NSW INJURY PROFILE: During to

I R M R C N S W I n j u r y R i s k Management Research centre NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 Andrew Hayen and Rebecca Mitchell NSW Injury Risk Management Research Centre The University of New South Wales Sydney, Australia ISBN 0 7334 2301 9 June 2006

Copyright The University of New South Wales. Suggested citation: Hayen, A. Mitchell, R. NSW Injury Profile: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004. Sydney: NSW Injury Risk Management Research Centre, The University of New South Wales, 2006.

contents / Contents List of Tables iii List of Figures vi Abbreviations viii Acknowledgements ix Executive Summary xi 1. Introduction 1 2. Methods 3 2.1 Definitions 3 2.1.1 Injury 3 2.1.2 Injury mechanism 3 2.1.3 Injury hospitalisation 4 2.1.4 Population data source 5 2.1.5 Injury data coding issues 5 2.2 Analysis 6 2.2.1 Top 10 causes of hospitalisation for NSW residents 7 2.2.2 Time trends 7 2.2.3 Age- and sex-specific rates 7 2.2.4 Injury mechanism subcategory-specific frequencies and rates 7 3. Injury hospitalisations in NSW 8 4. Falls 12 5. Motor vehicle transport 16 6. Struck by or struck against injuries 21 7. Self-harm 25 8. Cut or pierce-related injuries 29 9. Interpersonal violence 33 10. Poisoning 37 11. Non-motor vehicle road transport 41 12. Natural and environmental factors 45

ii / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 13. Foreign bodies 50 14. Fire and burns 53 15. Machinery injuries 57 16. Near-drowning 61 17. Conclusions and recommendations 65 18. References 70 Appendix 1. List of Ecodes by mechanism 72 Appendix 2. List of disease and injury categories for top 10 causes of hospitalisation 73 Appendix 3. List of Ecodes by injury mechanism and injury mechanism subcategory 74 Appendix 4. Age-specific rates of hospitalisation by age group and mechanism 77

list of tables / iii List of tables Table 1. Significant changes in injury-related hospitalisation rates by mechanism, NSW, 1989 1990 to 2003 2004 xii Table 2. Top 10 leading causes of hospitalisation by age group, NSW, 1999 2000 to 2003 2004 9 Table 3. Injury hospitalisations by mechanism in NSW, number, rate and CI, 1999 2000 to 2003 2004 11 Table 4. Injury hospitalisations by fall submechanism, NSW, number, rate and CI, 1999 2000 to 2003 2004 14 Table 5. Number of hospitalisations for falls by age group and cause, NSW, 1999 2000 to 2003 2004 15 Table 6. Injury hospitalisations by motor vehicle transport road user class, NSW, number, rate and CI, 1999 2000 to 2003 2004 18 Table 7. Number of hospitalisations for motor vehicle transport incidents by age group and cause, NSW, 1999 2000 to 2003 2004 20 Table 8. Injury hospitalisations by struck by/struck against injury cause, NSW, number, rate and CI, 1999 2000 to 2003 2004 23 Table 9. Number of hospitalisations for struck by/struck against injuries by age group and cause, NSW, 1999 2000 to 2003 2004 24 Table 10. Injury hospitalisations by method of self-harm, NSW, number, rate and CI, 1999 2000 to 2003 2004 27 Table 11. Number of hospitalisations for self-harm by age group and cause, NSW, 1999 2000 to 2003 2004 28 Table 12. Injury hospitalisations by cut/pierce injury cause, NSW, number, rate and CI, 1999 2000 to 2003 2004 31 Table 13. Number of hospitalisations for cut/pierce injuries by age group and cause, NSW, 1999 2000 to 2003 2004 32 Table 14. Injury hospitalisations by interpersonal violence method, NSW, number, rate and CI, 1999 2000 to 2003 2004 35 Table 15. Number of hospitalisations for interpersonal violence by age group and cause, NSW, 1999 2000 to 2003 2004 36

iv / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 Table 16. Injury hospitalisations by poisoning substance, NSW, number, rate and CI, 1999 2000 to 2003 2004 39 Table 17. Number of hospitalisations for poisoning by age group and cause, NSW, 1999 2000 to 2003 2004 40 Table 18. Injury hospitalisations by non-motor vehicle road transport-related cause, NSW, number, rate and CI, 1999 2000 to 2003 2004 43 Table 19. Number of hospitalisations for non-motor vehicle road transport-related injuries by age group and cause, NSW, 1999 2000 to 2003 2004 44 Table 20. Injury hospitalisations by natural and environmental factor-related cause, NSW, number, rate and CI, 1999 2000 to 2003 2004 47 Table 21. Number of hospitalisations for natural and environmental factor-related injuries by age group and cause, NSW, 1999 2000 to 2003 2004 49 Table 22. Injury hospitalisations for foreign bodies, NSW, number, rate and CI, 1999 2000 to 2003 2004 51 Table 23. Number of hospitalisations for foreign body-related injuries by age group, NSW, 1999 2000 to 2003 2004 52 Table 24. Injury hospitalisations by fire/burns type, NSW, number, rate and CI, 1999 2000 to 2003 2004 55 Table 25. Number of hospitalisations for injury due to fire/burns by age group and cause, NSW, 1999 2000 to 2003 2004 56 Table 26. Injury hospitalisations by machinery-related cause, NSW, number, rate and CI, 1999 2000 to 2003 2004 59 Table 27. Number of hospitalisations for machinery-related injuries by age group and cause, NSW, 1999 2000 to 2003 2004 60 Table 28. Injury hospitalisations by near-drowning location, NSW, number, rate and CI, 1999 2000 to 2003 2004 63 Table 29. Number of hospitalisations for near-drowning by age group and cause, NSW, 1999 2000 to 2003 2004 64 Table 30. Significant changes in injury-related hospitalisation rates by mechanism, NSW, 1989 1990 to 2003 2004 66

list of tables / v Table 31. Age-specific rates of hospitalisation per 100,000 population by age group and mechanism for all persons, NSW, 1999 2000 to 2003 2004 78 Table 32. Age-specific rates of hospitalisation per 100,000 population by age group and mechanism for males, NSW, 1999 2000 to 2003 2004 79 Table 33. Age-specific rates of hospitalisation per 100,000 population by age group and mechanism for females, NSW, 1999 2000 to 2003 2004 80

vi / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 List of figures Figure 1. Injury hospitalisation rates by sex, NSW, 1989 1990 to 2003 2004 8 Figure 2. Injury hospitalisation rates by age group and sex, NSW, 1999 2000 to 2003 2004 10 Figure 3. Hospitalisation rate for injury due to falls by sex, NSW, 1989 1990 to 2003 2004 13 Figure 4. Age-specific hospitalisation rate for falls by sex, NSW, 1999 2000 to 2003 2004 13 Figure 5. Hospitalisation rate for injury due to motor vehicle transport by sex, NSW 1989 1990 to 2003 2004 17 Figure 6. Age-specific hospitalisation rate for motor vehicle transport by sex, NSW, 1999 2000 to 2003 2004 17 Figure 7. Hospitalisation rate for injury due to struck by/struck against injuries by sex, NSW 1989 1990 to 2003 2004 22 Figure 8. Age-specific hospitalisation rate for struck by/struck against injuries by sex, NSW, 1999 2000 to 2003 2004 22 Figure 9. Hospitalisation rate for injury due to self-harm by sex, NSW 1989 1990 to 2003 2004 26 Figure 10. Age-specific hospitalisation rate for self-harm by sex, NSW, 1999 2000 to 2003 2004 26 Figure 11. Hospitalisation rate for injury due to cut/pierce injuries by sex, NSW 1989 1990 to 2003 2004 29 Figure 12. Age-specific hospitalisation rate for cut/pierce injuries by sex, NSW, 1999 2000 to 2003 2004 30 Figure 13. Hospitalisation rate for injury due to interpersonal violence by sex, NSW 1989 1990 to 2003 2004 34 Figure 14. Age-specific hospitalisation rate for interpersonal violence by sex, NSW, 1999 2000 to 2003 2004 34 Figure 15. Hospitalisation rate for injury due to poisoning by sex, NSW 1989 1990 to 2003 2004 38

list of figures / vii Figure 16. Age-specific hospitalisation rate for poisoning by sex, NSW, 1999 2000 to 2003 2004 38 Figure 17. Hospitalisation rate for injury due to non-motor vehicle road transport-related injuries by sex, NSW 1989 1990 to 2003 2004 42 Figure 18. Age-specific hospitalisation rate for non-motor vehicle road transport-related injuries by sex, NSW, 1999 2000 to 2003 2004 42 Figure 19. Hospitalisation rate for injury due to natural and environmental factor-related injuries by sex, NSW 1989 1990 to 2003 2004 46 Figure 20. Age-specific hospitalisation rate for natural and environmental factor-related injuries by sex, NSW, 1999 2000 to 2003 2004 46 Figure 21. Hospitalisation rate for injury due to foreign body-related injuries by sex, NSW 1989 1990 to 2003 2004 50 Figure 22. Age-specific hospitalisation rate for foreign body-related injuries by sex, NSW, 1999 2000 to 2003 2004 51 Figure 23. Hospitalisation rate for injury due to fire/burns by sex, NSW 1989 1990 to 2003 2004 55 Figure 24. Age-specific hospitalisation rate for injury due to fire/burns by sex, NSW, 1999 2000 to 2003 2004 55 Figure 25. Hospitalisation rate for injury due to machinery-related injuries by sex, NSW 1989 1990 to 2003 2004 58 Figure 26. Age-specific hospitalisation rate for machinery-related injuries by sex, NSW, 1999 2000 to 2003 2004 58 Figure 27. Hospitalisation rate for injury due to near-drowning by sex, NSW 1989 1990 to 2003 2004 62 Figure 28. Age-specific hospitalisation rate for near-drowning by sex, NSW, 1999 2000 to 2003 2004 62

viii / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 Abbreviations ABS CI Ecode HOIST ICD ICD-10 ICD-10-AM ICD-9 ICD-9-CM IPV IRMRC ISC MV MVT NCC NCCH Ncode NEC NEF NSW WHO Australian Bureau of Statistics Confidence interval External cause of injury code Health Outcomes and Information Statistical Toolkit International Classification of Disease International Classification of Diseases and Related Health Problems, 10 th Revision International Classification of Disease, 10 th Revision, Australian Modification International Classification of Disease, 9 th Revision International Classification of Disease, 9 th Revision, Clinical Modification Interpersonal violence NSW Injury Risk Management Research Centre Inpatient Statistics Collection Motor vehicle Motor vehicle transport National Coding Centre National Centre for Classification in Health Nature of medical condition or injury Not elsewhere classified Natural and environmental factors New South Wales World Health Organization

/ ix Acknowledgements The NSW Injury Risk Management Research Centre (IRMRC) is funded by the NSW Department of Health, the NSW Roads and Traffic Authority, and the NSW Motor Accidents Authority and supported by the University of New South Wales. Production of this report was funded through the IRMRC s core research program. The Centre for Epidemiology and Research of the NSW Department of Health provided the data used in this report, which were accessed via HOIST. We are also grateful for the use of some SAS macros developed by the Centre for Epidemiology and Research. Comments on the draft report were received from: Pam Albany, NSW Department of Health Kwame Atsu, Motor Accidents Authority Caroline Finch, NSW Injury Risk Management Research Centre Gwen Cosier, NSW Department of Health Andrew Graham, Roads and Traffic Authority Claire Monger, NSW Department of Health Maureen Owen, NSW Department of Health

EXECUTIVE SUMMARY / xi Executive Summary This report provides an overview of injury-related hospitalisations of NSW residents during 1989 1990 to 2003 2004. Injury-related morbidity data for this report were obtained from the NSW Inpatient Statistics Collection of the NSW Health Department. This report describes in detail the 12 most common injury-related mechanisms which represent 85.1% of injury-related hospitalisations, along with neardrowning events that resulted in hospitalisation. Injury, poisoning and certain other consequences of external causes represent a large proportion of all admissions to hospital for NSW residents, and were the sixth highest cause of hospitalisation for NSW residents for all age groups during 1999 2000 to 2003 2004. During this period, there were 503,530 admissions to hospital that were injury-related, giving a hospitalisation rate of 1,523 per 100,000 population. Males had one-and-a-half times the injury hospitalisation rate of females. Individuals 65 years and older (predominantly for fall-related injuries) and 15 24 years (particularly for motor vehicle transport and fall-related injuries) had the highest hospitalisation rates compared to all other age groups. Falls, motor vehicle transport-related injuries, struck by/struck against injuries, injuries resulting from self-harm, cut/pierce injuries, and injuries resulting from interpersonal violence were the most common injury-related hospitalisations during 1999 2000 to 2003 2004. Over the period 1989 1990 to 2003 2004, the overall injury rate remained fairly constant. However, yearly hospitalisation rates significantly decreased or increased for a number of injury mechanisms (Table 1). Priority areas identified for prevention activities are: falls motor vehicle transport incidents struck by/struck against injuries injuries sustained during an attempt at self-harm cut/pierce-related injuries injuries as a result of interpersonal violence.

xii / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 Table 1. Significant changes 1 in injury-related hospitalisation rates by mechanism, NSW, 1989 1990 to 2003 2004 Injury Mechanism Males Females Falls Motor vehicle transport Struck by/against Self-harm Cut/pierce - Interpersonal violence Poisoning Non-motor vehicle road transport - Natural/environmental factors Foreign bodies - Fire and burns Machinery Near-drowning All injury - - 1 An upwards facing arrow indicates a significant increase and a downwards facing arrow indicates a significant decrease. A dash indicates no significant trend.

EXECUTIVE SUMMARY/ xiii Recommendations to enhance the information collected regarding particular injury mechanisms include: recording the date of injury incorporating additional detail into the ICD-10-AM classification system for injuries resulting from: - falls - foreign bodies - fire and burns - struck by/struck against injuries - cutting/piercing injuries.

Introduction / 1 1 Introduction Injuries are a significant public health issue with over five million injury-related deaths worldwide each year (Krug et al, 2000). In Australia, injury is the leading cause of death for individuals aged 44 years or less (Kreisfeld and Harrison, 2005). Around 2,500 individuals are fatally injured in New South Wales (NSW) each year (Schmertmann et al, 2004). While injury-related mortality represents a portion of the injury burden in NSW, many more individuals are hospitalised in NSW following an injury. Together lifetime injury-related mortality and morbidity were estimated to cost $3.53 billion in NSW during 1998 1999 (Potter-Forbes and Aisbett, 2003). Injuries are preventable occurrences and through examining their frequency and causes, appropriate injury prevention strategies can be developed. There has been a range of interventions developed that are effective in preventing injuries, such as changes in legislation, regulation or policies, improved enforcement, environmental changes, improvements in design, and changes in individual behaviour (National Injury Prevention Advisory Council, 1999). The prevention of injury-related morbidity and associated disability in the community leads to cost savings in both direct and indirect costs associated with these incidents, including cost savings associated with medical treatment and long term care (Miller and Levey, 2000). Access to information on hospitalised injury and detailed analysis and reporting of this information assists in establishing the magnitude of hospitalised injury and in the identification of areas for injury prevention activities. This report provides an overview of the causes of hospitalisation for NSW residents. It describes the trend of hospitalised injury of NSW residents for selected injury mechanisms from 1989 1990 to 2003 2004, and presents a detailed analysis of select injury mechanisms during 1999 2000 to 2003 2004.

2 / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004

METHODS / 3 2 Methods 2.1 Definitions The following sections present the case definitions of injury, injury mechanism and morbidity used for the purposes of this report. 2.1.1 Injury According to Robertson (1998), An injury results when too much or too little energy (in the case of asphyxiation) is transferred to the human body, at rates or amounts that are above or below the tolerance of human tissues, resulting in damage. The World Health Organization (WHO) defines an injury similarly. An injury is a bodily lesion at the organic level resulting from acute exposure to energy (this energy can be mechanical, thermal, electrical, chemical, or radiant) interacting with the body in amounts or rates that exceed the threshold of physiological tolerance (Peden et al, 2001). Section 2.1.3 outlines the criteria used for putting into practice the definitions of injury used in this report. In order to recognise the physical nature of an injury (e.g. a broken leg) and the external cause of the injury (e.g. a fall), two separate sets of codes were developed by WHO as part of its work on an International Classification of Disease (ICD) coding structure (WHO, 1977; WHO, 1992). One set, known as diagnostic codes or Ncodes, describes the physical nature of an injury and provides important information from a clinical standpoint. The other set, known as external cause codes or Ecodes, provides important information for prevention purposes, by identifying the type of energy that caused the physical injury. Section 2.1.3 describes the case selection process using these codes for this report. 2.1.2 Injury mechanism Injuries are usually classified in terms of their external cause and intent. An injury mechanism (represented by an Ecode) is defined as the external object or circumstance that caused the injury, such as motor vehicle transport or drowning. The intent can be unintentional, intentional or undetermined. For example, the intent of an injury caused by a firearm could be unintentional, intentional (e.g. homicide, self-harm) or not able to be determined. Injury mechanisms that are intentional are either self-inflicted or inflicted by another person or persons. All injuries that are intentionally self-inflicted are grouped under an injury mechanism called self-harm. For example, a poisoning that is self-inflicted is considered to be self-harm and is therefore separated from poisonings that have occurred unintentionally. However, for this report, if the individual was aged less than 10 years the ingestion of a poisonous substance(s) was not considered to be a self-harm event and was included in the other injuries category as a young child s understanding of the concept of

4 / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 death and living is thought to be immature (Mishara, 1999). All injuries that are intentionally inflicted by another person or persons are grouped under an injury mechanism called interpersonal violence. Injury caused by the intentional use of a firearm on another person is considered to be interpersonal violence and is therefore separate from unintentional firearm injuries. Thirteen injury mechanisms are described in this report. Twelve of the mechanisms each resulted in more than 2,000 hospitalisations of NSW residents during 1999 2000 to 2003 2004, and accounted for 85.1% of injury-related hospitalisations. The remaining injury mechanism, near-drowning, resulted in fewer than 2,000 hospitalisations over this period, but is regarded as a national priority area and is included in this report (Australian Water Safety Council, 2004). The International Classification of Disease, version 9 clinical modification (ICD-9-CM) and the International Classification of Diseases, version 10 Australian modification (ICD-10-AM) Ecodes for the injury mechanisms included in this report are listed in Appendix 1. During the period of the report, various editions of ICD-10-AM were used to code hospital separations in NSW. 2.1.3 Injury hospitalisation Hospitalisation data were obtained from the NSW Inpatient Statistics Collection (ISC), a census (since July 1, 1993) of all services for admitted patients to public and private hospitals, private day procedures, and public psychiatric hospitals. The ISC is a financial year collection from 1 July through to 30 June of the following year. The ISC is maintained by the NSW Department of Health. Data were obtained via the Health Outcomes and Information Statistical Toolkit (HOIST). The ISC also contains data on hospitalisations of NSW residents that occurred in another state. However, these data were not available for 2003 2004. The number of interstate hospitalisations for this year was imputed based on hospitalisations for the previous three years. Details of the method used may be found in the The health of the people of NSW: Report of the Chief Health Officer (Population Health Division, 2004). Data for 1998 1999 and following years are for episodes of care in hospital, which end with the discharge, transfer or death of the patient, or when the service category for the admitted patient changed. Data for the years 1989 1990 to 1997 1998 are for periods of stay. Periods of stay end with the discharge, transfer, or death of the patient. The change from period of stay to episodes of care may cause a small rise in the apparent number of hospitalisations in the later years. Since 1998 1999, ISC data have been coded using the ICD-10-AM (National Centre for Classification in Health, 2000). For the years 1989 1990 to 1997 1998, ISC data were coded using ICD-9-CM (National Coding Centre, 1996).

METHODS / 5 Hospitalisations that satisfied the following criteria were included in the report: The hospitalisation was for a patient who was a resident of NSW A principal diagnosis in the ICD-10-AM range S00-T98 (1998 1999 to 2003 2004) or in the ICD-9-CM range 800-999 (for 1989 1990 to 1997 1998) An external cause code in the ICD-10-AM range V01-Y39 or Y85-Y98 (1998 1999 to 2003 2004) or in the ICD-9-CM range E800-E869, E880-E929, E950-E999 (for 1989 1990 to 1997 1998). In addition, hospital separations relating to transfers or statistical discharges were excluded. This was to partly eliminate multiple counts, which occur when an injured person has more than one hospitalisation for a given injury. In Table 2, hospitalisations with a principal diagnosis in the ICD-10-AM range S00-T98 and with an external cause of complications of care (ICD-10-AM: Y40-Y84, Y88; ICD-9-CM: E870-E879, E930- E948) are also included, along with those described above. These hospitalisations (i.e. including complications of care ) are referred to as Injury, poisoning and certain other consequences of external causes to distinguish these analyses from those presented in the rest of this report. Data in this report include 16,974 NSW residents who died whilst hospitalised due to injury, poisoning and certain other consequences of external causes between 1989 1990 and 2003 2004. 2.1.4 Population data source Age- and sex-specific population estimates as at 30 December of each year were obtained from the NSW Department of Health. These estimates are based on the Australian Bureau of Statistics (ABS) population estimates as at 30 June. More detail regarding ABS population estimates may be found in The Health of the People of NSW: Report of the Chief Health Officer (Population Health Division, 2004). 2.1.5 Injury data coding issues The data used in this report span a change in the coding scheme used to classify injury and disease. The ICD was initially formalised in 1893. Since 1948, it has been revised in its entirety approximately every 10 years by WHO. The two ICD revisions covered in this report are ICD-9-CM (NCC, 1996), which was in use in the ISC from 1989 1990 to 1997 1998, and ICD-10-AM (NCCH, 2000) used from 1998 1999 onward. In ICD-10 alphanumeric codes were introduced (e.g. A37, R01) to represent an injury or disease, superseding the numeric codes (e.g. 125, 802) used in ICD-9. The external cause of injury codes have been included within the alphanumeric structure of ICD-10, as opposed to the separate scheme in ICD-9 (i.e., use of E800-E999).

6 / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 At the time of separation from hospital, a consequence or nature of injury code is assigned by a medical coder on the patient s medical record. In ICD-9, there was a specific Ncode for each injury (i.e., 800 999) and the codes were organised by the type of injury (e.g. fracture, dislocation). In ICD-10, a unique Ncode still exists, but the codes are organised by the location of the body part injured (e.g. head) instead of the type of injury. For each injury Ncode and a few other disease Ncodes, an external cause of injury code (Ecode) must also be supplied to identify the cause or mechanism of the injury (e.g. drowning, fall, burn). Two major changes regarding Ecodes occurred between ICD-9 and ICD-10. In ICD-9, the person injured in a transport incident (e.g. motor vehicle) was secondary to the type of incident (e.g. collision with other motor vehicle). However, in ICD-10, the coding structure focuses firstly on the person injured and secondly on the type of incident. The second change in ICD-10 was the introduction of codes for the place where the injury occurred (e.g. home) and the activity at the time of the injury (e.g. playing sport). 2.2 Analysis Each of the following sections briefly describes the types of analysis conducted using the hospitalisation data. Three types of epidemiological analyses were conducted: number of hospitalisations age- and sex-specific rate of hospitalisations age-adjusted rates of hospitalisations. Age-specific rates were calculated by dividing the number of hospitalisations for a particular age group (e.g. under five years) by the population of the age group. Rates are presented as the number of hospitalisations per 100,000 population, except in the case of self-harm where rates are presented as the number of hospitalisations per 100,000 population aged 10 years or older. Age-adjustment is used to adjust for the effects of differences in the age-composition of populations across time or geographic region. In this report, age-adjustment was calculated using direct agestandardisation. An age-adjusted rate is a weighted sum of age-specific rates, where each weight is an age-specific population in the standard population. The estimated Australian residential population as at 30 June 2001 was used in this report as the standard population. Confidence intervals were calculated using the method of Dobson et al (1991). The following analyses are provided in this report. 2.2.1 Top 10 causes of hospitalisation for NSW residents All hospitalisations for 1999 2000 to 2003 2004 were grouped into disease and injury categories, using the principal diagnosis. The disease categories were based on the disease chapter headings in ICD-10.

METHODS / 7 The list of disease and injury categories used is at Appendix 2. The top 10 causes of hospitalisation tables were generated by ranking the frequencies of each disease and injury by age group. The following age groups were used to present frequencies for the top 10 leading causes of hospitalisation tables: under 1, 1 4, 5 9, 10 14, 15 24, 25 34, 35 44, 45 54, 55 64, and 65+ years. Note that the numbers of injury hospitalisations in the top 10 causes of hospitalisation table (Table 2) will differ substantially from those given in the rest of the report, because this table includes all hospitalisations with a principal diagnosis of injury, poisoning and certain other consequences of external causes. However, in the rest of this report, those hospitalisations with a principal diagnosis of injury, poisoning and certain other consequences of external causes must also have an external cause code in the range in the ICD-10-AM range V01-Y39, Y85-Y87 or Y89-Y98 (1998 1999 to 2003 2004) or in the ICD-9-CM range E800-E869, E880-E929, E950-E999 (for 1989 1990 to 1997 1998) to be considered as an injury-related hospitalisation (see section 2.1.3). 2.2.2 Time trends Age-adjusted rates for each injury mechanism were calculated annually from 1989 1990 to 2003 2004. A Poisson or negative binomial regression analysis (with population as an offset) was performed to examine the statistical significance of changes in the trend over the time period, and to calculate the annual percentage change in the rate of hospitalisations. This method takes into account changes in the age-structure of the population. Because of coding changes and changes in admission practices of hospitals, caution needs to be exercised in the interpretation of these trends. 2.2.3 Age- and sex-specific rates Age and sex-specific rates for five-year age groups were calculated for each injury mechanism for 1999 2000 to 2003 2004 and presented by sex and age group. 2.2.4 Injury mechanism subcategory-specific frequencies and rates The total number of hospitalisations for 1999 2000 to 2003 2004 for each injury mechanism was divided into subcategories specific to each injury mechanism. The list of injury mechanism subcategories by Ecode is at Appendix 3. The period 1999 2000 to 2003 2004 was chosen because hospital separations throughout this period were coded according to ICD-10-AM. Frequencies for the injury mechanism subcategories were also ranked by age group and presented in a Top 10 table format. The following age groups were used to present frequencies for the top 10 leading causes of injury hospitalisation tables: under 1, 1 4, 5 9, 10 14, 15 24, 25 34, 35 44, 45 54, 55 64, and 65+ years. Frequencies and rates for all persons, males and females were also calculated for each injury mechanism. This information was presented in a table and the subcategories were ranked by the number of hospitalisations in each subcategory.

8 / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 3. Injury hospitalisations in NSW During the period 1999 2000 to 2003 2004, injury, poisoning and certain other consequences of external causes were the sixth leading cause of hospitalisation of all NSW residents (Table 2). Injury, poisoning and certain other consequences of external causes were the leading cause of hospitalisation in those aged 10 14 years, the second leading cause of hospitalisation in those aged 5 9 years, and the third leading cause in those aged 1 4 years and 15 24 years. Injury, poisoning and certain other consequences of external causes were one of the top 10 leading causes of hospitalisation in all age groups. There were 606,954 hospitalisations with a principal diagnosis of injury, poisoning and certain other consequences of external causes during this period, which represented 6.3% of all hospitalisations. Data from 1989 1990 to 2003 2004 were used to describe the profile of injury-related hospitalisations for NSW residents. Hospitalisation data from 1999 2000 to 2003 2004 were used in the majority of the analyses, except for the trend analyses, which used hospitalisation data from 1989 1990 to 2003 2004. Hospitalisations for which the external cause was a complication of care are not reported in the following analyses. Over the period 1989 1990 to 2003 2004, there was no significant trend in the age-adjusted hospitalisation rate for injury (Figure 1). Hospitalisation rates for injury in males were significantly higher than those in females for every year in this period. Figure 1. Injury hospitalisation rates by sex, NSW, 1989 1990 to 2003 2004 Rate per 100,000 0 500 1000 1500 2000 Males Females 89-90 90-91 91-92 92-93 93-94 94-95 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 Year

Injury Hospitalisations in NSW / 9 table 2. top 10 leading causes of hospitalisation by age group 1, nsw, 1999 2000 to 2003 2004 Age group Age group Rank <1 1 4 5 9 10 14 15 24 25 34 35 44 45 54 55 64 65+ Total 1 2 Factors affecting health status 289,309 Perinatal conditions 129,082 3 Congenital Abnormalities 4 5 6 34,214 Respiratory diseases 34,241 Ill-defined conditions 23,226 Infectious diseases 16,042 7 Diseases of the digestive system 8 7,808 Genitourinary disease 5,197 9 Injury, poisoning and certain other consequences of external causes 10 4,002 Mental disorders 3,354 Respiratory diseases 79,118 Infectious diseases 41,191 Injury, poisoning and certain other consequences of external causes 29,162 Diseases of the ear 22,528 Ill-defined conditions 22,200 Diseases of the digestive system 18,692 Factors affecting health status 16,156 Congenital Abnormalities 13,920 Genitourinary disease 9,634 Nervous system diseases 7,420 Respiratory diseases 39,740 Injury, poisoning and certain other consequences of external causes 31,784 Diseases of the digestive system 21,116 Diseases of the ear 15,867 Infectious diseases 15,867 Factors affecting health status 11,872 Ill-defined conditions 10,408 Mental disorders 9,499 Congenital Abnormalities 7,252 Nervous system diseases 7,146 Injury, poisoning and certain other consequences of external causes 36,659 Diseases of the digestive system 20,490 Respiratory diseases 18,414 Mental disorders 12,840 Ill-defined conditions 10,690 Factors affecting health status 10,199 Skin diseases 7,138 Musculoskeletal 7,066 Infectious diseases 6,840 Cancer 5,835 Pregnancy 155, 576 Diseases of the digestive system 105,441 Injury, poisoning and certain other consequences of external causes 96,412 Mental disorders 61,641 Factors affecting health status 45,863 Genitourinary disease 37,978 Respiratory diseases 37,126 Ill-defined conditions 34,634 Musculoskeletal 33,570 Skin diseases 21,400 Pregnancy 398,268 Factors affecting health status 125,492 Diseases of the digestive system 115,463 Injury, poisoning and certain other consequences of external causes 85,484 Genitourinary disease 80,629 Mental disorders 73,196 Musculoskeletal 48,204 Ill-defined conditions 47,791 Respiratory diseases 30,312 Cancer 27,218 Factors affecting health status 187,319 Diseases of the digestive system 142,753 Pregnancy 129,734 Genitourinary disease 101,985 Injury, poisoning and certain other consequences of external causes 73,285 Mental disorders 69,042 Musculoskeletal 66,930 Ill-defined conditions 62,311 Cancer 56,018 Diseases of circulatory system 41,997 Factors affecting health status 229,991 Diseases of the digestive system 180,880 Genitourinary disease 96,805 Cancer 96,489 Musculoskeletal 83,598 Diseases of circulatory system 78,685 Ill-defined conditions 77,352 Mental disorders 63,877 Injury, poisoning and certain other consequences of external causes 61,568 Nervous system diseases 35,315 Factors affecting health status 327,087 Diseases of the digestive system 186,323 Cancer 125,831 Diseases of circulatory system 113,257 Musculoskeletal 87,557 Ill-defined conditions 78,814 Genitourinary disease 75,911 Injury, poisoning and certain other consequences of external causes 50,477 Respiratory diseases 45,522 Mental disorders 37,593 Factors affecting health status 821,799 Diseases of circulatory system 363,239 Diseases of the digestive system 326,239 Cancer 218,653 Diseases of the eye 230,143 Ill-defined conditions 181,100 Respiratory diseases 168,040 Musculoskeletal 153,310 Injury, poisoning and certain other consequences of external causes 138,121 Genitourinary disease 136,763 Factors affecting health status 2,066,087 Diseases of the digestive system 1,125,187 Pregnancy 685,587 Cancer 657,684 Diseases of circulatory system 626,927 Injury, poisoning and certain other consequences of external causes 606,954 Genitourinary disease 556,867 Ill-defined conditions 548,527 Respiratory diseases 515,487 Musculoskeletal 487,887 1 The number of injury, poisoning and consequences of external causes hospitalisations includes a small proportion of hospitalisations with a principal diagnosis of injury (S00-T98) that did not have an external cause assigned.

10 / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 The hospitalisation rate was higher for males for all ages up to 70 years of age. For those aged 70 years or older, females had a higher hospitalisation rate than males (Figure 2). Rates of hospitalisation varied by sex, age group and injury mechanism (Appendix 4). Figure 2. Injury hospitalisation rates by age group and sex, NSW, 1999 2000 to 2003 2004 Rate per 100,000 0 1000 2000 3000 4000 5000 6000 7000 Males Females 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ Age-group (years) Falls, motor vehicle-related incidents and struck-by/struck against incidents were the most common causes of injury hospitalisation in the period 1999 2000 to 2003 2004 (Table 3). Hospitalisation rates of males were higher for all injury mechanisms than the rates of females, except for self-harm, where the female rate was significantly higher than the male rate. The injury hospitalisation rate for males was about 55% higher than the rate for females during 1999 2000 to 2003 2004. During 1999 2000 to 2003 2004, just over 12% of injury hospitalisations (12.4% or 62,283 hospitalisations) were intentional (i.e. self-harm or interpersonal violence), and less than 1% (0.5%, or 2,372) of hospitalisations were of undetermined intent. The remainder of injury-related hospitalisations (87.2%, or 438,646 hospitalisations) were due to unintentional injuries.

Injury Hospitalisations in NSW / 11 Table 3. Injury hospitalisations by mechanism in NSW, number, rate and CI, 1999 2000 to 2003 2004 Injury mechanism All Persons Male Female N Rate 1 95%CI 2 N Rate 1 95%CI 2 N Rate 1 95%CI 2 Motor vehicle transport Falls 175,077 522.8 (520.4,525.3) 84,743 537.1 (533.4,540.7) 90,333 487.7 (484.5,491.0) 53,211 162.1 (160.7,163.4) 36,299 221.7 (219.4,224.0) 16,912 101.8 (100.3,103.4) Struck by/against 35,642 108.8 (107.7,109.9) 27,394 166.4 (164.5,168.4) 8,248 49.8 (48.7,50.9) Self-harm 3 32,582 115.1 (113.9,116.4) 12,941 91.5 (89.9,93.1) 19,640 139.6 (137.6,141.5) Cut/pierce 31,054 94.6 (93.6,95.7) 23,472 142.7 (140.8,144.5) 7,583 46.3 (45.3,47.4) Interpersonal violence 29,701 90.8 (89.8,91.8) 23,356 142.1 (140.3,143.9) 6,343 38.8 (37.8,39.8) Poisoning 16,828 51.2 (50.4,51.9) 8,739 53.3 (52.2,54.5) 8,088 49.0 (47.9,50.0) Non-motor vehicle road transport Natural/ environmental factors 15,177 46.6 (45.8,47.3) 10,216 61.9 (60.7,63.1) 4,961 30.9 (30.0,31.8) 13,789 41.9 (41.2,42.6) 8,332 50.8 (49.7,51.9) 5,457 32.9 (32.0,33.8) Foreign bodies 8,628 26.1 (25.6,26.7) 5,214 31.9 (31.0,32.8) 3,414 20.4 (19.8,21.1) Fire and burns 7,892 24.0 (23.5,24.5) 5,184 31.5 (30.7,32.4) 2,708 16.4 (15.8,17.0) Machinery 7,656 23.3 (22.7,23.8) 7,055 42.9 (41.9,43.9) 601 3.7 (3.4,4.0) Near-drowning 914 2.8 (2.6,3.0) 636 3.8 (3.5,4.1) 278 1.7 (1.5,1.9) injuries 75,150 228.3 (226.7,230.0) 50,511 308.5 (305.8,311.2) 24,636 145.3 (143.5,147.1) All injury 4 503,301 1,552.9 (1,518.7,1,527.1) 304,092 1873.8 (1867.1,1,880.5) 199,202 1,145.6 (1,140.5,1,1150.7) 1 Age-adjusted rate per 100,000 population. 2 95% confidence interval. 3 The rate for self-harm was calculated per 100,000 population aged 10 years of age or older. 4 The number of NSW residents hospitalised interstate during 2003 2004 was imputed and as a result the sum of the number of hospitalisations due to injury mechanisms may not equal the all injury total. Summary Injury, poisoning and certain other consequences of external causes was the sixth leading cause of hospitalisation involving NSW residents over the period 1999 2000 and 2003 2004, accounting for 6.3% of all hospitalisations. The yearly hospitalisation rate for injuries was stable over the period 1989 1990 to 2003 2004 and during 1999 2000 to 2003 2004 the rate of hospitalisation was 1,522.9 per 100,000 population. In 2003 2004, there were 103,640 hospitalisations of NSW residents following an injury, giving a hospitalisation rate of 1,534.9 per 100,000 population. Falls, motor vehicle transport and struck-by/struck against were the injury mechanisms that most commonly led to a hospitalisation during 1999 2000 to 2003 2004. These were also the most common mechanisms that led to the hospitalisation of males. For females, the most common mechanisms leading to hospitalisation were falls, self-harm, and motor vehicle transport.

12 / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 4. Falls This section describes injury hospitalisations due to unintentional falls. WHO classifies the cause of fall-related hospitalisations by the circumstance in which the fall occurs. The types of falls include falls on the same level (for example, due to tripping, stumbling), from one level to another, from a building or other structure, on stairs, while being carried, and on a ladder/scaffolding (WHO, 1977; WHO, 1992). In NSW, falls were the fourth leading cause of injury-related death during 1998-2002, with 1,093 deaths, giving a mortality rate of 3.4 per 100,000 population (Schmertmann et al, 2004). The lifetime cost of fatal and non-fatal falls in NSW has been estimated at $644 million $333 million in direct costs and $311 million in mortality and morbidity costs (Potter-Forbes & Aisbett, 2003). Data from 1989 1990 to 2003 2004 were used to describe the profile of hospitalisations due to falls for NSW residents. Hospitalisation data from 1999 2000 to 2003 2004 were used in the majority of the analyses, except for the trend analyses, which used hospitalisation data from 1989 1990 to 2003 2004. Falls were the leading cause of injury hospitalisation during 1999 2000 to 2003 2004, and accounted for approximately 34.8% of all hospitalisations due to injury (Table 3). During this period, there were 175,077 hospitalisations due to a fall. The age-adjusted hospitalisation rate for fall-related injuries was 522.8 per 100,000 population. Figure 3 shows the trend in the hospitalisation rate for falls from 1989 1990 to 2003 2004. The yearly hospitalisation rate was estimated to have increased significantly by 3.0% per year (95% confidence interval: 2.2% to 3.9%) for males and to significantly increase by 1.1% (95% confidence interval for the increase: 0.3% to 2.0%) for females over this period. Figure 4 shows the age-specific hospitalisation rate for falls between 1999 2000 and 2003 2004. People aged 65 years or older were at greatest risk of being hospitalised for a fall-related injury. Those aged 5 to 14 years also showed a slight increase in hospitalisation rates compared to individuals aged 15 to 65 years. Males had higher rates than females till 49 years, then females 55 years or older had higher hospitalisation rates than men.

Falls / 13 Figure 3. Hospitalisation rate for injury due to falls by sex, NSW, 1989 1990 to 2003 2004 Rate per 100,000 0 100 200 300 400 500 600 Males Females 89-90 90-91 91-92 92-93 93-94 94-95 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 Year Falls on the same level were the most common type of fall-related hospitalisations, accounting for nearly half of all fall-related hospitalisations (46.6%). The rate of hospitalisation for males for falls from a building and from ladders and scaffolding were both more than four times higher than the rate for females (Table 4). The age-adjusted hospitalisation rate for males was 10.1% higher than the rate for females between 1999 2000 and 2003 2004. Figure 4. Age-specific hospitalisation rate for falls by sex, NSW, 1999 2000 to 2003 2004 Rate per 100,000 0 1000 2000 3000 4000 5000 6000 Males Females 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ Age-group (years)

14 / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 The types of falls leading to hospitalisation by age group are shown in Table 5. For individuals aged less than 10 years, falls from one level to another were the most common type of fall. For individuals aged 10 years or older, falls on the same level were the most common type of falls leading to hospitalisation. Table 4. Injury hospitalisations by fall submechanism, NSW, number, rate and CI, 1999 2000 to 2003 2004 Fall submechanism All Persons Male Female N Rate 1 95%CI 2 N Rate 1 95%CI 2 N Rate 1 95%CI 2 Same level 81,577 243.1 (241.4,244.8) 37,478 238.7 (236.3,241.1) 44,094 235.6 (233.3,237.8) One level to another 32,212 97.5 (96.4,98.5) 17,375 107.5 (105.9,109.1) 14,835 85.3 (84.0,86.7) Stairs 12,665 37.8 (37.1,38.4) 5,358 33.8 (32.9,34.8) 7,302 41.1 (40.1,42.0) Building 6,079 18.5 (18.1,19.0) 4,781 29.1 (28.2,29.9) 1,297 7.9 (7.5,8.4) Ladder and scaffolding 6,017 18.0 (17.6,18.5) 5,080 31.1 (30.3,32.0) 931 5.4 (5.1,5.8) Fall while being carried 732 2.2 (2.1,2.4) 384 2.3 (2.1,2.5) 348 2.1 (1.9,2.4) Diving or jumping into water 697 2.1 (2.0,2.3) 537 3.3 (3.0,3.5) 159 1.0 (0.8,1.2) and unspecified 35,099 103.5 (102.4,104.6) 13,745 91.1 (89.6,92.7) 21,349 109.4 (107.9,110.9) All 3 175,077 522.8 (520.4,525.3) 84,743 537.1 (533.4,540.7) 90,333 487.7 (484.6,491.0) 1 Age-adjusted rate per 100,000 population. 2 95% confidence interval. 3 The number of NSW residents hospitalised interstate during 2003 2004 was imputed and as a result the sum of submechanisms may not equal the total. Summary Falls were the leading cause of hospitalisation due to injury involving NSW residents between 1999 2000 and 2003 2004, accounting for 34.8% of all injury-related hospitalisations. The yearly hospitalisation rate for fall-related injuries was estimated to have increased significantly by 3.0% per year for males during 1989 1990 to 2003 2004 and increase significantly by 1.1% per year in females. In 2003 2004, there were 36,860 hospitalisations of NSW residents following a fall, giving a hospitalisation rate of 532.4 per 100,000 population. More than two-fifths of those hospitalised following a fall (42.0%) were aged 65 years or older. Falls on the same level, other and unspecified falls, and falls from one level to another were the types of falls that most commonly led to a hospitalisation. Females had higher hospitalisation rates for falls down stairs than males, while males had higher rates of hospitalisation than females for falls from buildings or other structures and ladders and scaffolding. The overall hospitalisation rate for fall-related injuries was 10.1% higher for males than for females.

Falls / 15 Table 5. Number of hospitalisations for falls 1 by age group and cause, NSW, 1999 2000 to 2003 2004 Age group Age group Rank 0-4 5 9 10 14 15 24 25 34 35 44 45 54 55 64 65+ Total 1 One level to another 6,293 2 Same level 2,371 3 Stairs 752 4 Building 642 5 Fall while being carried 516 6 Ladder and scaffolding 56 7 Diving 16 and unspecified 1,231 One level to another 8,229 Same level 4,371 Building 750 Stairs 350 Fall while being carried 79 Ladder and scaffolding 75 Diving 66 and unspecified 1,535 Same level 8,216 One level to another 3,017 Building 464 Stairs 351 Diving 110 Fall while being carried 41 Ladder and scaffolding 32 and unspecified 1,392 Same level 8,329 One level to another 1,535 Building 945 Stairs 833 Ladder and scaffolding 271 Diving 216 Fall while being carried 36 and unspecified 1,525 Same level 5,440 One level to another 1,409 Stairs 1,079 Building 964 Ladder and scaffolding 565 Diving 120 Fall while being carried 19 and unspecified 1,558 Same level 4,541 One level to another 1,543 Stairs 1,272 Ladder and scaffolding 1,015 Building 802 Diving 71 Fall while being carried and unspecified 1,859 Same level 5,352 One level to another 1,614 Stairs 1,598 Ladder and scaffolding 1,289 Building 646 Diving 45 Fall while being carried and unspecified 2,336 Same level 6,580 Stairs 1,538 One level to another 1,436 Ladder and scaffolding 1,270 Building 434 Diving 22 and unspecified 2,690 Same level 36,132 One level to another 7,072 Stairs 4,793 Ladder and scaffolding 1,423 Building 408 Fall while being carried 26 Diving 21 and unspecified 20,868 Same level 81,573 One level to another 32,210 Stairs 12,660 Building 6,079 Ladder and scaffolding 6,011 Fall while being carried 732 Diving 695 and unspecified 35,094 1 Cell sizes represent fewer than five hospitalisations or data have been removed to prevent identification of cell sizes less than five.

16 / NSW INJURY PROFILE: A Review of Injury Hospitalisations During 1989 1990 to 2003 2004 5. Motor vehicle transport This section describes hospitalisations due to unintentional motor vehicle transport-related injuries. WHO classifies the cause of motor vehicle transport-related hospitalisations according to a number of criteria (WHO, 1977; WHO, 1992). Injury hospitalisations due to motor vehicle transport occur in either traffic situations (i.e., occurring on a public highway or street, which includes both the roadway and other land between property lines (NCCH (2000)) or non-traffic situations (i.e., occurring anywhere other than a public highway or street). Motor vehicle transport-related hospitalisations are also classified according to the type of road user involved (e.g. pedestrian, motorcyclists). Motor vehicle transport road-user classes include motor vehicle occupants (drivers and passengers), motorcyclists (riders and passengers), pedal cyclists, and pedestrians. The person who is injured is usually referred to by both the type of traffic situation (i.e., traffic or non-traffic) of the incident and their road-user class (e.g. pedestrian). In NSW, motor vehicle transport-related incidents were the second leading cause of death in NSW during 1998-2002, with 2,765 deaths, giving a mortality rate of 8.5 per 100,000 population (Schmertmann et al, 2004). The lifetime cost of fatal and non-fatal motor vehicle transport-related incidents in NSW has been estimated at $554 million - $61.6 million in direct costs and $493 million in mortality and morbidity costs (Potter-Forbes & Aisbett, 2003). Motor vehicle transport-related injury hospitalisation data from 1989 1990 to 2003 2004 were used to describe the profile of motor vehicle-related hospitalisations of NSW residents. Data from 1999 20000 to 2003 2004 were used in the majority of analyses, except for the trend analysis, which used data from 1989 1990 to 2003 2004. Motor vehicle transport was the second leading cause of injury hospitalisation for the period 1999 2000 to 2003 2004 and accounted for 10.6% of all injury hospitalisations (Table 3). During this period, there were 53,211 injury-related hospitalisations following a motor vehicle transport incident, at an age-adjusted hospitalisation rate of 162.1 per 100,000 population (Table 6). There were approximately 10,642 hospitalisations per year due to motor vehicle transport during 1999 2000 to 2003 2004. The age-adjusted hospitalisation rate for motor vehicle transport-related injury declined in both males and females during the period 1989 1990 to 2003 2004 (Figure 5). For males, the age-adjusted hospitalisation rate decreased significantly by 1.0% per year (95% confidence interval for the decrease: 0.4% to 1.5%). For females, the rate of hospitalisation decreased significantly by 1.9% per year during this period (95% confidence interval for the decrease: 1.4% to 2.5%).