Alcohol, tobacco & other drugs in Australia

Introduction

The availability of heroin in illicit drug markets in Australia has fluctuated over time. In the early 2000s, there was a rapid and considerable reduction in the availability of heroin (commonly referred to as the heroin shortage or drought) and this was associated with dramatic reductions in heroin-related overdoses (Degenhardt et al. 2004). Since then, the availability of heroin rose and later stabilised. 

This page focuses on the harms, availability, consumption of and treatment for heroin, as distinct from pharmaceutical opioids. Pharmaceutical opioids (including morphine, methadone and oxycodone) have been regulated by health professionals and used as pharmaceuticals for medical purposes, including pain management and treatment for opioid dependence (NDARC 2016). The reporting uses data from a range of sources, mostly national administrative and survey data. For related content on use of pharmaceutical opioids in this report, see Pharmaceutical drugs.

What data sources are available?

Data on heroin use is predominantly sourced from surveys while harms data such as ambulance attendances, hospitalisations and deaths are sourced from administrative or computer systems. 

For more information about each data source, see Technical notes.

What do we know about the availability of heroin in Australia?

Surveys of people who regularly inject drugs indicate that heroin is generally easy to obtain among people who regularly use it. Heroin accounts for a relatively small but rising proportion of the total illicit drug border detections, seizures and arrests in Australia each year. 

For detailed information on heroin availability, see Illicit drug markets and drug-related law enforcement activities.

What do we know about people who use heroin?

How many people use heroin and has it changed over time?

The National Drug Strategy Household Survey (NDSHS) shows that past year heroin use among the general population has remained low in Australia between 2001 (0.2%) and 2022–2023 (0.1%; estimate has a relative standard error of 25% to 50% and should be used with caution) (AIHW 2024, Table 5.6; Figure 1). Lifetime use has also remained low over the same period (1.6% in 2001 and 1.2% in 2022–2023) (AIHW 2024, Table 5.6). 

Does heroin use differ by age and gender?

The use of heroin is similar among males and females. In 2022–2023, 1.4% of males and 0.9% of females reported using heroin at least once in their lifetime (AIHW 2024, table 5.106). Data on heroin use by age is not available in the NDSHS due to the small number of people reporting use.

Are people using heroin with other drugs?

Data from the NDSHS does not include estimates for the proportion of people who used heroin with other drugs. However, data from the National Ambulance Surveillance System showed that in 2024, 39% of all ambulance attendances involving heroin involved multiple drugs.

Does heroin use differ by geographic area?

Data on heroin use by state and territory, remoteness area and socioeconomic areas is not available in the NDSHS due to the small number of people reporting use and a high margin of error in the results. 

Detailed information on heroin use by geographic area within Australia is available in State and territory data, Remoteness areas, and Socioeconomic areas.

What are the harms associated with heroin use?

Heroin is a central nervous system depressant. Like other opioids, it binds to receptors in the brain, sending signals to block pain and slow breathing (Table 1). Heroin can also trigger a range of negative short- and long-term effects, including drowsiness, constipation, and dry mouth resulting in tooth decay. Injection comes with a range of additional harms associated with the unsanitary sharing of injecting equipment, such as the transmission of blood borne viruses like Hepatitis C and HIV (Table 1).

Source: Adapted from ACIC 2019; Nielsen and Gisev 2017; NDARC 2017.

Heroin-related ambulance attendances

Data on alcohol and other drug-related ambulance attendances are sourced from the National Ambulance Surveillance System (NASS) and are currently available for six of the eight Australian states and territories (excluding Western Australia and South Australia). 

There were almost 4,100 heroin-related attendances among people aged 15 and over in 2024, a rate of 22 per 100,000 population. Among these attendances:

• over 7 in 10 (71%) were for males
• the highest rates of attendances were in people aged 35–44 (38 per 100,000 population) and 45–54 (40 per 100,000) (Table NASS3).

Between 2021 and 2023, the rate of heroin-related ambulance attendances remained relatively stable overall, ranging from 22 to 25 per 100,000 population. While the national number and rate of attendances remained stable between 2023 and 2024 (around 4,000 and 4,100 attendances, respectively, or 22 per 100,000 population in both years), there were increases in attendances across all individual states and territories with available data, except Victoria. 

In Victoria, attendances decreased from around 2,100 (38 per 100,000 population) in 2023 to around 1,300 (23 per 100,000 population) in 2024. This is explained by industrial action by paramedics in Victoria between March and September 2024, which resulted in fewer ambulance attendances being captured over that period. Therefore, the national data for 2024 is lower than expected and should be interpreted with caution (Table NASS3). 

Heroin-related hospitalisations

Opioid-related poisoning hospitalisations are generally more likely to involve pharmaceutical opioids than heroin. The number and rate of heroin-related poisoning hospitalisations has fluctuated over time, with an overall decrease from 994 hospitalisations (3.9 per 100,000 population) in 2018–19 to 494 (1.9 per 100,000 population) in 2022–23 following previous rises (Chrzanowska et al. 2025a). Additionally, in 2022–23: 

• Males had a higher number and rate of hospitalisations (359 hospitalisations; 2.7 per 100,000 population) than females (134 hospitalisations; 1.0 per 100,000). This has remained consistent over time.
• People aged 40–49 had the highest number of hospitalisations (158; 4.7 per 100,000) out of all age groups (Chrzanowska et al. 2025a).

Heroin-induced deaths

Opioids, including both licit and illicit substances, have been the leading class of drug present in drug-induced deaths in Australia for the last 2 decades. Of the 1,948 drug-induced deaths in Australia in 2024, 501 or 26% involved heroin (Table NMD2).

People who use heroin have a particularly high risk of overdose, especially when heroin is used in conjunction with other drugs like benzodiazepines (for example, alprazolam, diazepam) and alcohol. However, there are some challenges in interpreting the numbers of heroin deaths. Heroin can be difficult to identify at toxicology because it is rapidly metabolised to morphine by the body and these metabolites cannot be distinguished from other morphine sources (for example, codeine).

The rate of deaths involving heroin has overall declined since the late 1990s, when heroin consumption was at its peak in Australia (Degenhardt et al. 2004). Following declines in the 2000s, deaths involving heroin have shown an increase over the last decade, from 1.0 per 100,000 people in 2014 to 2.1 in 2018 and 2019. The rate has since decreased slightly to 1.9 in 2024, though this is a preliminary estimate and may increase with further revisions (Table NMD2, Figure 2).

There has also been a shift in the profile of opioid-induced deaths in Australia over time, away from pharmaceutical opioids and towards heroin (Table NMD2, Figure 2). In 2024, all opioids excluding heroin were involved in 62% (635 deaths) of opioid-induced deaths, down from 83% (917 deaths) in 2014. Heroin was involved in 49% of opioid-induced deaths in 2024 (501 deaths), up from 22% (237 deaths) in 2014 (Table NMD2).

In 2023, the rate of heroin-induced deaths was higher for:

• males than females (2.2 compared with 0.6 deaths per 100,000 population, respectively) 
• people aged 35–44 (2.8 per 100,000 people) or 45–54 (3.6 per 100,000) (Chrzanowska et al. 2025b).

How many people receive specialist treatment for heroin use?

Of the 5,462 clients who received treatment for heroin as their principal drug of concern in 2023–24:

• over 2 in 3 (68%) were male
• about two-thirds were aged either 30–39 (32% of clients) or 40–49 (33%)
• over 1 in 5 (22%) were Aboriginal and Torres Strait Islander (First Nations) people (AIHW 2025, tables SC.9–SC.11).

Additionally, of all treatment episodes provided to clients for their own use of heroin as their principal drug of concern in 2023–24:

• Around 4 in 5 (81% or 7,532) were for people who had previously received AOD treatment since 2013–14 (AIHW 2025, Table SCR.28a).
• Assessment only was the most common main treatment type (22% of episodes, or 2,028), followed by pharmacotherapy (20% or 1,835) and counselling (19% or 1,815) (AIHW 2025, Table Drg.54).

Treatment agencies with a sole function of prescribing or providing dosing services for opioid pharmacotherapy are excluded from the AODTS NMDS. Due to the multi-faceted nature of service delivery in this sector, these data are captured in the National Opioid Pharmacotherapy Statistics Annual Data (NOPSAD) collection.

NOPSAD data showed that, on a snapshot day in 2025, 28% of clients across Australia reported heroin as their opioid drug of dependence. However, these data should be interpreted with caution due to the high proportion of clients whose drug of dependence was ‘Not stated/not reported’ (52% of clients overall in 2025) (AIHW 2026b, Table S10). 

Where do I go for more information?

• Low-prevalence illicit drugs in the NDSHS
• National Opioid Pharmacotherapy Statistics Annual Data collection