Hospitalised Injuries Victoria, July 1992 - June 1998
Monash University Accident Research Centre Report #160 - 1999
Author: Voula Z. Stathakis
Full report in .pdf format [1.1 MB]
Abstract:
The aim of this report was to describe the all-age epidemiology of hospitalised injuries in Victorian public hospitals. This was done by frequencies and rates for the 6-year period July 1992 to June 1998, and by trend analysis for the 11-year period July 1987 to June 1998. A further aim was to identify potential areas for reducing the frequency and severity of injuries in the community.
Injury admissions to public hospitals recorded in the Victorian Inpatient Minimum Database (VIMD) were selected for analysis. Private hospital data were excluded as their coverage was not complete over the chosen study periods. The effect of Casemix funding on injury frequencies, rates and trends has been discussed and highlighted throughout the report. No adjustments to the data have been made since it was felt that such work would be beyond the scope of this project. Therefore, caution should be exercised with regard to the interpretation of the trends presented in this report.
The major causes of all-age injury hospitalisation were found to be falls, motor vehicle traffic accidents, intentional self-inflicted, hit/struck/crush, cutting/piercing and assault injuries. Falls were the leading cause of injury for all age groups except 15-24 year olds, for whom motor vehicle traffic related injuries were the leading cause of hospital admission. Injury rates were generally increasing over the study period from about age 60 years and upwards. Two other broad peaks in frequency occur in late adolescence/early adult years and the other in older age. In the previous 6-year time period, July 1987 to June 1993, the age groups among whom injury hospitalisations were most frequent were all younger than 35 years (20-24, 15-19, 25-29, 30-34, 10-14). For the time period July 1992 to June 1998, only two of the corresponding age groups were younger than 35 years and the remaining top five were over 65 years (70-74, 20-24, 75-79, 65-69, 25-29). This has significant implications for prevention programs and for the type of care required by the hospitalised injured population. The average annual all-age all-cause injury admission rate was 2,175 per 100,000 persons, representing a 34% increase on the same rate for the previous six-year period, of 1,620 per 100,000.
A log-linear regression model of rate data assuming a Poisson distribution of injuries was used to analyse trends for the 11-year period. It was found that the all-age all-cause injury rate was steadily climbing at a statistically significant rate. Transport injury rates, however, were declining annually by approximately 1.6%. Other significant increasing trends were observed for all-age falls, intentional self-inflicted injuries and assault injuries. The highest estimated annual percentage increase was observed for all-age intentional self-inflicted injuries at 8.6%. Trends for age-specific rates were varied, but most were significantly increasing or decreasing for selected injury causes. Notable changes included the decline in fire/burn injuries among under 5 year olds (-6.0% p.a.), increases in self-inflicted injuries among 15-24 (7.2% p.a.) and 25-64 (10.3% p.a.) year olds and most strikingly, the increasing trend for medical injuries among over 65 year olds with an annual percentage increase 13.2%.
Several recommendations have been made in the report including the necessity for further research and the review of progress towards present injury prevention targets and the setting of new targets. Recommendations have also been made with regard to data and coding issues relating to ICD9/ICD10 coding and the VIMD. The total cost and burden on society of injury needs to be quantified to facilitate the appropriate allocation of resources towards the development and implementation of effective injury prevention strategies.
Executive Summary
This report describes the all-age epidemiology of hospitalised injuries in Victorian public hospitals by frequencies and rates for the 6-year period July 1992 to June 1998 and by trend analysis for the 11-year period July 1987 to June 1998. Graphical representations for the entire VIMD collection, 11 years (July 1987 to June 1998) are also provided. A change of funding arrangements (Casemix) for public hospitals was introduced in July 1993. Any possible Casemix effects, covering the period 1992/93 to 1994/95, are highlighted on each graph, while the data itself has not been modified in any way. It appears that the Casemix effect for hospital admissions dissipates from the period 1995/96 onwards, however, in many instances, rates are still higher than pre-Casemix levels.
The section of this report relating to frequencies and rates of hospitalised injury is largely an update of the report Hospitalised Injuries Victoria, July 1987 to June 1993, (Watt1).
MAJOR CAUSES OF INJURY, FREQUENCY, RATES AND MALE/FEMALE DISTRIBUTION
- The major causes of all-age injury were falls, motor vehicle traffic accidents, intentional self-inflicted, hit/struck/crush, cutting/piercing and intentional assault injuries. Falls ranked first for all age groups except for 15-24 year olds, for whom motor vehicle traffic accidents headed the list ahead of falls at fourth position.
- The average annual frequency of admission to hospital for injury was 98,692, representing an average annual all-age all-cause rate of 2,175 per 100,000 persons, a 34% increase on the same rate for the previous six-year period (July 1987 to June 1993) of 1,620 per 100,000 persons.
- This time period is characterised by increasing rates of injury from about 60 years upwards and two broad peaks in injury frequency by 5-year age groups: one in late adolescence/early adult years, and the other in older age.
- Comparisons with the previous 6-year time period (July 1987-June 1993), revealed that injury rates were now generally higher, most noticeably from about age 55 years. Similarly, frequencies for each age group were higher particularly from about age 55 years onwards.
- These changes in frequency are due to the combined effect of increased rates and demographic changes. In the previous time period, July 1987 to June 1993, the age groups among whom injury hospitalisations were most frequent were all younger than 35 years (20-24, 15-19, 25-29, 30-34, 10-14). For the time period July 1992 to June 1998, only two of the corresponding age groups were younger than 35 years and the remaining top five were over 65 years (70-74, 20-24, 75-79, 65-69, 25-29).
- Females had a lower rate of admission for almost all causes of injury, the notable exceptions being various types of falls, horse riding and some means of self-inflicted injury.
TRENDS FOR INJURY RATES 11 YEARS (JULY 1987 TO JUNE 1998)
- The all-cause all-age injury rate was steadily climbing at a statistically significant rate (6.1% p.a.); average difference in rates over the 11-year period was +77 per 100,000. Excluding transport injuries, the increase in the all-cause all-age rate becomes slightly greater with an average difference of +88 per 100,000 and estimated annual percentage change of 7.0%.
- Transport injury rates were decreasing overall, but not significantly, the average difference over the 11-year period was 4 per 100,000 or 1.6% p.a.
- Other significantly increasing trends were observed for all-age falls, intentional self-inflicted and assault injuries. The highest estimated annual percentage change was observed for intentional self-inflicted injuries at 8.6%.
- The 0-4 year old all-cause injury rate was increasing significantly (2.9% p.a.), as was the trend for the 0-4 year old poisoning rate (3.2% p.a.). Significantly decreasing trends for this age group were recorded for near drowning, fire/burns, scalds, dog bites and motor vehicle traffic injury rates. There was also a non-significant increase in falls.
- For the 5-14 year age group, injuries overall appeared to be increasing. This significant increase also extended to falls, except for playground fall rates, which were increasing but not significantly. Decreasing rates were observed for pedestrian injuries (significant) and bicycle injury rates although not at a significant level.
- The all-cause injury rate for 15-24 year olds was increasing significantly at 2.3% p.a., as were intentional self-inflicted (7.2% p.a.) and assault injuries (3.5% p.a.), both displaying marked increases. Motor vehicle traffic and motorcycle rates (-4.3% p.a.) were declining while a non-significant increase was found for sports related fall and collision injury rates.
- For the 25-64 year age group, the all-cause injury rate was increasing significantly with an estimated annual increase of 6.4%, along with cutting/piercing (3.7% p.a.) and assault injuries (4.1% p.a.). The highest significant increase, however, was seen for intentional self-inflicted injury rates with an estimated annual percentage increase of 10.3%.
- Significant increases in the all-cause (8.7% p.a.), falls (4.6% p.a.), medical (13.2% p.a.) and intentional self-inflicted (4.0%) injury rates were observed for persons 65 years and over. Pedestrian injury rates had fallen significantly (-2.9% p.a.).
RECOMMENDATIONS
1. Use should be made of the data presented here to contribute to the review of progress towards meeting injury prevention targets and the setting of new targets for the next 5 years.
2. Future hospitalised injury reports should include public and private hospital data.
3. A formal detailed descriptive epidemiological study should be undertaken to determine the validity of data and reasons for the high and rising rate of medical injuries. This may provide useful information for the evaluation of medical procedures and hospital quality control programs.
4. A formal study should be undertaken to thoroughly determine the effect of Casemix funding on admission policy in public hospitals and how this might affect any analysis of the VIMD for injury prevention purposes.
5. Casemix effect modelling for public hospital admissions is strongly needed for accurate trend analysis studies.
6. The Department of Human Services should continue to monitor the quality of all the data being included in the VIMD (now known as the VAED), and make the results of their monitoring readily available. This is particularly important considering the changeover to ICD10 coding. Particular attention should be paid to place and activity codes for injury cases.
7. These coding validation studies need to be repeated on a regular basis for comparison purposes for all coding fields including E-codes at all character levels, i.e. the entire code.
8. Existing coding systems should be improved to facilitate identification of all product-related injury, and more specific identification of occupational and sports injury.
9. Consumer product safety seems to have been relatively neglected in research and prevention of injury. There are a sufficient number of injuries apparently resulting from consumer products to warrant further major effort in this area.
10. The data presented in this report indicate the need for further investment in injury prevention.
11. Linkage of health surveillance systems is recommended to clearly enable the identification of sequences and causes of injury events, to provide more detail for each case, and to improve data reliability.
12. Hospital admission data should be used to inform the development of an Injury Cost and Consequences Model.
Sponsoring Organisation(s): This project was funded by the Victorian Health Promotion Foundation.