About the datasets

• Australian Bureau of Statistics - Death Unit Record
• Victorian Coroner's Facilitation System
• Victorian Admitted Episodes Dataset
• Victorian Emergency Minimum Dataset
• Original Victorian Injury Surveillance System Database

VISU has progressively established substantial holdings of statewide and national injury data at three levels of severity - deaths (Australia ), hospital admissions and emergency department presentations (Victoria).

Databases of incident cases held and utilised by VISU

*Deaths: additional data sources: 
(1) The National Coronial Information System, will supersede the VCFS and incorporate Victorian Coronial data.

Dataset details

Australian Bureau of Statistics - Death Unit Record File

The Australian Bureau of Statistics (ABS) death unit record file (DURF)  consists of information supplied by State Registrars of Births, Deaths and Marriages. Each death registered in Australia is classified by the ABS according to the World Health Organisation (WHO) International Classification of Diseases (ICD) coding system. 

Coroners' Facilitation System

The Coroner's Facilitation System is a database containing all Victorian unnatural deaths for the period July 1989 to June 1995.  Data are collated from the findings of the State Coroner. These include those that were unexpected, unnatural or violent, or which resulted from accident or injury.

Victorian Public Hospital Admissions due to injury (VAED)

Source: Victorian Admitted Episodes Dataset (VAED, formerly VIMD), 1987/88 - Dec. 2006 (19.5 years)

The Victorian Admitted Episodes Dataset is a collection of data on acute hospital admitted patient information. To fund public hospitals equitably under the casemix system and to maintain morbidity data on all Victorians, the Department of Human Services requires a minimum dataset which accurately reflects health service activities and meets the requirements of the Health Act 1958 (General Amendment 1988). These data collection requirements include:

• Causes, effects and nature of illness among Victorians
• Determinants of good and ill-health
• Utilisation of health services in Victoria.

To meet these objectives, all public and private acute hospitals, including acute facilities in rehabilitation, extended care institutions and day procedure centres, are required to report the relevant minimum dataset of admitted patient activity. These data are transmitted to Allegiance Systems Australia and are compiled into the Victorian Admitted Episodes Dataset (VAED). The release of data is permitted only under certain terms and conditions. There are limitations placed upon the specificity of the data released to maintain the rights of patients to privacy in relation to personal details.

This document containing this summary along with further detailed information about the VAED can be found at www.health.vic.gov.au/hdss/vaed/index.htm.

Monash University Accident Research Centre (MUARC) has been acquiring some of the data collected on a year by year basis. The data we have purchased is considered "external cause data", as it contains only cases where an external cause is coded for the patient. Each hospital collects data at admission and at separation. This is entered by the hospital and checked and corrected by the hospital.

Readmissions within 30 days are excluded. Each record in the database represents an episode of care, and not necessarily one incident. A patient may be transferred between and within hospitals for various episodes of care and will therefore be represented by more than one record. This is estimated to account for 10% of the database.

The VAED held by MUARC contains more than 2.4 million records for the 20.5-year period from July 1987 to December 2007. Variables provided in standard output include: age, sex, injury cause, location, activity, nature of main injury, body region, intent, birth place, address postcode and separation type.

Victorian Emergency Minimum Dataset (VEMD)

The VEMD records details of injuries treated at the Emergency Departments of the 38 Victorian public hospitals with 24-hour emergency departments. The total number of cases on the database to date (December 2008) is in excess of 3,148,000 cases.

The 38 hospitals currently contributing data to the collection, along with starting dates are shown below:

Variables collected include: age, sex, injury cause, location, activity, nature of main injury, body region, human intent, birth place, preferred language, address postcode, departure status, referral on departure and a 250 character text description of the injury event.

Limitations of VEMD data 

Identification of specific data, especially products and activities, is reliant on the item of interest being recorded in the 100 character injury narrative.  Hence the number of recorded cases for some injuries is likely to be an underestimate of the true incidence of emergency department presentations eg. only 45% of sporting injuries (identified by the 'Activity when injured' code) recorded in 1996 and 1997 noted the type of sporting activity being undertaken eg, netball, tennis.

Original VISS Database

The Original VISS database records details of injuries treated at the Emergency Departments of the six VISS hospitals. The data is based on information provided by the injured persons (or their parent) and the attending doctor. The total number of cases in the database is approximately 173,330. Data collection in each hospital began at different times:

Caveats

Because VISS data are collected from clusters of hospitals rather than from the whole state, they are most useful in  determining proportions of factors such as locations or products associated with injuries of various severities.  It is therefore important to bear in mind several points in drawing conclusions from the data:

• Only cases attending the VISS hospitals are recorded in our database.  Many injuries are treated at other Melbourne hospitals or clinics.
• Comparisons of injury rates are difficult because we do not have information about how many people are actually at risk for a particular injury.  That is, we do not know how many people are exposed to specific hazards, for what time period, how many of these are male or female, what their ages are, or where the exposure to these hazards occurs (e.g. in the home, at school, on the road and so on). 
• Only injury deaths that actually occurred within the Emergency Department or patients who were dead on arrival (DOA) are reported in the VISS database. Some hospitals do not include DOAs in their hospital statistics. Deaths that occurred after the patient left the Emergency Department are not identified by VISS and would have been coded as an admission or a transfer. During the time of the VISS collection approximately 100 Victorian children and 1600 adults died from injury each year.