NDARC Technical Report No. 49 (1997)


This  report  analysed  data  on  trends in  opioid overdose mortality between 1979 and 1995 inclusive. Its major aim was determine whether there had been a statistically significant  increase in the rate of such overdose deaths. A secondary aim was to examine trends in the demographic characteristics of persons who died  of  opioid overdose, defined as deaths due to opioid dependence and accidental opiate poisoning.

The number of opioid overdose deaths rose from 70 in1979 to 550 in1995. 78% of the deaths were male. The rate (per million of the adult population aged 15 to 44) significantly increased from 10.7 to 67.0. The increase was more marked among males than females, increasing 6.8 times for males (from 15.3 in 1979 to 104.6 in 1995) and 4.7 times for females (from 5.9 in 1979 to 27.9 in 1995).

Overdose mortality rates varied between different jurisdictions. Among males, New South Wales consistently accounted for around a half of all overdose fatalities, and its rate was almost twice that in Victoria, and three times that in the remaining states. There was a similar pattern among females.

The average age at death increased from 24.2 years in 1979 to 30.1 years in 1995. The highest mortality rate was observed among those aged between 25 and 24 years. The rate of increase in overdose mortality was greater among men and women aged 35 to 44 years and 25 and 34 years than among those aged between 15 and 24 years. An analysis by birth cohort showed that 46% of male overdose deaths and 50% of female overdose deaths in the period occurred among those born between 1960 and 1969. Deaths among persons born between 1950 and 1959 accounted for 38% of male and 33% of female deaths.

Most of the increase in opioid overdose mortality between 1979 and 1995 occurred among persons who initiated heroin use in the 1970s and 1980s. Recentinitiations of heroin use among those born between 1970 and 1979 have begun to be reflected in an increased rate of opioid overdose deaths. If the mortality experience of initiates born since 1970 replicates that of the two earlier cohorts (1950-59 & 1960-69) there will be another epidemic of overdose mortality in 10 to15 years.

Increased heroin purity and the use of alcohol and benzodiazepines in conjunction with heroin have probably played some role in the increased overdose mortality but the magnitude of the contributions awaits further research. Research is required on changes in: the numbers of opioid dependent persons; heroin purity;   risk factors among fatal and non-fatal opiate overdoses; and patterns of heroin and polydrug use among heroin users, especially during the 1991-1995 periods.

These data suggest that more effort needs to be made to recruit older dependent heroin users into methadone treatment which substantially reduces the risks of overdose. Those who decline treatment need to be persuaded of the overdose risks of using opioids in combination with alcohol and benzodiazepines. Consideration also needs to be given to trialling naloxone distribution to heroin users to reduce overdose fatalities, and to trialling “injecting rooms” in some high risk locations.




Wayne Hall, Shane Darke
Date Commenced
26 Sep 1997
Resource Type
Technical Reports