Alcohol and other drug use - feature analysis:
People with a history of specialist alcohol and other drug treatment services who have died
Web article
Last updated:
Citation
AIHW (Australian Institute of Health and Welfare) (2026) People with a history of specialist alcohol and other drug treatment services who have died, AIHW, Australian Government, accessed 17 June 2026.
This article is part of Alcohol and other drug use - feature analysis
- People who received specialist Alcohol and Other Drug Treatment Services in their last year of life
- People with a history of specialist alcohol and other drug treatment services who have died This page
Content advice
- The AIHW respectfully acknowledges the people who have died and are described in this article. Each statistic represents a person – with a family and community grieving for their loss.
- The information presented in this article contains information and data on alcohol or other drugs and deaths by suicide. If you or someone you know needs help, contact the National Alcohol and Other Drug Hotline on 1800 250 015 or Lifeline on 13 11 14. Go to the support services page for a list of support services.
- The AIHW supports the use of the Mindframe guidelines on responsible, accurate and safe reporting on alcohol and other drug use. Please consider these guidelines when reporting on these topics.
Key findings
Between 1 July 2012 and 30 June 2023, almost 37,700 people with a history of specialist alcohol and other drug (AOD) treatment services for their own drug use, died – around 27,200 males and 10,500 females. Among this population:
- Around 2,800 people died while receiving ongoing treatment – about 5 deaths each week.
- The median age at death was 51 years.
- Mortality was almost 4 times as high as the expected mortality in the general Australian population.
- The average number of potential years of life lost per death was 26 years.
- 3 in 5 deaths were from potentially avoidable causes.
- Accidental poisoning, suicide and liver disease were the top 3 leading underlying causes of death.
- ‘Policing and justice’ was the most mentioned psychosocial factor in accidental poisoning deaths, while ‘personal history of self-harm’ was the most common among deaths by suicide.
- While a greater number of people who died were treated for alcohol only, a higher proportion of people treated for heroin or pharmaceuticals died.
Deaths data are a vital measure of a population’s health, providing insights into disease patterns and factors contributing to mortality across different groups and over time (AIHW 2023). This article examines national mortality patterns and causes of death among people with a history of publicly funded, specialist alcohol and other drug (AOD) treatment services (for more information on specialist AOD treatment services see Alcohol and other drug treatment services).
Research has shown that people who received AOD treatment or those that received specialist AOD treatment services in Australia are particularly vulnerable cohorts, experiencing higher mortality compared to the general population (Tisdale et al. 2021; Bista et al. 2021) and lower estimated life expectancies (Lewer et al. 2020).
The findings from this analysis provide insights into the context and scale of excess mortality among people with a history of specialist AOD treatment services. Findings are intended to inform targeted prevention activities and support improvements in service delivery and outcomes for clients.
Definitions
Data presented in this article are about people with a history of publicly funded, specialist alcohol and other drug (AOD) treatment services who died between 1 July 2012 to 30 June 2023. It has been sourced from the NACS linked dataset. See Technical notes for more information.
- The focus of this report is on people with a history of specialist AOD treatment services, who received treatment at least once for their own drug use, and died (‘AODTS clients’). This means people who received treatment and/or assistance solely for someone else’s drug use are not included. This is because their client profile is likely to differ and warrants separate analyses.
- People aged 10 and over are included. This is consistent with the scope of the Alcohol and Other Drug Treatment Services National Minimum Data Set (AODTS NMDS), which only includes persons of this age range.
- The term ‘non-AODTS’ population (that is, persons who have not received specialist AOD treatment services) and ‘general Australian population’ (that is, all Australians aged 10 or over, including AODTS clients) have been used in this article for ease of reading. See Technical notes for further information.
- The phrases ‘publicly funded specialist AOD treatment services’ and ‘specialist AOD treatment services’ have been used interchangeably throughout this article for ease of reading.
Data coverage limitations
The AODTS NMDS data collection introduced a statistical linkage key in 2012–13 which enabled the number of clients receiving treatment to be counted (that is, at the person-level) (AIHW 2024). Therefore 2012–13 is the earliest financial year of data available in this analysis. This means for our follow-up time periods, particularly among people who died ‘over 2 years’ since their last AOD treatment episode, the length of historical data available will depend on a client’s year of death.
For example:
- Among people who died in 2015–16, to be included in the ‘over 2 years’ follow-up period, they would have had to complete their treatment episode in 2013–14. This means there is only two observable years of historical data to draw from.
- Comparing this to AODTS clients who died in 2022–23, they would have had to complete their treatment episode before 2020–21. This means there are nine observable years of historical data to draw from as a client could have received a treatment episode anytime between 2012–13 and 2020–21.
Therefore, caution should be taken when comparing follow-up periods across time and to one another, particularly before 2016–17, as earlier year counts are likely to be underestimated. This means an observed increase in deaths over time among people with a history of specialist AOD treatment services, for example, ‘more than 2 years’ from their last treatment episode, may not reflect an actual increase in deaths.
Comparing results to other publications
The age distribution of people who receive specialist AOD treatment services are very different to that of the general population. Some of the results from this analysis may not be directly comparable to that of the general Australian population.
As noted above, the period of follow-up data is limited by the availability of data. As the AODTS NMDS data collection continues and more deaths data are available for linking, the results of future analysis exploring similar measures could change.
What is the profile of people with a history of specialist AOD treatment services who have died?
-
Among the almost 37,700 people with a history of specialist AOD treatment services who died between 1 July 2012 and 30 June 2023, nearly half were aged 40–59.
Some of the results from this section may not be directly comparable to that of the general Australian population. For more information see Data coverage limitations.
Between 1 July 2012 to 30 June 2023, about 37,700 people with a history of specialist alcohol and other drug (AOD) treatment services for their own drug use died – around 27,200 males (72%) and 10,500 (28%) females (Figure 1 and Table S1). The median age at death across the study period was 51 years of age (the same for males and 52 years of age for females).
Age at death
Of the nearly 37,700 people with a history of specialist AOD treatment services who died between 2012–13 to 2022–23:
- The highest proportion of deaths were among clients aged 50–59 (24%).
- This was followed by clients aged 40–49 (23%) at death and 60–69 at death (17%).
- The lowest proportion of deaths were among AODTS clients aged 10–19 (1.0%).
Use of specialist AOD treatment services prior to death
Over the study period (2012–13 to 2022–23):
- Around 1 in 14 people died while receiving an AOD treatment service (7.4%).
- 1 in 3 clients received their last AOD treatment episode up to a year before their death (33%).
- Around 1 in 6 clients received their last AOD treatment episode between 1 and 2 years before their death (16%).
- Just over 2 in 5 people received their last AOD treatment episode more than 2 years before their death (43%).
This was similarly observed among male and female AODTS clients.
Use of specialist AOD treatment services prior to death, by age group
Between 1 July 2012 and 30 June 2023:
- In each age group and for both males and females (excluding those aged 10–19 at death), a high proportion of AODTS clients received their last treatment episode more than 2 years before their death (34–58% for males; 28–62% for females; 33–59% for persons).
- Among male and female AODTS clients aged 10–19 at death, over 1 in 4 clients had received their last treatment episode up to 6 months before their death (29% and 28% respectively).
- Among male and female AODTS clients aged 70+ at death, over 1 in 2 clients had received their last treatment episode over 2 years before their death.
Figure 1: Deaths among people with a history of alcohol and other drug treatment services (for their own drug use), by sex and age-group, 2012–13 to 2022–23
Dashboard displaying an interactive vertical bar graph with the option to filter between different statistical measures (number or per cent) and time from last treatment episode to death
Potential years of life lost
Potential years of life lost (PYLL) is a measure of premature mortality. If dying before the age of 75 is considered premature, then a person who died before the age of 51 would have lost 24 potential years of life (for more information see Deaths in Australia).
Caution should be used when interpreting time series data within, and differences between, follow-up periods as they are not comparable. They are presented to show the characteristics of the people who have died within each follow-up period and therefore should not be compared. See Data coverage limitations for more information.
-
On average 26 years of potential life were lost
among each person who received specialist AOD treatment services and died between 1 July 2012 and 30 June 2023.
Using the age of 75 as the life expectancy cut-off, there were about 921,000 PYLL among people with a history of specialist AOD treatment services who died over the study period (Table S2). This is an average of 26 PYLL per client that died and represents 10.8% of the total PYLL due to premature mortality in Australia between 2012–13 and 2022–23.
Potential years of life lost, by sex
Between 1 July 2012 to 30 June 2023, out of the total PYLL (that is, including both AODTS clients and the non-AODTS population):
- Males with a history of specialist AOD treatment services accounted for 12.4% of all PYLL.
- Females with a history of specialist AOD treatment services accounted for 8.0% of all PYLL (Table S2).
These sizable contributions to PYLL reflect both the low median age at death of males and females with a history of specialist AOD treatment and the predominance of largely preventable underlying causes of death (see Age at death and How many deaths were potentially avoidable). Findings may not be directly comparable to PYLL in the general Australian population due to differences in the underlying age distribution of people who receive specialist AOD treatment services.
In each year of the study period, male AODTS clients accounted for a majority (72–74%) of PYLL, reflecting in part, the larger proportion of male clients who died compared to female clients (see What is the profile of people with a history of specialist AOD treatment services who have died).
Potential years of life lost, by service use prior to death
In 2022–23:
- People who received their last treatment episode up to a year before their death lost on average around 28 years of potential life and contributed one third of PYLL over the study period (33% or about 42,400 PYLL). For more information on this group, see People who received specialist Alcohol and Other Drug Treatment Services in their last year of life.
- AODTS clients who received their last treatment episode more than a year before their death accounted for around two thirds of PYLL (around 67% or 84,500 PYLL). Findings related to these follow-up periods should be interpreted with caution as distributions observed are likely to change if more time was available to observe all clients until death.
How do deaths among people with a history of specialist AOD treatment services compare to the general Australian population?
A standardised mortality ratio (SMR) is calculated by dividing the number of observed deaths over the number of expected deaths. It shows the magnitude of difference between what the study population actually experienced and what it would have experienced if it had the same (in this case) age and sex specific rates as the general Australian population (reference population). That is, it helps provide an understanding on excess mortality in the population of interest.
SMRs and age-specific rate ratios (ASRR) presented below have been calculated, using death as the outcome, from the start of each client’s first ever treatment episode in the study period. This differs to data presented on follow-up time elsewhere in this article, which counts from a client’s last ever treatment episode before death. A small number of clients were excluded when calculating SMRs due to certain criteria not being met yet may have been included (where relevant) in other sections of this article. For example, if a client had an unknown date of birth or negative person-time they were excluded when calculating SMRs and ASRRs. This means that the number of observed deaths presented in this section will differ from other death counts presented in this article. For more detailed information on how SMRs and ASRRs were calculated, see Technical notes.
-
Excess deaths
Over the study period, mortality among people with a history of specialist AOD treatment services was almost 4 times as high as the expected mortality of the general Australian population.
Between 1 July 2012 and 30 June 2023, mortality (due to any underlying cause) among people with a history of specialist AOD treatment services for their own drug use was:
- Almost 4 times as high as the expected mortality in the general Australian population (that is, all Australians aged 10 or over, including AODTS clients).
- 3.7 times as high as the expected mortality in the general Australian population for males.
- 4.4 times as high as the expected mortality in the general Australian population for females (Table 1).
Among people with a history of specialist AOD treatment services who died, mortality was 20% higher among females compared to males (Table 1).
| Sex | Observed deaths | Expected deaths(a) | Standardised mortality ratio (95% confidence interval) |
|---|---|---|---|
| Males | 26,457 | 7,238 | 3.7 (3.6–3.7) |
| Females | 10,148 | 2,281 | 4.4 (4.4–4.5) |
| Persons (total) | 36,605 | 9,519 | 3.8 (3.8–3.9) |
(a) expected deaths refers to the deaths that would have been expected if it had the same age and specific rates of the general Australian population. For more information refer above to What are standardised mortality ratios and how have they been calculated or Technical notes.
Source: AIHW NACS linked dataset 2025; Table S3.
Age–specific rate ratios
In each age-group, male and female AODTS clients consistently had higher mortality compared to the expected mortality in the general Australian population, indicating substantial excess mortality (see glossary in Technical notes for further information) (Table S3.1).
Across the study period:
- Among males, AODTS clients aged 10–19 had the largest relative difference in mortality with an age-specific rate ratio (ASRR) of 6.9, indicating that the age-specific rate was almost 7 times as high compared to the general Australian population for this sex and age group. This was followed by 30–39 year olds with an ASRR of 5.1 and 40–49 year olds with than ASRR of 4.7.
- Among females, AODTS clients aged 30–39 had the largest relative difference in mortality with an ASRR of 8.2, indicating that the age-specific rate was over 8 times as high compared to females of the same age-group in the general Australian population. This was followed by 40–49 year olds with an ASRR of 6.9 and 20–29 year olds with an ASRR of 6.8.
- Female AODTS clients had higher ASRRs in each age group compared to male AODTS clients, excluding ages 10–19 and 80–99. This indicates a greater relative excess of deaths among females receiving specialist AOD treatment compared with females in the general Australian population, than male AODTS clients.
What were the most common underlying causes of death?
The leading underlying causes of death presented in this section are classified using an AIHW-modified version of A method for deriving leading causes of death (Becker et al. (2006)). For more information see Technical notes – classifying causes of death.
Caution should be used when interpreting time series data within, and differences between, follow-up periods as they are not comparable. They are presented to show the characteristics of the people who have died within each follow-up period and therefore should not be compared. See Data coverage and limitations for more information.
-
Whether a client received their last AOD treatment service one year before, or more than 2 years before their death, accidental poisoning was the most common underlying cause of death.
Among AODTS clients who died between 1 July 2012 to 30 June 2023:
- Accidental poisoning (16%) was the most common underlying cause of death, followed by suicide and liver disease (both 12%) (Table S4 and Figure S2). Accidental poisoning involves a person unintentionally poisoning themselves by exposure to noxious substances, which includes accidental drug overdose (see Accidental poisoning for more information).
- These underlying causes of death consistently remained in the top three in each follow-up period (albeit liver disease was ranked second followed by suicide for clients who received their last AOD treatment episode ‘up to 6 months’ before their death) (Table S4.1 and Figure 2).
- Among male clients, accidental poisoning (16%), suicide (13%) and liver disease (11%) were also the most common underlying causes of death across the study period.
- In each follow-up period, these same causes of death remained in the top three, however among clients who received their last AOD treatment episode ‘more than 2 years’ before their death, coronary heart disease (9.5%) replaced liver disease (8.1%) as the third most common underlying cause of death.
- Among female clients, accidental poisoning (15%) was the most common underlying cause of death, followed by liver disease (14%) and suicide (11%) across the study period.
- In each follow-up period, the same causes of death and their rankings remained consistent except for clients who received their last AOD treatment episode ‘up to 6 months’ before their death, as liver disease was the top ranked cause (20%) followed by accidental poisoning (18%) and suicide (13%).
Specialist AODTS clients among all deaths, by underlying cause
As a proportion of all deaths across the study period (that is, including both AODTS clients and the non-AODTS population by underlying cause, where age at death was 10 years or older) over the study period, AODTS clients accounted for:
- 2 in every 5 deaths due to accidental poisoning (40%).
- Almost 1 in every 3 deaths due to mental and behavioural disorders due to psychoactive substance use (28%).
- 1 in every 5 deaths due to liver disease (20%).
- Around 1 in every 7 deaths due to suicide (14%) (Table S4 and Figure 2).
Male AODTS clients accounted for 42% of all male accidental poisoning deaths in Australia, 27% of all male deaths due to mental and behavioural disorders due to psychoactive substance use, 20% of all male deaths due to liver disease and 14% of all male deaths due to suicide (Table S4).
Female AODTS clients accounted for 35% of all female accidental poisoning deaths, 29% of all female deaths due to mental and behavioural disorders due to psychoactive substance use, 19% of all female liver disease deaths and 14% of all female suicide deaths (Table S4).
Figure 2: Deaths among people with a history of specialist alcohol and other drug treatment services (for their own drug use), by sex and top 10 causes of death, 2012–13 to 2022–23
Dashboard displaying an interactive horizontal bar graph with the option to filter between different statistical measures (number and proportion), sex and time from last treatment episode to death
How many deaths were potentially avoidable?
A death can be classified as potentially avoidable due to causes such as diabetes, certain cancers (for example prostate and breast) suicide and accidental poisoning. For more information, see A guide to the potentially avoidable deaths indicator in Australia and Australian Health Performance Framework.
Caution should be used when interpreting time series data within, and differences between, follow-up periods as they are not comparable. They are presented to show the characteristics of the people who have died within each follow-up period and therefore should not be compared. See Data coverage limitations for more information.
-
Out of all potentially avoidable deaths across the study period, people with a history of specialist AOD treatment services accounted for around 1 in every 13 deaths.
Over the 11-year study period (1 July 2012 to 30 June 2023), there were around 21,100 potentially avoidable deaths (PADs) among people aged 10-74 with a history of specialist AOD treatment services for their own drug use (Table S5). This equates to 3 in 5 deaths among AODTS clients being potentially avoidable (60%). Out of all PADs across the study period, AODTS clients accounted for about 1 in every 13 deaths (Table S5).
Among people with a history of specialist AOD treatment services who died:
- Clients had a higher proportion of potentially avoidable deaths (57–63% of all causes of death) compared to the non-AODTS population (43–47%) in each year of the study period – particularly among females (Table S5 and Figure S3).
- In 2022–23, more than half (around 57%) of all deaths were potentially avoidable, equating to around 8 potentially avoidable deaths per day.
The top 3 underlying causes of PADs among people with a history of specialist AOD treatment services who died were:
- Accidental poisoning which accounted for around 1 in every 6 deaths (17%, 5,895 deaths).
- Suicide which accounted for 1 in every 7 deaths (13%, 4,688).
- Coronary heart disease which accounted for 1 in every 15 deaths (6.7%, 2339).
These were the same underlying causes of PADs among male AODTS clients, while chronic obstructive pulmonary disease was the third ranked potentially avoidable cause of death among female AODTS clients.
Figure 3: People with a history of specialist alcohol and other drug treatment services (for their own drug use) who died, by sex and proportions of potentially avoidable deaths, 2012–13 to 2022–23
Dashboard displaying an interactive vertical bar graph with the option to filter between sex and each financial year of the study period.
Most common psychosocial factors for select causes of death
Non-medical factors, such as those related to housing or education, can play a significant role in a person’s life and death. Since 2017, information on some of these factors (described as psychosocial factors) has been collected for coroner-referred deaths by the Australian Bureau of Statistics (for more information see Psychosocial risk factors as they related to coroner-referred deaths in Australia). Understanding these factors can highlight a need to implement or improve social policies or interventions for specific groups of people. For more information on AIHW groupings of psychosocial factors and examples, see Psychosocial factors contribute to death.
-
1 in every 3 deaths
mentioned 'policing and justice' among AODTS clients who died by accidental poisoning.
-
Almost 1 in every 2 deaths
mentioned 'personal history of self-harm' among AODTS clients who died by suicide.
Accidental poisoning
Between 1 January 2017 and 30 June 2023, out of all deaths due to accidental poisoning among people with a history of specialist AOD treatment services, around 4 in 10 deaths in coronial findings mentioned at least 1 psychosocial factor (40%, or 1,741 deaths) (Table S6). The average number of psychosocial factors for accidental poisoning deaths among AODTS clients who died was 1.7.
Out of all accidental poisoning deaths among people with a history of specialist AOD treatment services involving 1 or more psychosocial factors:
- 1 in 3 deaths (33%) mentioned ‘policing and justice’ (for example imprisonment or other legal circumstances).
- Over 1 in 5 deaths (22%) mentioned ‘personal history of self-harm’ (for example suicidal and non-suicidal deliberate self-injury).
- Over 1 in 5 deaths (21%) mentioned ‘support systems’ (for example estrangement from family or social isolation).
These were the same most mentioned psychosocial factors for both males and females, albeit differing rankings. ‘Policing and justice’ were proportionally higher among males (37% and 22%, respectively), while ‘personal history of self-harm’ (28% and 19%, respectively) and ‘support systems’ were more commonly mentioned among female AODTS clients (27% and 19%, respectively) (Table S6 and Figure 4).
Suicide
Between 1 January 2017 and 30 June 2023, out of all deaths due to suicide among people with a history of specialist AOD treatment services, about 8 in 10 deaths in coronial findings mentioned at least 1 psychosocial factor (81% or 2,960 deaths). The average number of psychosocial factors for deaths by suicide among AODTS clients that died was 2.3.
Out of all suicide deaths among people with a history of specialist AOD treatment services involving one or more psychosocial factors:
- Over 2 in 5 deaths (44%) mentioned ‘personal history of self-harm’.
- Around 2 in 5 deaths (41%) mentioned ‘suicidal ideation’ (for example thoughts of suicide with or without intent).
- Almost 2 in 5 deaths (39%) mentioned ‘intimate partner’ (for example separation/divorce or domestic violence).
These were the same top ranked psychosocial factors mentioned among female AODTS clients who died (56%, 38%, 34%, respectively). Among male AODTS clients, ‘suicide ideation’ was the most common psychosocial factor mentioned (42%), followed by ‘intimate partner’ (40%) and ‘personal history of self-harm’ (40%) (Table S6 and Figure 4).
While it is well established that accidental poisoning and suicide deaths are more likely to mention psychosocial factors, due to the complex nature of circumstances surrounding the death, the proportions of deaths by these causes among people with a history of specialist AOD treatment services are higher (see What were the most common underlying causes of death) and the leading psychosocial factors for these select causes are different to those in the broader Australian population (see Deaths in Australia, Psychosocial factors contribute to death).
Figure 4: Deaths among people with a history of specialist alcohol and other drug treatment services (for their own drug use), by sex, selected underlying cause of death and top psychosocial factors, 1 January 2017 to 30 June 2023
Dashboard displaying an interactive horizontal bar chart divided by sex, with the option to filter between different statistical measures (number and per cent) and selected underlying causes of death
What were the most common principal drug(s) of concern?
A person can receive multiple treatment episodes over their lifetime, each for the same or a different Principal Drug of Concern (PDOC). For example, a person may receive treatment for alcohol use in one treatment episode, followed by another treatment episode for heroin use. These treatment episodes could occur in the same year, and/or in different years. For more information on AODTS client deaths and PDOC see People who received specialist Alcohol and Other Drug Treatment services in their last year of life or Technical notes for PDOC groupings.
The information below does not indicate whether a client's PDOC was related to, caused by, or present in their death. Examining deaths involving alcohol and other drugs among AODTS clients is yet to be undertaken.
-
While a greater number of people with a history of specialist AOD treatment services who died were treated for alcohol, a higher proportion of people treated for heroin or pharmaceuticals died.
Treatment history and mortality, by PDOC
Over the study period, almost 790,000 people received specialist AOD treatment services for their own drug use. Almost 70% of these clients received treatment for either alcohol only (around 247,000 people), cannabis only (almost 181,000 people) or meth/amphetamines only (around 108,000 people). This, in part, is reflected in the large number of deaths for alcohol and cannabis among people with a history of specialist AOD treatment services.
Of the nearly 37,700 people with a history of specialist AOD treatment services who died over the study period:
- Over 18,900 people had received treatment for alcohol only (50%).
- About 2,900 people had received treatment for cannabis only (7.6%).
- About 2,900 people had received treatment for alcohol and another drug (7.6%) (Table S8 and Figure 5).
In contrast, when looking at the proportion of deaths out of all clients with a treatment history for a particular PDOC(s) over the study period, the most prominent treatment histories included pharmaceuticals or heroin (with, or without another drug).
Between 1 July 2012 and 30 June 2023, out of all clients who received specialist AOD treatment services for:
- Pharmaceuticals only, around 1 in 10 people died (10% or 2,665 deaths).
- Heroin and another drug, around 1 in 11 people died (9.1% or 1,177 deaths).
- Heroin only, around 1 in 11 people died (8.8% or 1,612 deaths).
- Pharmaceuticals and another drug, around 1 in 12 people died (8.5% or 1,496 deaths) (Table S8 and Figure 5).
This PDOC treatment history pattern was similarly observed among males and females, albeit differing rankings among female clients receiving specialist AOD treatment services. These findings serve to highlight mortality differences among PDOC treatment histories and warrants further analyses.
Figure 5: Deaths among people with a history of specialist alcohol and other drug treatment services (for their own drug use), by sex and principal drug of concern, 1 January 2017 to 30 June 2023
Dashboard displaying an interactive horizontal bar graph with the option to filter between different statistical measures (number and per cent) and sex
AIHW (Australian Institute of Health and Welfare) (2023) Deaths in Australia, AIHW, Australian Government, accessed 10 February 2026.
AIHW (2024) Alcohol and other drug treatment services NMDS, 2023–24; Quality Statement [website], accessed 6 February 2026.
AIHW (2025) Alcohol and other drug treatment services in Australia annual report 2023–24, AIHW, Australian Government, accessed 25 March 2026.
Australian Bureau of Statistics, Measuring Australia’s excess mortality by remoteness areas during the COVID-19 pandemic until December 2023, ABS, Australian Government, accessed 21 May 2026.
Bista S, Nathan S, Rawstorne P, Palmer K, Ferry M, Williams M, Hayen A (2021) ‘Mortality among young people seeking residential treatment for problematic drug and alcohol use: A data linkage study’, Drug and Alcohol Dependence, 1:228, DOI: 10.1016/j.drugalcdep.2021.109030. Epub 2021 Sep 17. PMID: 34592701.
Lewer D, Jones NR, Hickman M, Nielsen S, Degenhardt L (2020), ‘Life expectancy of people who are dependent on opioids: A cohort study in New South Wales, Australia’, Journal of Psychiatric Research, 130: 435–440, DOI: 10.1016/j.jpsychires.2020.08.013.
Tisdale C, De Andrade D, Leung J, Chiu Vivian, Hides L (2021) ‘Utilising data linkage to describe and explore mortality among a retrospective cohort of individuals admitted to residential substance use treatment’, Australasian Professional Society on Alcohol and other Drugs, 40(7): 1202–1206, DOI: 10.1111/dar.13279
NACS linked dataset
For this analysis, data from the NACS linked dataset have been restricted to the study period 1 July 2012 to 30 June 2023. For more information see Data source.
Data presentation
Throughout this article:
- Values presented in the columns and rows of tables may not sum to the totals shown due to missing and not stated values, as well as rounding.
- Totals reported include missing and not stated values, unless otherwise noted.
- The percentages shown in the tables are calculated excluding the missing and not stated values, unless indicated otherwise.
- Percentage distributions may not sum to 100 due to rounding.
The Australian Institute of Health and Welfare (AIHW) has strict confidentiality policies which have their basis in section 29 of the Australian Institute of Health and Welfare Act 1987 (AIHW Act) and the Privacy Act 1988 (Privacy Act). Data may be suppressed for either confidentiality reasons or where estimates are based on small numbers, resulting in low reliability. Information that results in attribute disclosure will also be suppressed.
Deaths data and statistics
Data for the number of deaths is sourced from the NACS linked dataset. Deaths in this report are presented by year of occurrence.
In the version of the NDI database used in the NACS, data were considered final for those deaths registered prior to 2021, revised for deaths registered in 2021 and preliminary for deaths registered in 2022 and 2023.
The number of deaths in this report are not comparable to other AIHW publications, such as Deaths in Australia, due to:
- Different data sources (National Death Index compared with the AIHW National Mortality Database).
- Year of death occurrence compared with year of death registration.
Glossary
- The term ‘non-AODTS population’ refers to persons who have not received specialist AOD treatment services over the study period (whether for their own drug use or someone else’s). That is, the population includes people who may use drugs but not attended a specialist AOD treatment service and people who do not use drugs.
- The term ‘general Australian population’ is inclusive of all Australians aged 10 or over. This means it includes people who may have attended a specialist AOD treatment service, a private clinic, people who use drugs and not attended either of these services, and people who do not use drugs.
- The term excess mortality generally refers to the difference between the total (observed) number of deaths and the expected number of deaths over the same time period (ABS 2025).
Standardised mortality ratios
Follow-up time (in person years)
Follow-up time (in person-years) was determined by counting from a client’s first treatment episode in the study period, up until date of death or censor date (whichever came first). The censor date was set to the end of our study period (30 June 2023). The first treatment episode was chosen based on the assumption that time in treatment is not risk-free and therefore we treated the period from that start date as a continuous period of risk regardless of whether a client was actively in treatment or between episodes.
To ensure accurate estimation of standardised mortality ratios, the ageing of a client over the study period was accounted for by calculating follow-up time based on age boundaries. This means a client may have contributed person-time to multiple age groups as they aged over the study period. Person-years were then aggregated based on sex and 10-year age groupings.
Clients who had the same death (or censor) date and first treatment start date were included.
Exclusions
To ensure accuracy and correct estimation of SMRs and ASRRs, clients may have been excluded due to:
- Inconsistent or incomplete date of birth.
- Negative person-time (for example due to a treatment episode following a date of death).
- Age below 10 at the time of a client’s treatment start date.
- This means the total number of (observed) deaths used to calculate SMRs and ASRRs will differ to other death counts published in this article.
Method type
Standardised mortality ratios were calculated using the indirect age standardisation method where observed deaths were divided by expected deaths. The analysis was conducted using the SAS procedure stdrate.
Reference population
The number of deaths and total population counts for Australia between calendar years 2012 and 2023 were obtained from the Australian Bureau of Statistics. Mean yearly death rates for each age group during this period was used as the age-specific rates (ASRs) in the reference population.
Client counts
The number of clients presented in this report will not necessarily match totals presented in other alcohol and other drug reports. This is because the linkage process can identify people through SLKs of lower quality than what is accepted within the AODTS NMDS annual reporting.
Count of clients who received specialist AOD treatment and died are in scope if the client had their last specialist AOD treatment episode and died anytime between 1 July 2012 to 30 June 2023. The death could have taken place:
- Within 90 days prior to the end date of their last treatment episode.
- The same day as the end date of their last treatment episode.
- Following the end date of their last treatment episode.
Sex
Sex is based on the sex reported to Medicare. If this was not available, the most frequently reported sex in the AODTS NMDS for a client was assigned. If there was not a dominant proportion, the sex of a client was based on the sex that was recorded for their most recent AODTS episode. Sex was not available for some records.
Using these datasets, it is important to note that it is not known if the people completing these records interpreted sex to mean sex at birth or gender identity. While there were instances where sex was recorded as ‘other’ or ‘missing’ in the NACS linked dataset, due to small numbers, these records have not been disaggregated into their own category.
Age at death
For age-based analyses in this report, if a client’s month and year of their date of birth could not be determined, they were excluded from the client count. Exact date of birth information is not available in the NACS dataset, with only month and year of birth included from base data sources. Date of birth is approximated as the 15th day of the birth month for all clients. Where available, the month and year of birth information is sourced from the Medicare Consumer Directory (MCD), and from either the SHSC or AODTS for clients who have not died. Date of birth information is not available for some clients due to data quality concerns.
Age at death is only able to be calculated for people who either link to the MCD or to the SHSC and AODTS datasets. This excludes groups such as infants (who do not have an MCD listing as they are not yet enrolled in Medicare), overseas visitors and temporary residents, asylum seekers and refugees. More information about who is eligible for Medicare is provided at Enrolling in Medicare.
Follow-up time (last treatment episode to death)
Some information in this article is presented according to the amount of time since a client’s last closed treatment episode to their death date:
- ‘During treatment’ refers to people who died within 90 days prior to their last treatment episode (based on either start or end date of treatment episode), or any person who died between the start date and end date of their last treatment episode (inclusive). This classification does not imply any relationship between the treatment provided and the individual’s death. In some cases, a specialist AOD treatment service agency may keep an episode open for up to 3 months (90 days) while there has been inconsistent, minimal or no contact with a client (for more information see Alcohol and Other Drug Treatment Services National Minimum Data Set Data Collection Manual 2024–25).
- ‘Up to 6 months’ refers to people who died within 182 days (inclusive) of the end date of their last treatment episode.
- ‘Between 6–12 months’ refers to people who died between 182 days and 365 days (inclusive) of the end date of their last treatment episode.
- ‘Between 1–2’ years refers to people who died between 365 and 730 days (inclusive) of the end date of their last treatment episode.
- ‘Over 2 years’ refers to people who died over 730 days from the end date of their last treatment episode.
Principal drug(s) of concern
Principle drug/s of concern (PDOC) have been grouped accordingly:
| Broad PDOC category | Australian Standard of Classification of Drugs of Concern |
|---|---|
| Alcohol | 2100–2199 |
| Cannabis | 7000–7199, 3200–3299 |
| Heroin | 1202 |
| Meth/amphetamines | 3100–2199 |
| Pharmaceuticals | 1101, 1102, 1201, 1305, 1203, 1100, 1199, 1200, 1298–1299, 1300–1304, 1306–1399, 0005, 1000, 1400–1499, 2400–2499, 2000, 2200–2299, 2300–2399, 2500–2599, 2900–2999, 4000–4999 |
| Other | 3405, 3903, 3906, 3000, 3300–3399, 3400–3404, 3406–3499, 3500–3599, 3600–3699, 3700–3799, 3800–3899, 3900–3901, 3902, 3904–3905, 3999, 0006, 6000–6999, 0003, 5000–5999, 9000–9999 |
Across someone’s treatment history, if a client had alcohol, heroin and codeine (pharmaceuticals) listed as their PDOCs, this has been labelled as ‘3+ pdocs’. If a client had methadone (pharmaceuticals) and oxycodone (pharmaceuticals) listed, this has been labelled as ‘Pharmaceuticals only’. If a client had ecstasy (other), cocaine (other) and cannabis listed as their PDOCs, this has been labelled as ‘Cannabis and another drug’ and ‘Other and another drug’. As clients can be treated for two or more PDOCs, these people will be counted more than once. For example, a person treated for ‘alcohol’ and ‘heroin’ will be included in results for ‘Alcohol and another drug’ and ‘Heroin and another drug’.
Information on PDOC in this article has been presented as a combination of all PDOCs a person has received treatment for, at any point between 1 July 2012 and 30 June 2023. This is because some drug use and drug combinations may indicate a greater rate of death. To only include the PDOC a client received in their last treatment episode prior to death would mask any differences the combination of PDOCs may have on a person’s outcomes. Due to the nature of this study, the service and PDOC history will not be as complete for those clients who died earlier in the reporting period, as those clients who died later in the reporting period. Similarly, the analysis here only presents a client’s PDOC history over the entire study period, which may mask changes over time for specific PDOC combinations.
Considering this, the distribution of clients by PDOC across each financial year of the study period was examined as the overall proportion of deaths by PDOC history can be sensitive to changes in these distributions. This is because if a PDOC becomes more common later in the study period, then this PDOC would include relatively more clients that entered the study period in more recent years and therefore have a shorter follow-up time. This would lower the overall death proportion for that PDOC. Therefore, differences in the overall death proportions by PDOC may be influenced by follow-up length differences rather than a PDOC indicating potentially greater harm than another. While the distribution of clients by PDOC did vary across financial years, these changes were not significant.