Alcohol, tobacco & other drugs in Australia

The health burden of alcohol and other drug use is considerable and includes hospitalisation from injury and other disease, mental illness, pregnancy complications, injection-related harms, overdose and mortality.

Drug-induced deaths

Drug-induced deaths are defined as those that can be directly attributable to drug use and includes both those due to acute toxicity (for example, drug overdose) and chronic use (for example, drug-induced cardiac conditions) as determined by toxicology and pathology reports (ABS 2021). 

See also the Technical notes. Multiple drug types may have been reported on a single death record as associated causes of death. As a result, the sum of each drug type may be more than the total number of deaths.

Preliminary Australian Bureau of Statistics data on Causes of Death show that in 2020:

• There were 1,842 drug-induced deaths (1,187 males and 655 females), fewer than in 2019 (1,874; 1,188 males and 686 females).
• After adjusting for population growth and ageing the rate of drug-induced deaths (7.2 per 100,000 population) decreased from 2019 (7.5 per 100,000 population).
• There was a higher rate of drug-induced deaths in regional Australia than capital cities. The rate of drug-induced deaths in Remainder of states (7.2 per 100,000 population, 578 deaths) was higher than Capital cities (7.0 per 100,000 population, 1,237 deaths).
• The median age at death was higher for females than males (48.1 and 44.4 years respectively)
• Over two-thirds (67%) of drug-induced deaths were considered accidental (1,233 deaths) and 23% (428 deaths) were considered intentional (ABS 2021). This compares to 1,244 accidental drug-induced deaths and 442 intentional drug-induced deaths in 2019 (66% and 24% respectively).

Detailed analysis of the AIHW National Mortality Database by the Australian Institute of Health and Welfare (AIHW) showed that in 2019:

• Benzodiazepines continued to be the largest contributor to drug-induced deaths. In 2019, there were 811 drug-induced deaths involving benzodiazepines (a rate of 3.3 per 100,000 population) (Figure IMPACT1; Table S1.1a).

Figure IMPACT1: Number or age-standardised rate (per 100,000 population) of drug-induced deathsᵃ, by drug type or drug class, 1997 to 2019

View data tables >

In 2019, the highest rates of drug-induced deaths were for people aged 45–54 (15.1 deaths per 100,000 population) and 35–44 (14.2 deaths per 100,000 population). This has changed since the 1990s, when the rate of deaths was highest for people aged 25–34 (Chrzanowska et al. 2021).

Common drug classes and types identified in drug-induced deaths

Over the past 2 decades, benzodiazepines have remained the most commonly-identified single drug type in drug-induced deaths, and opioids the most common drug class (Figure IMPACT1; Table S1.1a). AIHW analysis of the National Mortality Database showed that, in 2019:

• Opioids were the most commonly identified drug class, present in 3 in 5 (60.5% or 1,129) drug-induced deaths—a rate of 4.6 per 100,000 population. Opioids include a number of drug types including heroin, opiate based analgesics (such as codeine and oxycodone) and synthetic opioid prescriptions (such as tramadol and fentanyl).
• Benzodiazepines were the most commonly identified single drug type, present in over 2 in 5 (43% or 811) deaths. It is important to note that benzodiazepines may not have been recorded as the underlying cause of death and are commonly reported as an associated cause in deaths due to other drug types.
• Other drugs commonly identified in drug-induced deaths included depressants excluding alcohol (51% or 944 deaths) and psychostimulants excluding cocaine (25% or 469 deaths).

Preliminary mortality data indicate that there has been a recent change in the main type of opioid identified in drug-induced deaths. Over the past decade, opioid-induced deaths were more likely to be due to prescription drugs than illegal drugs, and there has been a rise in the number of deaths with a prescription drug present. However, the proportion and rate of opioid-induced deaths relating to illegal opioids—opium and heroin—is increasing.

• NDARC analysis indicates that the proportion of opioid-induced deaths attributed to illegal opioids only rose from 17% in 2014 to 32% in 2019, while the proportion attributed to prescription opioids only fell from 75% to 56% (Chrzanowska et al. 2021).
• Preliminary data from AIHW analysis of the National Mortality Database indicate that in 2019, there was a slightly higher rate of drug-induced deaths involving heroin (1.9 per 100,000 population) than natural and semi-synthetic opioids (1.8 per 100,0000) for the first time since 1997 (Figure IMPACT1; Table S1.1a).

Since 1997, the most common other drug involved in opioid-induced deaths was benzodiazepines (55% or 614 deaths in 2019). The majority (78%) of opioid-induced deaths were accidental (Chrzanowska et al. 2021).

Poly drug use

In 2020, 5 or more drugs were present in 19.3% of all drug-induced deaths, and a further 17.9% reported 3 drugs present. Females were more likely than males to have 5 or more drugs present at toxicology (23.0% and 17.3%, respectively) (ABS 2021).

Geographic trends

AIHW analysis of the National Mortality Database showed that:

• The rate of drug-induced deaths in 2019 was marginally higher in Major cities (7.4 per 100,000 population) compared with Regional and remote areas (7.2 per 100,000 population). The lowest rate of drug-induced deaths was recorded in Remote and very remote areas (3.9 per 100,000 population, compared with 7.4 for Major cities and 7.5 for Regional areas) (AIHW unpublished).
• The rate of drug-induced deaths has fluctuated since 2009 in both Major cities and Regional and remote areas. However, when comparing the rates for 2009 and 2019, in Regional and remote areas the rate increased by 19% while the rate in Major cities increased by 9%.

Psychosocial risk factors

Psychosocial risk factors, recorded for coroner-referred deaths in the National Mortality Database, are ‘social processes and social structures which can have an interaction with individual thought or behaviour and health outcomes’ (ABS 2019). Risk factors may not be mutually exclusive and therefore deaths with multiple psychosocial risk factors recorded will be counted in more than one category.

In 2019, at least one psychosocial risk factor was recorded in one-third (33%) of drug-induced deaths. For intentional drug-induced deaths, this proportion was more than 3 in 5 (61%) (Table S1.1b).

Personal history of self-harm was the most commonly identified risk factor (12%), followed by relationship issues including: disruption of family by separation and divorce (5%); disappearance and death of a person in the primary support group (3.6%); and problems in relationship with spouse or partner (3.2%) (Table S1.1c). However, there were some notable differences when they were examined by intent, age and sex:

• Release from prison was ranked the 6th most common psychosocial risk factor for all drug-induced deaths, however, for accidental deaths, this was the second most common risk factor identified (Table S2.71a).
• Limitation of activities due to disability was the most common risk factor for people aged 65 years and older, (Table S2.71b).
• Disruption of family by separation and divorce was the second most common risk factor for males, while for females it was disappearance and death of a family member (Table S2.71c).

There were also differences in the most commonly identified risk factors across drug types:

• For cocaine, disruption of family by separation and divorce was the most commonly identified risk factor.
• Release from prison was the second most common risk factor identified in drug-induced deaths involving methadone and all psychostimulants and was equal second for deaths involving heroin.
• Limitation of activities due to disability was the third most common risk factor identified in deaths involving natural and semi-synthetic opioids (Figure IMPACT2).

Figure IMPACT2: Leading psychosocial risk factors identified in drug-induced deaths, by drug class or drug type, 2019

View data tables >

Associated causes of death

Associated causes of death refer to conditions other than the underlying cause of death and can include diseases that are part of the chain of events leading to death, risk factors and co-morbid conditions (ABS 2020). People with multiple associated causes of death recorded will be counted in more than one category.

AIHW analysis of the National Mortality Database showed that of drug-induced deaths in 2019 (Table S1.1c):

• 1 in 2 (52%) had mental and behavioural disorders due to psychoactive substance use as an associated cause of death.
  • Of the 1,460 mentions of these disorders as an associated cause of death, most were mental and behavioural disorders due to the use of opioids or multiple drug use and use of other psychoactive substances (Table S2.71d).
• 1 in 3 (33%) had mood (affective) disorders as an associated cause of death.
  • Of the 671 mentions of mood (affective) disorders as an associated cause of death, the majority were for depressive episode and the remaining 16% were for bipolar affective disorder (Table S2.71d).

Deaths due to harmful alcohol consumption

Alcohol-induced deaths are defined as those that can be directly attributable to alcohol use (i.e. where an alcohol-related condition is recorded as the underlying cause of death), as determined by toxicology and pathology reports. This may be the result of a chronic condition directly related to alcohol use (e.g. alcoholic liver cirrhosis) or from an acute condition directly related to harmful consumption (e.g. alcohol poisoning).

Alcohol-related deaths include deaths directly attributable to alcohol use (as defined above) and deaths where alcohol was listed as an associated cause of death (e.g. a motor vehicle accident where a person recorded a high blood alcohol concentration) (ABS 2018). See also the Technical notes.

Australian Bureau of Statistics data on Causes of Death show that in 2020:

• The number of alcohol-induced deaths reached a 10 year peak of 1,452 (1,056 males and 396 females) after a previous high of 1,399 deaths in 2017.
• After adjusting for population growth and ageing the rate of alcohol-induced deaths (5.2 per 100,000 population) increased from 2019 (5.0 per 100,000 population).
• The majority (73% or 1,056 deaths) were recorded for males (396 deaths for females).
• After adjusting for population differences, Remainder of states had a higher rate of deaths than Capital cities (6.4 and 4.5 per 100,000 population respectively). However, Remainder of states had a lower number of alcohol-induced deaths (630 deaths in Remainder of states and 749 deaths in Capital city areas) (ABS 2021).

Australian Institute of Health and Welfare (AIHW) analysis of the AIHW National Mortality Database (Table S1.1d) showed that in 2019:

• People were 2.4 times as likely to have alcohol certified at death as an associated cause (4,504 alcohol-related deaths) than to have died from an alcohol-induced death (1,317). This has increased from 1.4 times as likely in 2010.

Burden of disease

The Australian Burden of Disease Study 2018 shows that tobacco use (8.6%) and alcohol use (4.5%) collectively accounted for 13.1% of the total burden of disease in Australia in 2018. Declines were observed for most leading risk factors with the exception of illicit drug use for which the age-standardised DALY rate increased by 35% between 2003 and 2018 (AIHW 2021c).

Burden of disease analysis is used to compare the impact of different diseases, conditions or injuries on a population (AIHW 2021ca). It combines the burden of living with ill health (non-fatal burden) with the burden of dying prematurely (fatal burden). This is measured through the calculation of disability-adjusted life years (DALY) – one DALY is one year of 'healthy life' lost due to illness and/or death.

Tobacco, alcohol and illicit drug use contribute to increased chronic disease, injury, poisoning and premature death and are among the leading risk factors contributing to disease burden in Australia (AIHW 2021c) (Figure IMPACT3). Analysis of data from the Australian Burden of Disease Study 2018 for tobacco, alcohol and illicit drug use based on the latest evidence of linked diseases indicated the following:

• Tobacco, alcohol and illicit drug use collectively accounted for 16.1% of the total burden of disease in Australia in 2018.
• Tobacco use contributed to 8.6% of the total burden of disease in Australia in 2018. It was responsible for 39% of the burden of respiratory diseases, 22% of cancers, 10.7% of cardiovascular diseases, and 6.2% of infections (Table S2.58).
• Alcohol use contributed to 4.5% of the total burden of disease in Australia in 2018 and was the leading risk factor for males aged 15–44 (12.3% compared to females 3.9%). Males experienced a greater proportion of disease burden attributable to alcohol use than females.
• Alcohol use was responsible for 100% of the burden due to alcohol use disorders, 40% of liver cancer burden, 25% of road traffic injuries—motor vehicle occupant burden, 19% of chronic liver disease burden and 14% of suicide burden (Table S2.63) .  
• Illicit drug use contributed to 3.0% of the total burden of disease in Australia in 2018, most of which was experienced by people aged 15–44. Males aged 15–44 experienced a greater proportion of total disease burden attributable to illicit drug use than females in this age group (10.6% compared to 4.4%).
• Illicit drug use was responsible for 100% of the burden of drug use disorders (excluding alcohol) and 72% of the poisoning burden. It was also responsible for 74% of the acute Hepatitis C burden, 33% of the acute Hepatitis B burden, and 7.2% of the HIV/AIDS burden.
• Opioid use accounted for the largest proportion (31%) of the illicit drug use burden, followed by amphetamine use (24%), cocaine (10.9%) and cannabis (10.2%). In addition, 17.8% of the burden was from diseases contracted through unsafe injecting practices (Table S2.69) (AIHW 2021c).

The Australian Burden of Disease Study 2018: key findings for Aboriginal and Torres Strait Islander people was released in October 2021. Full results from the Study are planned for release in early 2022.

Analysis of key findings from the Study shows risk factors contributing to the burden of disease for Indigenous Australians included:

• Tobacco use was the highest risk factor, 11.9% of total burden
• Alcohol use was the second highest risk factor, 10.5% of total burden
• Illicit drug use was the 4th highest risk factor, 6.9% of total burden (AIHW 2021d).

Figure IMPACT3: Burden of disease due to alcohol or illicit drug use, by selected linked disease and sex, 2018

View data tables >

Injury and hospitalisation

Recent consumers of alcohol (past 12 months)

• According to the 2019 National Drug Strategy Household Survey (NDSHS):
  • 1.2% of consumers aged 14 years or older had been injured while under the influence of alcohol and required medical attention and 0.4% required admission to hospital for their injuries. 
  • less than 1% of consumers aged 14 years or older required medical attention (0.3%) or hospitalisation (0.2%) because they were so intoxicated (Table S1.4).
• People who consumed alcohol in risky quantities (lifetime or single occasion risk) were far more likely to require medical attention or admission to hospital due to injuries sustained while drinking or due to intoxication. This was even higher among people aged 14 years or older who consumed 11 or more standard drinks at least monthly with 4.9% requiring medical attention for their injuries (AIHW 2020) (Table S1.5).

People who have recently used illicit drugs (past 12 months)

2019 NDSHS data on injury and hospitalisation for people who have recently used illicit drugs have a high relative standard error and should be interpreted with caution (AIHW 2020).

• Data from the 2019 NDSHS shows that 1.2% of people aged 14 and over who have recently used illicit drugs reported that they had injured themselves while under the influence of illicit drugs and required medical attention and 0.5% said their injury was serious enough to require hospitalisation (Table S1.6).
• Less than 1% of people who have recently used illicit drugs reported that they had overdosed and required medical attention (0.9%) or hospitalisation (0.4%) (Table S1.6). This was higher among people who had used meth/amphetamines in the previous 12 months (2.0% had overdosed and required medical attention and 0.8% required hospitalisation) (AIHW 2020) (Table S1.7).

Hospitalisations

A hospitalisation is an episode of admitted patient care ending with discharge, death, or transfer, or a portion of a hospital stay (AIHW 2021a). Drug-related hospitalisations are hospitalisations where the principal diagnosis (considered to be responsible for an episode of admitted patient care to hospital) relates to a substance use disorder or direct harm due to selected substances (AIHW 2018). For more information, see the Glossary and Technical notes.

In 2019–20, there were 11.1 million hospitalisations in Australia’s public and private hospitals (AIHW 2021a). Drug-related principal diagnoses accounted for 1.3% of these hospitalisations (Table S1.8a). This is a similar proportion to the previous year, when 1.2% of hospitalisations had a drug-related principal diagnosis.

AIHW analysis of the National Hospital Morbidity Database (NHMD) showed:

• In 2019–20, there were 140,709 drug-related hospitalisations.
• The total number of drug-related hospitalisations increased between 2015–16 and 2019–20. However, taking into account population growth, the rate of drug-related hospitalisations was lower in 2019–20 (550.5 per 100,000 population) than in 2015–16 (565.1 per 100,000) (Table S1.8b; Figure IMPACT4).
• Overnight separations continued to be more common than same-day separations, accounting for 60% of all drug-related hospitalisations (Table S1.8a).

Figure IMPACT4: Number or crude rate (per 100,000 population) of hospitalisations with a drug-related principal diagnosis, by drug type, 2015–16 to 2019–20

View data tables >

Common drug types and classes identified in drug-related hospitalisations

In the 5 years to 2019–20, alcohol remained the most common drug type involved in drug-related hospitalisations (Table S1.8b; Figure IMPACT4). AIHW analysis of the NHMD showed that, among hospitalisations with a drug-related principal diagnosis in 2019–20:

• Alcohol was the most commonly involved drug, accounting for 53% (74,511) of hospitalisations.
• Amphetamines and other stimulants accounted for 12.9% (18,157) of hospitalisations. Over 3 in 4 (77%) of these hospitalisations were related to methamphetamines.
• Antiepileptic, sedative-hypnotic and antiparkinsonism drugs (excluding alcohol) accounted for 7.6% (10,685) of hospitalisations. Almost half (47%) of these hospitalisations were related to benzodiazepines.
• Other drugs involved in hospitalisations include opioids (5.4% or 7,597 hospitalisations), non-opioid analgesics (4.8% or 6,783) and cannabinoids (4.7% or 6,640) (Table S1.8a; Figure IMPACT4).

The most recent analysis of the NHMD by the National Drug and Alcohol Research Centre (NDARC) identified 61,780 drug-related hospitalisations in 2018–19 (Man et al. 2021). This is lower than the number estimated from AIHW analysis for that period (140,570 hospitalisations; Tables S1.8b). This is primarily due to the exclusion of alcohol-related hospitalisations from the NDARC analysis (see Box IMPACT2).

According to the NDARC analysis:

• The 5 most common drugs recorded in drug-related hospitalisations in 2018–19 were similar to the AIHW analysis (excluding alcohol) for the same year. These 5 drugs remained the most commonly recorded since 2012–13.
• The age-standardised rate of drug-related hospitalisations was higher in 2018–19 than in 1999–2000, but remained stable from 2017–18 (Man et al. 2021).

Overdose and misuse

Overdose and misuse of alcohol and other drugs (AOD) are public health concerns that affect the community on many levels. Surveillance and monitoring of AOD overdose and misuse can help to form an evidence base in relation to trends and emerging patterns of harms (Moayeri et al. 2020).

Data from the Illicit Drug Reporting System (IDRS) and the Ecstasy and Related Drugs Reporting System (EDRS) show rates of self-reported overdose among people who regularly use stimulant drugs (EDRS) and who regularly inject drugs (IDRS). In 2021:

• 1 in 10 (9%) IDRS participants reported experiencing a non-fatal heroin overdose in the last 12 months (Sutherland et al. 2021b).

• 16% of EDRS participants reported experiencing a non-fatal stimulant overdose in the last 12 months (Sutherland et al. 2021a).

Data collection for 2021 took place from April–August for the EDRS and June–July for the IDRS. Due to COVID-19 restrictions being imposed in various jurisdictions during data collection periods for both the IDRS and the EDRS, interviews in 2020 and 2021 were delivered face-to-face as well as via telephone. This change in methodology should be considered when comparing data from the 2020 and 2021 samples relative to previous years.

Ambulance attendances

Data on alcohol and other drug-related ambulance attendances are sourced from the National Ambulance Surveillance System for Alcohol and Other Drug Misuse and Overdose. Data for 2020 are currently available for New South Wales, Victoria, Queensland, Tasmania and the Australian Capital Territory. Data are reported for 4 snapshot months per year, specifically March, June, September and December. Please see the data quality statement for further information.

For the 5 jurisdictions that supplied data in 2020:

• Across all jurisdictions, the highest number (and rate) of ambulance attendances were alcohol intoxication-related (with rates ranging from 143.3 per 100,000 population in New South Wales to 198.9 per 100,000 population in Queensland).

• Ambulance attendance rates were considerably lower for all other drugs reported here, including meth/amphetamines (10.7 per 100,000 population in Tasmania to 27.4 per 100,000 population in Victoria).

• With the exception of NSW, benzodiazepine and opioid analgesic-related ambulance attendances were predominantly for females, while attendances for alcohol and other drugs across all jurisdictions were predominantly for males.

• Around 90% of benzodiazepine-related ambulance attendances resulted in transfer to hospital (ranging from 85% to 94% of attendances), while heroin had the lowest rates of transfer to hospital, ranging from 44% to 81% of attendances.

• The involvement of multiple drugs (excluding alcohol) was reported in over half of all benzodiazepine and opioid analgesic-related ambulance attendances (Table S2.81).

Mental health conditions

There is a strong association between illicit drug use and mental health issues. According to the 2019 NDSHS:

• between 2016 and 2019 there was an increase in the proportion of people who had recently used an illicit drug (in the past 12 months) experiencing high or very high levels of psychological distress (from 22% to 26%)
• the proportion of people who recently used drugs who had been diagnosed with or treated for a mental health condition in the previous 12 months remained stable at around 26% (Table S1.9).

Over half (58%) of participants of the 2021 EDRS reported mental health issues in the preceding 6 months, an increase from 52% in 2020. The primary issue of concern reported among this population of people who regularly use ecstasy and other stimulants was anxiety (71%) and depression (62%) (Sutherland et al. 2021a) (refer to Box HARM1 for more information). 

Data collection for the EDRS for 2021 took place from April–August. Due to COVID-19 restrictions in various jurisdictions, interviews were delivered via face-to-face interview and by telephone. This change in methodology should be considered when comparing data from the 2020 and 2021 samples relative to previous years.

Pregnancy complications

Supporting the health and wellbeing of women throughout pregnancy helps to ensure healthy outcomes for mothers and their babies. Encouraging healthy behaviours during pregnancy can reduce the risk of adverse outcomes for mothers and their babies.

Tobacco smoking in pregnancy is the most common preventable risk factor for pregnancy complications. Smoking is associated with poorer perinatal outcomes including low birthweight, being small for gestational age, pre-term birth and perinatal death.

Data from the National Perinatal Data Collection showed that, in 2019 compared to babies of mothers who did not smoke, babies of mothers who smoked at any time during pregnancy were more likely to be:

• Low birthweight (12.2% compared with 6.0% of liveborn babies).
• Small for gestational age (15.7% compared with 8.7% of liveborn singleton babies).
• Born pre-term (12.8% compared with 8.0%) (AIHW 2021b).

Alcohol consumption during pregnancy is also associated with adverse impacts for development of the fetal brain. Fetal alcohol spectrum disorder (FASD) is the term used to describe the effects of prenatal alcohol exposure including fetal alcohol syndrome (FAS). There are currently no data available indicating the prevalence of FASD in Australia, however there have been some jurisdictional based studies which reported birth prevalence of FAS of between 0.01 and 0.68 per 1000 live births. Higher prevalence of FAS is commonly found among Indigenous communities, likely reflecting socioeconomic factors and patterns of alcohol use (Burns et al. 2013).

The latest Australian Guidelines to reduce Health risks from Drinking Alcohol advise that to prevent harm from alcohol to their unborn child, women who are pregnant or planning a pregnancy should not drink alcohol (NHMRC 2020).

The 2019 NDSHS showed that 65% of pregnant women (aged 14–49) abstained from drinking alcohol during their pregnancy; this is an increase from 40% in 2007 and 56% in 2016. The remaining women reported that they reduced their drinking during pregnancy (35%) compared with when they were not pregnant, and less than 1% reported drinking the same (AIHW 2020).

Questions on substance using during pregnancy were updated in the 2013 NDSHS to provide a more accurate picture of drinking during pregnancy—see 2019 NDSHS Technical notes for further information. Each question collects information about slightly different concepts. The measure about what women consumed before and after knowledge of pregnancy is likely to give the most accurate estimate on the amount of alcohol consumed during pregnancy but has only been collected since 2013.

Results from the 2019 NDSHS showed that among pregnant women aged 14–49 who were unaware of their pregnancy, about 1 in 2 (55%) consumed alcohol before they knew they were pregnant, and this declined to 14.5% once they knew they were pregnant. Among all pregnant women, regardless of whether they knew they were pregnant, 3 in 10 (30%) reported drinking alcohol during pregnancy and this has declined from 42% in 2013 (AIHW 2020).

Injection-related harms

The 2019 NDSHS estimates that a very low proportion of the Australian general population aged 14 and over have injected drugs, either in their lifetime (1.5%) or in the past 12 months (0.3%) (AIHW 2020) (tables S2.31 & S2.32). However, people who inject drugs are at a higher risk of health problems, including HIV and hepatitis C (UNODC 2021).