Do AR-DRGs adequately describe the trauma patient episode in New South Wales, Australia?
Abstract: The use of Diagnosis Related Groups (DRGs) may not be an accurate tool to provide reimbursement for trauma services. This study aimed to determine whether Australian Refined Diagnosis Related Groups (AR-DRGs) adequately describe the trauma patient episode and to identify AR-DRG groupings where reimbursement was not commensurate with actual cost. The AR-DRG allocated costs and actual costs of a sample of 206 trauma patient episodes were reviewed during a three-month period. Of the AR-DRG groups identified in the patient episodes, 62.8% were not commensurate with actual cost incurred, equating to an overall loss of $113,921 from under-funded acute trauma patient episodes over a three-month period. Assault-related penetrating trauma, traffic-related and sport-related incidents were all inadequately reimbursed using AR-DRGs compared with the actual cost of treatment. Cases involving female patients, patients aged 45 years or less and those with moderate injuries were similarly underfunded. AR-DRGs are not adequate to describe the extent of injuries experienced by trauma patients and there is a need to investigate alternative funding models for trauma services.

Keywords (MeSH): Trauma; Injuries; Healthcare Costs; Hospital Economics; Diagnosis Related Groups; Treatment Costs; Australia.
Article Type: Report
Subject: Trauma centers (Services)
Trauma centers (Information management)
Medical care, Cost of (Measurement)
Authors: Curtis, Kate
Mitchell, Rebecca
Dickson, Cara
Black, Deborah
Lam, Mary
Pub Date: 02/01/2011
Publication: Name: Health Information Management Journal Publisher: Health Information Management Association of Australia Ltd. Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2011 Health Information Management Association of Australia Ltd. ISSN: 1833-3583
Issue: Date: Feb, 2011 Source Volume: 40 Source Issue: 1
Topic: Event Code: 360 Services information; 260 General services Canadian Subject Form: Trauma centres; Trauma centres Computer Subject: Company systems management
Geographic: Geographic Scope: Australia Geographic Name: New South Wales Geographic Code: 8AUST Australia; 8AUNS New South Wales
Accession Number: 252944956
Full Text: Introduction

To reimburse hospitals for treatment provided, healthcare funding models attempt to estimate the cost of hospital treatment for specific types of clinical procedures. An episode-based funding model (also known as casemix-based funding) has been adopted for acute healthcare services in New South Wales (NSW), Australia (NSW Health Department 2010). Episode-based funding means that a health care facility is allocated a predetermined financial payment for each type of patient episode (Curtis, Bollard & Dickson 2002). In acute care in Australia, types of patient episodes are defined using Australian Refined Diagnosis-Related Groups (AR-DRGs). This type of funding model provides a financial incentive for hospitals to avoid providing unnecessary services or extending a patient's hospital stay (MacKenzie et al. 1991).

For some health conditions, such as those encountered in rehabilitation or palliative care, AR-DRGs have not been found to be good indicators of the 'true' types of patient episodes (Eager & Harvey 2001). The same can be said for trauma patients, where AR-DRGs have not always been found to be appropriate indicators of the care that is required for these patients (Campbell et al. 1995).

Approximately 2,300 individuals who are severely injured (Injury Severity Score [ISS] >15) receive specialist care at trauma centres in NSW each year (NSW Institute of Trauma and Injury Management 2010), yet these patients represent a small proportion (20-30%) of the overall burden of trauma to trauma centres. For example, at St George Public Hospital (SGPH) in South East Sydney, there are around 250 patient admissions with an ISS>15 each year. There are, however, a further 750 trauma patients that require admission to hospital (South Eastern Sydney Illawarra Area Health Service 2008). This scenario is representative of most other trauma centres (see for example Westmead Hospital Trauma Service 2008; Liverpool Hospital Trauma Service 2008; American College of Surgeons 2008). Severely injured trauma patients often have complex care needs as they often have multiple injuries, in more than one body region (Jacobs & Jacobs 1992). For this reason the determination of appropriate AR-DRGs derived from the clinical coding process can be difficult (Curtis, Bollard & Dickson 2002).

In an environment of increasing healthcare costs and competition for finite resources, economic data relating to the cost of injury and illness are becoming integral factors in the development of health services policy. Therefore, it is important to have appropriately costed hospital treatment/service utilisation funding models and subsequent resource allocation to avoid under-resourcing of the hospital sector. Funding for the acute inpatient portion of trauma patients' care which takes place outside of the emergency department (ED) and intensive care unit (ICU) is based on AR-DRG cost weights, derived from NSW Health Program and Product Data Collection (PPDC) costing results. Accordingly, it is necessary to establish whether AR-DRGs adequately describe the acute trauma patient episode.


This pilot study aims to determine whether AR-DRGs adequately describe the in-hospital trauma patient episode and to identify AR-DRG groupings where reimbursement is not commensurate with actual cost.


The pilot study method used for this research is described in detail elsewhere (Curtis et al. 2009) and is outlined here.

Patient identification

All patients presenting to the ED of SGPH during a three-month period (November 2006 to January 2007) were assessed by the duty Trauma Case Manager (TCM) to determine if they fulfilled trauma criteria (Figure 1). Inter-hospital transfer patients were identified by the TCM and the nurse-in-charge of each ward and included in the study.

For the patients that met trauma criteria (n=206), information was collected by the duty case manager during the course of their daily patient round and medical record review for the hospital trauma database. This information routinely includes: patient demographics, the mechanism of injury, injury severity, injuries, length of stay and outcome. The patient's medical record number and date of admission were used to link to episode level costing and utilisation data produced by the hospital's patient costing system Access database.


Casemix costings

Financial information was obtained from the hospital's casemix unit to calculate the cost of each patient's admission. Patient costing was conducted in accordance with 2006-07 NSW PPDC Standards AR-DRG version 5.0 (NSW Health Department 2007). The Trendstar Decision Support System, which employs both clinical costing and cost modelling methodologies in the cost allocation process, calculated costs associated with each episode.

Financial information was extracted from the hospital's general ledger. Cost centres were identified as either overhead or patient care cost centres (PCCC). Suitable overhead allocation statistics were selected for each overhead cost centre, in accordance with guidelines in NSW PPDC. Examples of overhead statistics include: staffing full-time equivalents, weighted floor space and weighted sterile supply items. Overhead costs were allocated to the PCCCs and ultimately to the patients utilising the resources of that PCCC. Inpatient fractions were utilised for each PCCC where total patient utilisation was not available.

The clinical information in Trendstar is based on patient data from the hospital Patient Administration System, which is accessed via an interface with the Health Information Exchange (HIE). The HIE is NSW Health's data warehouse and acts as a repository for a number of data collections. The additional clinical information, such as theatre, prostheses, pharmacy and allied health, was sourced from a variety of interfaced, non-interfaced and paper-based systems.

Data management and analysis

The patients' injury severity was coded using the Abbreviated Injury Scale (AIS), the most widely used anatomic injury severity scale in the world (Association for the Advancement of Automotive Medicine 1990). It is used in epidemiological research, trauma centre studies predicting survival probability, patient outcome evaluation and healthcare systems research. To determine a patient's Injury Severity Score (ISS), the coder allocates a code and score to each of the patient's injuries. The higher the AIS score allocated, the more serious the injury (Association for the Advancement of Automotive Medicine 1990). Patients were grouped according to their ISS in an effort to place patients with similar severity of injury for comparison. An ISS greater than 15 was chosen as the accepted standard of severe injury. An ISS of 9-15 was considered moderate to serious, and less than nine minor to moderate (Holbrook, Hoyt & Anderson 2001).

Each patient was allocated a unique identifier to allow de-identification. The hospital trauma and casemix datasets were combined, entered into and analysed with SPSS, version 18 (SPSS 2009). Descriptive statistics were performed on each patient demographic element.


Males accounted for just less than three-quarters (73.8%) of trauma admissions and the mean age of patients was 44 years (sd 22.3; range 0-92). Around half (49.5%) of patients had a minor to moderate ISS. Just over half (51.5%) of the trauma presentations were traffic-related (Table 1). The median length of stay (LOS) was three days (range 1-126), while the median AR-DRG allocated LOS was 3.7 days (range 1.2-127.6).

For the AR-DRGs allocated to the 206 trauma patients, the average in-hospital AR-DRG reimbursement was $12,751 (range: $1,370-$147,988), while the average actual cost was $15,104 (range: $782-$195,022). The average trauma patient LOS was 10.1 days (sd 18.3) and the overall AR-DRG allocated LOS was 8.9 days (sd 14.1). Patients who were aged greater than 45 years, severely injured (ISS>15) or whose cause of injury was: pedestrian, motor cycle crash, self harm or a fall from <1 metre had the largest positive variance in average LOS. The average cost varied for different patient characteristics. Procedures that were performed on females, younger patients, and less severely injured patients were all inadequately funded using AR-DRGs compared with the actual cost of treatment. Traffic-related incidents (excluding pedestrians), assault-related penetrating trauma, sport-related and incidents involving self-harm were all under-funded using the current cost weights and price allocated to AR-DRGs to fund the cost of treatment (Table 2).

Of the 43 AR-DRG groupings identified in this group of patients, there were 27 (62.8%) AR-DRG groupings where reimbursement was not found to be commensurate with actual cost incurred, totalling -$400,912; and 16 (37.2%) where the type of treatment was overfunded, totalling $286,991. This equated to an overall loss of $113,921 from under-funded trauma treatment costs. Under-funded procedures commonly involved treatment for head injuries, chest trauma, musculoskeletal disorders, injuries to extremities, and severe burns (Figure 2). For example, a patient with an AR-DRG of E66A (Major Chest Trauma A >69 + Cc) who had three fractured ribs, a fractured sternum, an upper limb laceration, soft tissue injuries of the ankle and cervical spine had a treatment reimbursement of $7,193 but had an actual treatment cost of $94,617. Also, a patient with DRG B79Z (skull fractures) had also sustained a small subdural haemorrhage, sub-arachnoid haemorrhage, cerebral contusions and a scalp haematoma. The DRG cost allocation was $3,953 and the actual cost was $26,705. This patient was most likely incorrectly coded.


The allocation of AR-DRGs for a trauma patient can be a complicated process as often these patients have multiple injuries (Taheri et al. 1999). This study has shown that AR-DRGs do not adequately represent the trauma patient episode in this pilot sample. This is most likely due to the complex nature of trauma that usually results in multiple injuries not represented by a defined event or single AR-DRG, such as for an isolated limb fracture. Studies conducted in relation to the costing of various healthcare services in Australia do not specifically include or describe the full scope of trauma (Hyder, Meddings & Bachani 2009; Aucar & Hicks 2005; Antioch & Walsh 2000). However, findings from this pilot study are supported by the international literature, which has consistently demonstrated that the AR-DRG method of describing the complex trauma patient's injuries is insufficient and underestimates the true cost of the patient's treatment requirements (Aucar & Hicks 2005, Grotz et al. 2004; Pape et al. 2009; Jacobs & Jacobs 1992).

In the current study, the average AR-DRG cost increased as injury severity increased. However, less severely injured patients appeared to be under-funded using AR-DRGs compared with those that were severely injured. This result could partially be explained by the limitations of the ISS system, as it only allows the scorer to include one injury per body region (e.g. head, chest, abdomen), which may lead to an underestimation of the patient's overall anatomic injury severity, because the patient's most severe injuries may not be represented if they have more than one severe injury in a single body region. The New Injury Severity Score (NISS) (Association for the Advancement of Automotive Medicine 1990) scores the patient's three most severe injuries, regardless of body region. While the NISS was not available for this analysis, the NISS has been reported to more accurately represent the severity of injury and subsequent risk of mortality (Osler, Baker & Long 1997, Brenneman et al. 2009). This could further be explained by the variance in LOS. The actual LOS was on average three days longer than the AR-DRG allocated LOS for severely injured patients.

Assault-related penetrating trauma, burns, injuries to pedal- and motor-cyclists and motor vehicle occupants were all under-funded in the current study, they also had the largest variance in AR-DRG LOS. These injury mechanisms have also been found to be costly in other studies in the United States (Clancy et al. 1994, Dimick et al. 1986, Siegel et al. 1993). Industrial injuries, falls and sport-related injuries all appeared to be over-funded in the current study, although there was no difference in compensability status or injury severity. Around one-third of falls for each type of height involved head injuries which could account for some of the funding imbalance for this injury mechanism.

The example of the patient with DRG B79Z highlights that clinical coding errors can occur and contribute to funding variance. There are no national standards for auditing clinical coding, and no state-wide audit was performed on NSW data in 2005-06, although with the introduction of Episode Funding one of the 'works in progress' is commissioning external audits of coding, costing and reporting standards and application (NSW Health Department 2010). Hospitals in NSW perform formal audits on The International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification (ICD-10-AM) coded data at a local level. Data edits are monitored regularly and consistent errors are identified and rectified by individual hospitals.

At the study hospital, ICD-10-AM coding of trauma patients is undertaken with the assistance of the trauma database. The trauma database contains data obtained by TCM on their daily rounds and is edited and maintained by a dedicated data manager. This was instigated to improve the accuracy of coding the complex trauma patient (Curtis, Bollard & Dickson. 2002).


Cost containment and effective financial management has become increasingly important for hospitals in recent years. These findings suggest that in relation to the complex patient, improved financial modelling is required. While several trauma centres in Australia have taken the lead in cost containment and efficacy of trauma patient care using strategies, such as trauma case management (Curtis et al. 2006), clinical pathways (Sesperez et al. 2001) and dedicated admitting teams (Ursic et al. 2009), without accurate and adequate resource funding in the first instance, many of these initiatives, while clinically significant, may not impact greatly on the overall financial management of the trauma casemix.

This study did have some limitations. As it was a pilot study, the findings may have limited generalisability to other trauma centres. However, overall AR-DRGs have been found to underestimate the true direct cost of trauma in other locations (Aucar & Hicks 2005, Grotz et al. 2004, Pape et al. 2009, Jacobs & Jacobs 1992). There could have been seasonal variance in the type of injury mechanisms experienced as the study only assessed the cost of trauma cases during a short three month period. However, the proportion of injury mechanisms is similar to the NSW state annual profile (NSW Institute of Trauma and Injury Management 2010).

This study has highlighted the fact that AR-DRGs are not adequate to describe the extent of injuries experienced by trauma patients and has shown that there is a need to now develop adequate funding models for trauma. In order to establish such a model the financial and clinical burden of trauma casemix in NSW's designated major trauma centres (NSW Health Department 2009) needs to be clearly described. There is a need for a larger, multicentre study to estimate the direct cost of the acute treatment of trauma and to identify factors, such as demographics and circumstances of the injury event, that are associated with higher treatment costs across trauma centres.


R Mitchell was supported by an ARC-linkage postdoctoral fellowship (LP0990057).


American College of Surgeons (2008). National trauma bank annual report 2005 [Online]. Available at: (accessed 4 Sept 2008).

Antioch, K. and Walsh, M. (2000). Funding issues for Victorian hospitals: the risk-adjusted vision beyond casemix funding. Australian Health Review 23: 145-153.

Association for the Advancement of Automotive Medicine (1990). The Abbreviated Injury Scale 1990 Revision. Des Plaines, IL, Association for the Advancement of Automotive Medicine.

Aucar, J. A. and Hicks, L. L. (2005). Economic modeling comparing trauma and general surgery reimbursement. The American Journal of Surgery 190: 951-960.

Brenneman, F., Boulanger, B., McLellan, B. and Redelmeier, D. (2009). Measuring injury severity: time for a change? Journal of Trauma 44: 580-582.

Campbell, A., Vittinghoff, E., Morabito, D., Paine, P., Shagoury, C., Praetz, P., Grey, D., McAninch, J. and Schecter, W. (1995).

Trauma centres in a managed care environment. Journal of Trauma Injury Infection and Critical Care 39: 246-253.

Clancy, T., Misick, L., Covinton, D., Churchill, M. and Maxwell, J. (1994). The financial impact of intentional violence on community hospitals. The Journal of Trauma 37: 1-4.

Curtis, K., Bollard, L. and Dickson, C. (2002). Coding errors and the trauma patient--is nursing case management the solution? Australian Health Review 25: 73-80.

Curtis, K., Dickson, C., Black, D. and Nau, T. (2009). The cost and compensability of trauma patients. Australian Health Review 33: 84-92.

Curtis, K., Zou, Y., Morris, R. and Black, D. (2006). Trauma case management: improving patient outcomes. Injury 37: 626-632.

Dimick, A., Potts, L., Charles, E., Wayne, J. and Reed, I. (1986). The cost of burn care and implications for the future on quality of care. The Journal of Trauma 26: 260-265.

Eager, K. and Harvey, R. (2001). Australia not ready to implement 'best practice' health financing model. Health Cover 10: 52-56. Grotz, M., Schwermann, T., Lefering, R., Ruchholtz, S., Schulenburg,

J. M. G. V. D., Krettek, C. and Pape, H. C. (2004). G-DRG reimbursement for polytrauma patients--a comparison with the comprehensive hospital costs using the german trauma registry. DRG-entlohnung beim polytrauma--Ein vergleich mit den tatsachlichen krankenhauskosten anhand des DGU-traumaregisters 107: 68-75.

Holbrook, T., Hoyt, D. and Anderson, J. (2001). The impact of major in-hospital complications on functional outcome and quality of life after trauma. Journal of Trauma Injury Infection and Critical Care 50: 91-95.

Hyder, A., Meddings, D. and Bachani, A. (2009). MENTOR-VIP: piloting a global mentoring program for injury and violence prevention. Academic Medicine, 84: 793-796.

Jacobs, B. and Jacobs, L. (1992). The effect of the new trauma DRGs on reimbursement. Journal of Trauma Injury Infection and Critical Care 33: 495-503.

Liverpool Hospital Trauma Service (2008). Liverpool Hospital Trauma Site. Trauma Registry Statistics 1995-1999 [Online]. Available at: (accessed 4 Sept 2008).

MacKenzie, E., Steinwachs, D., Ramzy, A., Ashworth, J. and Shankar, B. (1991). Trauma case mix and hospital payment: the potential for refining DRGs. HSR: Health Services Research 26: 5-26.

NSW Health Department (2007). 2006-2007 NSW program and product data collection standards incorporating subacute and non-acute and national minimum dataset for mental health establishments data collection requirements. North Sydney, NSW Health Department.

NSW Health Department (2009). Selected specialty and statewide services plans Number Six. NSW Trauma Services. North Sydney, NSW Health Department.

NSW Health Department (2010). NSW Health Episode Funding Policy 2008/2009 [Online]. Available at: au/policies/pd/2008/PD2008_063.html (accessed 16 March 2010).

NSW Institute of Trauma and Injury Management (2010). The NSW Trauma Registry profile of serious to critical injuries 2007. Sydney, NSW Institute of Trauma and Injury Management.

Osler, T., Baker, S. and Long, W. (1997). A modification of the injury severity score that both improves accuracy and simplifies scoring. Journal of Trauma Injury Infection and Critical Care 43: 922-925.

Pape, H. C., Neugebauer, E., Ridley, S. A., Chiara, O., Nielsen, T. G. and Christensen, M. C. (2009). Cost-drivers in acute treatment of severe trauma in Europe: a systematic review of literature. European Journal of Trauma and Emergency Surgery 35: 61-66. Sesperez, J., Wilson, S., Jalaludin, B., Seger, M. and Sugrue, M. (2001). Trauma case management and clinical pathways: prospective evaluation of their effect on selected patient outcomes in five key trauma conditions. Journal of Trauma 50: 643-649.

Siegel, J., Mason-Gonzalez, S., Dischinger, P., Read, K., Cushing, B., Badellino, M., Goodarzi, S., Smialek, J., Heatfield, B., Robinson, R., Parkinson, K., Hill, W., Jackson, J., Livingston, D., Bents, F. and Clark, C. (1993). Causes and costs of injuries in multiple trauma patients requiring extrication from motor vehicle crashes. The Journal of Trauma 35: 920-931.

South Eastern Sydney Illawarra Area Health Service (2008). SESIAHS Trauma Service Overview [Online]. Available at: (accessed 4 Sept 2008].

SPSS (2009). SPSS: Statistical Program for the Social Sciences version 18.0. Chicago, SPSS.

Taheri, P, Butz, D., Watts, C., Giffes, L. and Greenfield, L. (1999). Trauma services: a profit center? Journal of the American College of Surgeons 188: 349-354.

Ursic, C., Curtis, K., Zou, Y. and Black, D. (2009). Improved trauma patient outcomes after implementation of a dedicated trauma admitting service. Injury 40: 99-103.

Westmead Hospital Trauma Service (2008). Westmead Hospital Trauma Site. Trauma Statistics 2003 Annual Report [Online]. Available at: registry_data.php (accessed 4 Sept 2008).

Corresponding author

Kate Curtis RN, PhD

Clinical Nurse Consultant

Trauma Service, St George Hospital

Gray St, Kogarah NSW 2217


Ph +612 91132686

F: +612 91133974


Clinical Associate Professor, Sydney Nursing School, University of Sydney

Visiting Senior Fellow, NSW Injury Risk Management Research Centre, and Conjoint Associate Professor, St George Clinical School, Faculty of Medicine, University of New South Wales Honorary Professorial Fellow, George Institute for Global Health

Cara Dickson BSc(Hons)

Performance Analyst

St George Hospital

Gray Street, Kogarah NSW 2217


Rebecca Mitchell PhD

Research Fellow

Department of Aviation & New South Wales Injury Risk

Management Research Centre

University of New South Wales

Sydney NSW 2052


Deborah Black PhD

Professor and Associate Dean, Staff Development

Faculty of Health Sciences

University of Sydney

PO Box 170

Lidcombe NSW 1825


Mary Lam PhD


Faculty of Health Sciences

University of Sydney

PO Box 170

Lidcombe NSW 1825

Table 1: Patient characteristics of the study group, St George
Hospital, November 2006 to January 2007 (n = 206)

PATIENT CHARACTERISTICS          (n=206)     %

Male                                152     73.8
Female                               54     26.2

Age group
<16                                   9      4.4
16-25                                47     22.8
26-45                                59     28.6
46-65                                49     23.8
>65                                  42     20.4

Injury Severity Score
ISS <9                              102     49.5
ISS 9-15                             42     20.4
ISS >15                              62     30.1

Injury mechanism
Motor vehicle crash--driver          48     23.3
Motor vehicle crash--passenger       20      9.7
Motor cycle crash                    16      7.8
Pedestrian                           13      6.3
Pedal cyclist                         9      4.4
Assault/stabbing/gunshot wound       17      8.3
Fall < 1 metre                       24     11.7
Fall 1-5 metres                      31     15.0
Fall >5 metres                        5      2.4
Sport-related                        14      6.8
Burns                                 5      2.4
Industrial work                       3      1.5
Self-harm                             1      0.5

Table 2: Average DRG cost, average actual cost and difference,
average LOS, DRG, allocated LOS and ISS by patient characteristics,
St George Hospital, November 2006 to January 2007 (n=206)

                                AVERAGE   AVERAGE
                                  DRG      ACTUAL      DIFFERENCE
PATIENT                           COST      COST
CHARACTERISTICS                    $         $         $         %

Male                            14,778    14,258        520      3.5
Female                           6,629    17,390    -10,761   -162.3

Age group
<16                              2,808    13,818    -11,010   -392.1
16-25                            7,510    12,783     -5,273    -70.2
26-45                           12,094    20,554     -8,460    -70.0
46-65                           16,825     9,201      7,624     45.3
>65                             16,841    16,837          4      0.0

Injury Severity Score
ISS <9                           5,902    14,275     -8,373   -141.9
ISS 9-15                         9,167    17,937     -8,770    -95.7
ISS >15                         26,221    14,651     11,570     44.1
Injury mechanism
Motor vehicle crash--driver     14,537    15,835     -1,298     -8.9
Motor vehicle crash--passenger   2,664     6,040     -3,376   -126.7
Motor cycle crash                7,873    16,827     -8,954   -113.7
Pedestrian                      17,378    16,786        592      3.4
Pedal cyclist                    6,031    27,340    -21,309   -353.3
Assault/stabbing/gunshot wound   4,810    34,761    -29,951   -622.7
Fall < 1 metre                  13,359    10,998      2,361     17.7
Fall 1-5 metres                 14,363     9,833      4,530     31.5
Fall >5 metres                  29,468    10,533     18,935     64.3
Sport-related                   20,918    10,858     10,060     48.1
Burns                            2,859    17,801    -14,942   -522.6
Industrial work                 50,406    15,119     35,287     70.0
Self-harm                        1,670     2,106       -436    -26.1

PATIENT                           AVERAGE     ALLOCATED     AVERAGE
CHARACTERISTICS                  LOS (sd)     LOS (sd)     ISS (sd)

Male                            11.3 (20.0)  10.0 (15.5)  12.1 (10.8)
Female                           6.7 (11.6)   5.7  (7.9)   9.5 (10.4)

Age group
<16                              2.4  (1.4)   2.7  (1.8)   4.6  (2.7)
16-25                            3.5  (2.9)   5.5  (6.4)  10.3 (11.5)
26-45                            8.4 (14.4)   7.2 (11.5)  11.2 (10.2)
46-65                           14.6 (28.0)  12.5 (21.6)  12.8 (12.4)
>65                             16.1 (17.9)  12.3 (12.4)  13.0  (8.9)

Injury Severity Score
ISS <9                           4.5 (11.6)   4.8  (8.4)   3.7  (1.9)
ISS 9-15                         7.5  (8.9)   7.1  (8.2)  10.8  (2.3)
ISS >15                         21.1 (25.8)  16.8 (20.2)  24.7  (9.7)
Injury mechanism
Motor vehicle crash--driver      9.6 (17.4)   9.7 (13.6)  10.3 (10.7)
Motor vehicle crash--passenger   2.7  (1.9)   2.8  (2.6)   5.2  (4.2)
Motor cycle crash                8.9  (7.2)   6.5  (6.3)  20.9 (16.1)
Pedestrian                      23.8 (30.9)  14.6 (18.3)  10.1  (7.7)
Pedal cyclist                    3.3  (2.9)   4.9  (4.7)   8.3  (6.5)
Assault/stabbing/gunshot wound   3.6  (3.0)   4.2  (2.3)   8.4  (7.3)
Fall < 1 metre                  13.2 (17.3)   9.4 (10.1)  14.5  (9.4)
Fall 1-5 metres                 10.8 (20.4)   9.1 (10.5)  12.7  (9.7)
Fall >5 metres                  13.0 (10.6)  17.8 (15.9)  12.0  (9.6)
Sport-related                   10.1 (14.9)  11.1 (15.2)  11.7 (10.0)
Burns                            5. 6 (8.6)   2.4  (0.9)  10.2 (10.0)
Industrial work                 44.7 (70.4)  43.6 (72.8)  21.3 (30.9)
Self-harm                        4.0    (0)   1.5    (0)   1.0    (0)

Note: DRG = diagnosis related group; LOS = length
of stay; ISS = injury severity score
Gale Copyright: Copyright 2011 Gale, Cengage Learning. All rights reserved.