NDARC Technical Report No. 163 (2003)
Aims: This report compares trends in the characteristics of male and female accidental drug-induced deaths among Australians over the period 1997 to 2001, including demographics, underlying and contributing cause of death and drugs noted in toxic quantities. This report also aims to review the current method for determining the number of accidental opioid deaths in Australia.
Method: Cause of death (COD) data, coded according the International Statistical Classification of Diseases and Related Problems – 10th revision (ICD-10), was obtained from the Australian Bureau of Statistics (ABS) for the years 1997 to 2001. The following variables were assessed: sex, age groups, state of residence and underlying and contributing causes of death. The ABS definition of drug-induced death (including accidental) was utilised. COD data coded according to ICD-9 was also obtained from the ABS for the years 1997 to 1998 to permit opioid-induced death mapping analyses between the two versions.
(a) Accidental drug-induced deaths
From 1997 to 2001 there were 5539 accidental drug-induced deaths. Accidental deaths were the leading cause (68%) of drug-induced deaths registered in 2001. In 2001, the rate of accidental drug-induced deaths continued to decrease from 1999, where it peaked at 93.3 to 45.9 deaths per million people aged 15 years and over.
(b) Demographic characteristics
The majority of accidental drug-induced deaths were likely to be male (74%). Males and females aged 25-34 made up the largest proportion of accidental deaths over time. The majority (around 70%) of male and female accidental drug-induced deaths occurred in NSW and Victoria.
(c) Underlying causes of death
The most frequently assigned ICD-10 codes for underlying cause of death for accidental deaths were: accidental poisoning due to multiple drug use (X44 - 29%), accidental poisoning due to narcotics (X42 - 23%), opioid drug use disorder (F11 - 23%) and multiple drug use disorder (F19 - 17%).
(d) Drugs noted in toxic quantities at death
The most common drugs noted in toxic quantities at time of death was the same for males and females. However there was a different pattern in the proportion of drug poisoning: opioids (77% vs. 60%), sedative-hypnotics (22% vs. 30%) and antidepressants (8% vs. 22%).
(e) Contributing causes of death
For contributing causes of death, mood disorders were present in 3.3% of males and 8.7% of females, and acute hepatitis C was present in 6.8% of male and 5.2% of female deaths. Other contributing diseases were not commonly mentioned.
(f) Classifying accidental opioid deaths
Three lines of evidence supported the inclusion of multiple drug use disorder (F19) cross-classified with opioids (opioid poisoning and opioid use disorder) when calculating the number of accidental opioid deaths: its equivalent code multiple drug poisoning (X44) cross classified with opioids was already included; multiple drug use disorder (F19) often had opioid poisoning as a contributing cause in the majority of deaths; and the mapping analysis showed that opioid codes in ICD-9 mapped to multiple drug use disorder cross-classified with opioids in ICD-10. While the revised calculation of accidental opioid deaths increased the total number of accidental drug-induced deaths attributed to opioids in Australia, the general trends did not change. In 2001 the rate of accidental opioid deaths in Australia was 40 per million persons aged 15 to 44 years, a 67% decrease compared to the rate in 1999 (which was 122 per million).
There has been a steady increase and resulting decrease in the number and rate of accidental opioid drug-induced deaths among Australians between 1997, 1999 and 2001. This is consistent with data from other sources that suggest coinciding changes in the supply of heroin in Australia over this period. Males aged 25-35 continued to predominate accidental drug-induced deaths in states that are populous for opioid-related use. A high incidence of acute hepatitis C was demonstrated. Multiple drug deaths were common. Mono-intoxication was likely in opioid-related deaths and very unlikely in other drug-deaths. Benzodiazepines featured in a notable proportion of accidental druginduced deaths. Finally, accidental drug-related deaths, particularly opioids, are a preventable public health problem. Consistent, accurate and timely monitoring of the changes in the patterns of these deaths over time provides an opportunity to inform key stakeholders so as to contribute to appropriate responses to this issue.
Citation: Barker, B. and Degenhardt, L. (2003) Accidental drug-induced deaths in Australia 1997 - 2001, Sydney: National Drug and Alcohol Research Centre.