creat-ng additional codes. As a result, different cod-

B

Ttpi(jii2sptlu

rapb(tstht

I

Fittbha

500

ing frames were used to classify the activity andlocation variables during the three time periodsof 1999—2000,4 2000—2001 to 2001—2002,5 and2002—2003 to 2003—2004.6 For consistency, placeof occurrence was classified using ICD-10-AM ver-sion one.4 Activity at time of incident was onlyanalysed for the years 2002—2003 to 2003—2004using ICD-10-AM version 3.6 As appropriate, injurytypes were grouped by bodily region.

Data analysis

Analysis was performed using SAS.11 Age- and sex-specific population estimates as at 31 December ofeach of the years under study were obtained fromthe NSW Health Department. These estimates arebased on the Australian Bureau of Statistics (ABS)population estimates as at 30 June.12 Directly agestandardised rates were calculated using the esti-mated Australian residential population as at 30June 2001 as the standard population. Ninety-fiveper cent confidence intervals (95% CI) were calcu-lated assuming a Poisson distribution.13 To examinethe association between age group and body loca-tion of injury, a chi-squared test of independencewas used.14 Differences between age-standardisedrates for males and females were assessed for sig-nificance using the method described in Armitageet al.14

Results and discussion

While information needs from hospitalised injurydata can vary depending on who requires the infor-mation and for what purpose, all injury profession-

als seek information of high quality and of enoughdetail to allow the identification of injury preven-tion priorities and/or to assist in evaluation of theeffectiveness of injury prevention programs.

boti

Table 1 Sport- or leisure-related struck by/struck against95% CI, 1999—2000 to 2003—2004

Year Persons Males

N Ratea 95% CIb N

1999—2000 1469 22.8 21.7—24.0 12902000—2001 1287 19.9 18.8—21.0 11252001—2002 1396 21.4 20.3—22.5 12022002—2003 2063 31.2 29.8—32.5 17412003—2004 1966 29.8 28.5—31.2 1671

Total 8181 25.1 24.5—25.6 7029a Age-standardised rate per 100,000 population.b 95% confidence interval.

R. Mitchell, A. Hayen

asic demographics

he type of information that can be obtained fromhe NSW ISC to assist with the identification ofrevention measures for sport- or leisure-relatednjuries, includes basic demographic informationsuch as age and sex), which can be used in con-unction with demographic data to calculate injuryncidence rates to illustrate the magnitude of thessue. For example, overall during 1999—2000 to003—2004, there were 35,416 hospitalisations fortruck by/struck against injuries, at a rate of 107.8er 100,000 population (95% CI 106.6—108.9). Ofhese, 8181 (23.1%) were recorded as sport- oreisure-related, at a rate of 25.1 per 100,000 pop-lation (95% CI 24.5—25.6).

Males had a significantly higher hospitalisationate of sport- or leisure-related struck by/struckgainst injuries than females (�2 = 4221.9, d.f. = 1,< 0.0001) with over three-quarters of the strucky/struck against hospitalised injuries of males85.9%) (Table 1). Males aged 15—19 years hadhe highest age adjusted hospitalisation rates fortruck by/struck against injuries during 1999—2000o 2003—2004. For females, girls aged 10—14 yearsad the highest hospitalisation rates during thisimeframe (Fig. 1).

njury mechanism and type of injury

undamental information regarding the type ofnjury mechanism and information regarding theype of injury experienced can be obtained fromhe NSW ISC. For sport- and leisure-related strucky/struck against injuries these included: beingit, struck, kicked, twisted, bitten or scratched bynother person (33.9%), striking against or struck

y sports equipment (29.4%), and striking againstr bumped into by another person (24.9%) werehe most common types of struck by/struck againstnjuries (Table 2). Head injuries (38.3%), wrist and

injury hospitalisations by year, NSW, number, rate and

Females

Ratea 95% CIb N Ratea 95% CIb

39.7 37.5—41.9 179 5.6 4.8—6.534.3 32.3—36.3 162 5.1 4.3—5.936.4 34.4—38.6 194 5.9 5.1—6.851.9 49.4—54.4 322 9.8 8.8—11.050.1 47.7—52.5 295 9.0 8.0—10.1

42.5 41.5—43.5 1152 7.1 6.7—7.5

Getting more out of being struck 501

F stru1

hlcbipa

bmtrl

ia

cmiass

igure 1 Age adjusted rates for sport- or leisure-related999—2000 to 2003—2004.

and injuries (19.6%), and injuries to the knee andower leg (15.8%) were the most common prin-ipal diagnoses following admission for a strucky/struck against injury (Fig. 2). Injury type var-ed significantly by age group (�2 = 458.5, d.f. = 90,< 0.001), with head injuries common in individualsged less than 24 years.

However, for sport- and leisure-related strucky/struck against injuries, the lack of specific infor-

ation regarding, for example, the type of object

hat struck the person, or the circumstances sur-ounding or contributing to each injury mechanism,imits injury prevention activities aimed at reduc-

fitth

Table 2 Mechanism of sport- or leisure-related struck by/sper cent, 1999—2000 to 2003—2004

Mechanism

Struck by thrown, projected or falling object (W20)

Striking against or struck by sports equipment (W21)Striking against or struck by bat and racquet (W21.0)Striking against or struck by ball (W21.1)Striking against or struck by object or structure on or neaStriking against or struck by other sports equipment (W21Striking against or struck by unspecified sports equipment

Striking against or struck by other objects (W22)Caught, crushed, jammed or pinched in or between objectsHit, struck, kicked, twisted, bitten or scratched by anotherStriking against or bumped into by another person (W51)Crushed, pushed or stepped on by crowd or human stamped

Total

ck by/struck against injury hospitalisations by sex, NSW,

ng the incidence of injuries due to struck by/struckgainst injuries.

There are several ways in which improvementsould be made, such as by expanding the infor-ation that is currently recorded in the existing

npatient data collection on injuries. One option tochieve this would be an assessment of the fea-ibility of collecting relevant additional variables,uch as those specified in the International Classi-

cation of External Causes of Injuries (ICECI), par-icularly regarding the type of object that struckhe person.15 The ICECI is a multi-axial, modular,ierarchical system for classifying external causes

truck against injury hospitalisations, NSW, number and

Number Per cent

182 2.2

2408 29.4117 4.9758 31.5

r sports area (W21.2) 28 1.2.8) 243 10.1(W21.9) 53 2.2

487 6.0(W23) 197 2.4person (W50) 2775 33.9

2036 24.9e (W52) 96 1.2

8181 100.0

502 R. Mitchell, A. Hayen

isure

plunodplnpsptmmti

S

AiehD

Figure 2 Principal body location of injury of sport- or leNSW, per cent, 1999—2000 to 2003—2004.

of injuries. It was developed by an internationalcollaboration of injury experts to complement theInternational Classification of Diseases (ICD) codingof injuries and enable further detail on the circum-stances of the incident to be recorded, includinginformation on the risk and (lack of) protective fac-tors that led or contributed to the injury eventoccurring.

Activity and location of the injury event

Some information regarding the activity and loca-tion of the injury event is available in the NSWISC. For sport- and leisure-related struck by/struckagainst injuries, common activities reported to beundertaken at the time of the incident were teamball sports (58.3%) and team bat or stick sports(11.7%) (Table 3). Sports and athletic areas (82.2%),schools, other institution and public administra-tive areas (2.6%), and the home (1.3%) were themost common specified locations where the injuryoccurred. Unspecified locations accounted for only9% of the places where the struck by/struck againstinjury occurred (Table 4).

In recent editions of ICD-10-AM, there has been

particular attention to the addition of activitycodes to indicate different types of sport- orleisure-related activities that were performed atthe time of the injurious incident. For example, at

tmpm

-related struck by/struck injury hospitalisations by sex,

resent there are around 260 types of sport- andeisure-related activities (such as, cricket, rugbynion, basketball) that can be classified. Unfortu-ately, there is no specific detail regarding the typef activity being performed at the time of the inci-ent, which is often necessary to identify effectivereventive actions. For example, if the sport- oreisure-related activity was identified as cricket,o information is available regarding whether thelayer was batting, bowling, or fielding when thetruck by/struck against injury occurred, nor whathase of activity was being undertaken, such asraining or competition. Nevertheless, having infor-ation about the sport played at time of injury is aajor advance for sports injury researchers in Aus-

ralia, but this has not yet been adopted universallynternationally in ICD-10.

ources of additional information

nother option to gain further detail regarding thenjurious incident may be to combine or link differ-nt data sources together to obtain a more compre-ensive picture of injury events and outcomes.16,17

ata linkage involves combining information from

wo or more different data sources either throughanual methods, deterministic procedures, orrobabilistic linking techniques.18 Combining infor-ation from different data sources regarding the

Getting more out of being struck 503

Table 3 Activity of incident for sport- or leisure-related struck by/struck against injury hospitalisations, NSW,number and per cent, 2002—2003 to 2003—2004

Activity Number Per cent

Team ball sports 2350 58.3Football (including Australian Rules, Rugby Union, Rugby League, Soccer) 2146 53.3Basketball 124 3.1

Team bat or stick sports 471 11.7Baseball 57 1.4Cricket 269 6.7Field hockey 64 1.6Softball 34 0.8

Team water sports (e.g. water polo) 6 0.1Boat sports 35 0.9

Individual water sports 238 5.9Fishing 15 0.4Surfing and boogie boarding 127 3.2Swimming 61 1.5

Ice and snow sports 43 1.1Individual athletic sports 29 0.7Aesthetic activities (e.g. dancing) 8 0.2Racquet sports (e.g. tennis) 33 0.8Target and precision sports 58 1.4Combat and martial arts 146 3.6Power sports 19 0.5Equestrian activities 12 0.3Adventure sports (e.g. hiking) 4 0.1Wheeled motor sports 11 0.3Non-wheeled motor sports (e.g. cycling, roller balding, scooters) 42 1.0Aero sports (e.g. hang gliding) 4 0.1Other school-related sports 25 0.6Other sport, exercise and leisure activities 403 10.0Other activities 86 2.1Not known/missing 6 0.1

Total 4029 100.0

Table 4 Location of incident for sport- or leisure-related struck by/struck against injury hospitalisations, NSW,number and per cent, 1999—2000 to 2003—2004

Location Number Per cent

Home 108 1.3Residential institution 10 0.1School, other institution and public administrative area 216 2.6Sports and athletics area 6723 82.2Street and highway 14 0.2Trade and service area 38 0.5Industrial and construction area 2 0.0Farm 3 0.0Other specified places 330 4.0Unspecified place 736 9.0Not known/missing 1 0.0

Total 8181 100.0

1isNaiAhdcs

A

TMpRAPt

dmca

504

same injury event is often able to provide anincreased understanding of an event and/or out-come as the combined data sources can often pro-vide a temporal ordering to the sequence of eventsfrom the causal factors leading to a injury-relatedincident, to a description of the incident itself andof its outcome.

The addition of a narrative text field to comple-ment the coded data may be another alternativethat has been investigated elsewhere with somesuccess.19,20 The use of data text mining of nar-rative text fields is not a new concept to the NSWHealth system as this technique has been used toexamine records of presentations to a number ofemergency departments and has potential use forinjury prevention efforts.21

Additionally, the NSW ISC does not capture infor-mation relating to date of injury. If such a variablewere introduced, it could be used with probabilisticdata linkage methods (or a unique patient iden-tifier), to determine whether a given episode ofcare for a patient was the first admission to hospi-tal for a particular injury. This would enable betterestimates of the incidence of sports- and leisure-related activity.

Limitations

This paper only reviewed one type of injury, sport-or leisure-related injuries, and one type of injurymechanism, struck by/struck against injuries. It ispossible that analyses of other injury mechanismswill reveal additional issues and outline further rec-ommendations to improve hospitalisation data forinjury prevention purposes. This paper also onlyreviewed a number of variables from the NSW ISCand does not represent the total information thatis available for analysis.

One limitation of the analysis conducted is theuse of the whole population estimates to calculateinjury incidence rates as these are likely to resultin an underestimate of the true incidence rate. Infact, only 28.4% of NSW residents reported takingpart in some form of organised sporting activity inthe 12 months prior to 1999—2000.22

Conclusion

This examination of sport- or leisure-related struckby/struck against injury hospitalisation data sought

to provide an indication of several areas forimprovement both in the collection of hospitali-sation data in NSW and in the ICD-10-AM exter-nal causes classification system. In NSW, during

R

R. Mitchell, A. Hayen

999—2000 to 2003—2004 injury was the sixth lead-ng cause of hospitalisation with 606,954 hospitali-ations, representing 6.3% of all hospitalisations ofSW residents (A Hayen and R Mitchell unpublishednalysis). Injury is thus an important area for thedentification of appropriate prevention strategies.lthough, these recommendations for improvementave been illustrated using NSW hospitalisationata, they have national and international impli-ations as ICD is one of the most widely used clas-ification system for hospital morbidity records.23

Practical applications

• To develop targeted injury prevention strate-gies for hospitalised sport- or leisure-relatedstruck by/struck against injuries additionalinformation is required.

• Both the activity and location codes inthe existing ICD-10-AM classification systemrequire further enhancement to allow for addi-tional detail to be collected.

• Use of the International Classification of Exter-nal Causes of Injuries would augment hospitalinjury-related morbidity data currently usingICD-10-AM.

• Injury-related hospitalisation data would ben-efit from the addition of narrative text fieldsand from data linkage with other data collec-tions that include information on the circum-stances of injury.

cknowledgements

he authors are supported by the NSW Injury Riskanagement Research Centre, with core fundingrovided by the NSW Health Department, the NSWoads and Traffic Authority and the Motor Accidentsuthority. Rebecca Mitchell is also supported by ah.D. scholarship from Injury Prevention and Con-rol Australia.

The authors wish to thank the Centre for Epi-emiology and Research at the NSW Health Depart-ent for providing the data from the Health Out-

omes and Information Statistical Toolkit (HOIST)nalysed in this study.

eferences

1. Australian Government Department of Health and Ageing.Sports safety in Australia. Canberra: Commonwealth of Aus-tralia; 2003.

G

1

1

1

1

etting more out of being struck

2. World Health Organization. ICD-9 International Classifica-tion of Diseases, 9th revision. Geneva: World Health Orga-nization; 1977.

3. World Health Organization. ICD-10 International Classifica-tion of Diseases, 10th revision. Geneva: World Health Orga-nization; 1992.

4. National Centre for Classification in Health. ICD-10-AM. 1sted. Sydney: National Centre for Classification in Health;1998.

5. National Centre for Classification in Health. ICD-10-AM. 2nded. Sydney: National Centre for Classification in Health;2000.

6. National Centre for Classification in Health. ICD-10-AM. 3rded. Sydney: National Centre for Classification in Health;2002.

7. National Centre for Classification in Health. ICD-10-AM. 4thed. Sydney: Commonwealth Department of Health; 2004.

8. Langley J, Chalmers D. Coding the circumstances ofinjury: ICD-10 a step forwards or backwards? Inj Prev1999;5:247—53.

9. MacIntyre C, Ackland M, Chandraraj E. Accuracy of injurycoding in Victorian hospital morbidity data. Aust N Z J PublicHealth 1997;21(7):779—83.

0. Population Health Division. The Health of the People ofNSW. Report of the Chief Health Officer. Sydney: NSW HealthDepartment; 2004.

1. SAS Institute. SAS: statistical software, version 8.02. Cary,North Carolina: SAS Institute; 2000.

2. Centre for Epidemiology and Research. Health Outcomes and

Information Statistical Toolkit (HOIST). NSW Department ofHealth; 2006.

3. Dobson A, Kuulasmaa K, Eberle E, Schere J. Confidence inter-vals for weighted sums of Poisson parameters. Stat Med1991;10:457—762.

505

14. Armitage P, Berry G, Matthews J. Statistical methods inmedical research. 4th ed. Cornwell: Blackwell Science;2002.

15. WHO Collaborating Center on Injury Surveillance. Inter-national Classification of External Causes of Injury WorldHealth Organization; 2005; Available from: URL:http://www.iceci.org/. [cited 29 June 2005].

16. Graitcer PL. Injury surveillance. In: Halperin W, Baker E, edi-tors. Public health surveillance. New York: John Wiley & SonsInc.; 1992. p. 142—56.

17. Horan JM, Mallonee S. Injury surveillance. Epidemiol Rev2003;25:24—42.

18. Clark DE. Practical introduction to record linkage for injuryresearch. Inj Prev 2004;10(3):186—91.

19. Lincoln AE, Sorock GS, Courtney TK, Wellman HM, Smith GS,Amoroso PJ. Using narrative text and coded data to develophazard scenarios for occupational injury interventions. InjPrev 2004;10(4):249—54.

20. Williamson A, Feyer A-M, Stout N, Driscoll T, Usher H.Use of narrative analysis for comparisons of the causesof fatal accidents in three countries: New Zealand, Aus-tralia, and the United States. Inj Prev 2001;7(Suppl. I):i15—20.

21. Muscatello D, Churches T, Kaldor J, Zheng W, Chiu C, CorrellP, et al. An automated, broad-based, near real-time publichealth surveillance system using presentations to hospitalEmergency Departments in New South Wales, Australia. BMCPublic Health 2005;4(141).

22. Australian Bureau of Statistics. Participation in Sport

and Physical Activities. (Cat. No. 4177.0) Canberra: ABS;2000.

23. Harrison J. Injury classification: balancing continuity andutility. Inj Control Saf Promot 2000;7(1):51—63.