Alcohol, tobacco & other drugs in Australia

Key findings

View the pharmaceuticalS fact sheet >

In Australia, pharmaceutical drugs are available via a prescription from a registered healthcare professional or over-the-counter (OTC) from pharmacies and other retail outlets, and are widely used to prevent, treat and cure injury and illness. When used appropriately, pharmaceutical drugs are associated with considerable reductions in morbidity and mortality and are an important pillar of public health. However, pharmaceutical drugs are subject to use other than prescribed use.

The non-medical use of pharmaceutical drugs is an ongoing concern internationally, with different pharmaceutical opioids being misused in different regions.

The use of opioids (including the use of pharmaceutical opioids for non-medical purposes) in Australia in 2019 (3.3 per cent of the adult population) remains higher than the global average (1.2% of the global population aged 15–64 years). In 2019, the majority of pharmaceutical misuse in Australia continued to be for codeine (UNODC 2021).

Dramatic increases in opioid-related deaths associated with rising use in the US have led to the problem being declared an 'epidemic' (CDC 2017; U.S. Department of Health & Human Services 2017). In the US in 2019, 3.6% of the population aged 12 and older had misused pharmaceutical opioids at least once in the past year (UNODC 2021).

Availability

The Australian Criminal Intelligence Commission’s (ACIC) Illicit Drug Data Report (IDDR) includes data on prescription pharmaceutical detections at the Australian border (ACIC 2021). The importation of prescription pharmaceuticals is primarily undertaken by individuals for personal use and without criminal intent (ACIC 2021).

The number of pharmaceutical detections reported at the Australian border only reflect detections of benzodiazepines and opioids (including morphine, buprenorphine, methadone and oxycodone) (ACIC 2021). In 2019–20 there were a total of 1,112 pharmaceutical detections, a decrease of 8% since 2010–11 (1,211 detections) (ACIC 2021). The majority (73%) of these detections were for benzodiazepines (810 in 2019–20). However, though they remain the minority of overall pharmaceutical detections, there has been a 695% increase in the number of pharmaceutical opioid detections over the past decade (38 in 2010-11, increasing to a record high of 302 in 2019–20 (ACIC 2021).

Prescription drug dispensing

AIHW analysis of recent data from the Pharmaceutical Benefits Scheme (PBS) provides information on dispensing of prescription opioids, benzodiazepines and gabapentinoids in Australia (Box PHARMS2). These numbers largely represent medicines being prescribed for and used for their intended purposes. However, drug-related harms are often associated with drug prescribing rates (such as described for opioids in Roxburgh et al. (2017)) so it is important to monitor prescription rates in the context of harm reduction.

Recent data indicate that opioids continue to be dispensed at higher rates than benzodiazepines or gabapentinoids, but rates of dispensing have fallen or remained stable for all 3 drug classes over the past 5 years. Detailed information on specific drug classes and patient characteristics is outlined below.

Dispensing of selected medicines by drug class and generic name

Data from the PBS indicate that in 2021–22:

• Around 13.3 million opioid scripts were dispensed to 2.9 million patients for pain relief, a rate of around 51,400 scripts and 11,200 patients per 100,000 population (not including prescriptions for the treatment of opioid dependence as these data are not available in the PBS) (Figure PHARMS1; tables PBS1–4). 
  • Oxycodone had the highest rate of scripts dispensed (18,200 scripts per 100,000 persons, compared with 13,200 for codeine). However, more patients were dispensed codeine than oxycodone (5,900 per 100,000 persons and 4,400 per 100,000 persons respectively). This may indicate that a higher number of PBS scripts for oxycodone were dispensed per patient than for codeine (Figure PHARMS1).
• Around 4.9 million benzodiazepine scripts were dispensed to 1.4 million patients, a rate of 18,900 scripts and 5,400 patients per 100,000 population. Diazepam had the highest rates of dispensing among benzodiazepine drugs (Figure PHARMS1; tables PBS21–24). 
• Approximately 4.2 million gabapentinoid prescriptions were dispensed to 613,000 patients, a rate of around 16,300 scripts and 2,400 patients per 100,000 population. Pregabalin had the highest rates of dispensing among gabapentinoid drugs (Figure PHARMS1; tables PBS45–48).

Rates of dispensing across all drug classes were typically highest in Inner regional and Outer regional areas. See Data by region for information on PBS dispensing by remoteness area and state/territory.

Figure PHARMS1: PBS prescriptions dispensed or patients who were dispensed selected opioids, benzodiazepines or gabapentinoids, by drug class and generic name, 2012–13 to 2021–22 (number or rate per 100,000)

Between 2012–13 and 2021–22:

• Opioid dispensing rates increased overall up to 2016–17 but have declined since. This was consistent for most opioid drugs except tapentadol, where rates of scripts dispensed increased consistently following its listing on the PBS in 2013–14 (Figure PHARMS1; tables PBS2 and PBS4).
• Benzodiazepine dispensing rates steadily declined across the period. This was consistent across most benzodiazepine drugs, except diazepam where the rate of patients increased to 2018–19 before declining to 2021–22 (Figure PHARMS1; tables PBS22 and PBS24).
• Gabapentin dispensing rates increased to 2017–18 but have been declining since around 2018–19 (Figure PHARMS1; tables PBS46 and PBS48). This varied by gabapentinoid drug:
  1. Pregabalin dispensing rates increased by 12 times (from an age-standardised rate of 1,300 to 14,600 per 100,000 and a crude rate of 1,400 to 16,400 per 100,000) between 2012–13 to 2017–18, before falling to an age-standardised rate of 13,400 per 100,000 and a crude rate of 15,700 per 100,000 in 2021–22. 
  2. Gabapentin dispensing rates decreased from 2012–13 to 2016–17 (age-standardised rate of 455 to 430 per 100,000 and a crude rate of 500 to 475 per 100,000), then increased to 2021–22 (age-standardised rate of 480 per 100,000 and a crude rate of 550 per 100,000) (Figure PHARMS1).

For information on concomitant dispensing of opioids and benzodiazepines, see Poly drug use. Dispensing of prescription opioids, benzodiazepines and gabapentinoids may have been affected by recent policy changes. Refer to the Policy context section for more information.

Patient characteristics

Opioid, benzodiazepine and gabapentinoid dispensing varied by the age and sex of patients. Across all drug categories:

• Rates of dispensing increased with increasing age and were highest for people aged 80 and over (Figure PHARMS2). See Older people: Pharmaceuticals for more information on PBS dispensing by patient age.
• Females had higher rates of dispensing than males. This was similar over time and across most age groups (Figure PHARMS2).

Figure PHARMS2: PBS prescriptions dispensed or patients who were dispensed selected opioids, benzodiazepines and gabapentinoids, by age and sex of patients, 2012–13 to 2021–22 (number or rate per 100,000)

Defined daily doses dispensed for opioid drugs

PBS data indicate that 82.4 million defined daily doses were dispensed in 2021–22 for all opioids, a rate of 8.7 S-DDDs/1,000 population/day. Oxycodone (2.1 S-DDDs/1,000/day), codeine and tramadol (1.8 and 1.7 S-DDDs/1,000/day respectively) had the highest rates of dispensing (tables PBS17–18).

Between 2012–13 and 2021–22, the total rate of S-DDDs dispensed for opioids fell from around 13 S-DDDs/1,000 population/day in 2012–13 (Table PBS18). Trends in dispensing varied by opioid drugs: 

• Rates of dispensing decreased for codeine, fentanyl, methadone, morphine and tramadol (methadone and buprenorphine rates do not include prescriptions for the treatment of opioid dependence as these data are not available in the PBS). 
• The rate of S-DDDs dispensed initially increased for buprenorphine (peaking at 0.8 S-DDDs/1,000/day in 2017–18), hydromorphone (peaking at 0.4 in 2014–15) and oxycodone (peaking at 3.4 in 2013–14), before declining to 2021–22.
• There was a steady increase in the rate of S-DDDs dispensed for tapentadol, rising from <0.1 to 1.1/1,000 population/day between 2013–14 and 2021–22 (Table PBS18).

Decreased rates of opioid dispensing measured in terms of S-DDDs may be related to several recent policy changes, including the de-listing of some medicines, increased restrictions on prescribing some opioids, and the introduction of smaller pack sizes for some PBS-listed immediate-release opioid formulations in June 2020. Refer to the Policy context section for more information.

Opioid dispensing for cancer treatment and palliative care

Opioids are effective for managing strong cancer pain and in the context of palliative care. However, there is limited evidence that opioid medicines are beneficial for the long-term treatment of non-cancer pain conditions (for example, back pain). Long-term opioid use is also linked to increased risk of harm such as overdose (RACGP 2020). Australian clinical guidelines recommend that opioids only be used for chronic non-cancer pain if a patient is experiencing moderate/severe pain that is not adequately managed with other treatments (RACGP 2020).

PBS data indicate that most opioid dispensing in Australia is not related to cancer or palliative care. In 2021–22, around 2.7 million scripts were dispensed in relation to palliative care (10,500 scripts per 100,000 population) and 769,000 scripts were dispensed in relation to cancer treatment (3,000 scripts per 100,000). By comparison, 10.0 million scripts were dispensed not in relation to palliative care or cancer treatment (39,000 scripts per 100,000) (tables PBS19–20). Note that the number of opioid prescriptions related to palliative care and cancer may be underestimated, as some palliative care patients may not have received any PBS items from the Palliative Care Schedule, and some patients receiving treatment for cancer may not have received cancer treatment-related drugs through the PBS. 

The rate of opioid dispensing for palliative care declined 2018–19 and 2020–21, while rates of dispensing for the other two categories (cancer treatment and neither palliative care nor cancer treatment) declined between 2017–18 and 2020–21. Between 2020–21 and 2021–22, the rate of opioid dispensing for palliative care increased from 8,400 to 10,500 prescriptions per 100,000 persons (Table PBS20). 

Additions to PBS listings for palliative care were made on 1 June 2021 (NPS Medicine Wise 2023).

Consumption

Data from the 2019 NDSHS showed that:

• Nearly 1 in 25 (4.2%) Australians aged 14 and over had used a pharmaceutical for non-medical purposes in the previous 12 months (AIHW 2020, Table 4.6)
• 1 in 9 (11.7%) Australians aged 14 and over had used a pharmaceutical for non-medical purposes in their lifetime (AIHW 2020, Table 4.2)
• 1 in 5 (21%) used pharmaceutical drugs daily or weekly for non-medical purposes
• Pharmaceutical pain-killers/pain-relievers and opioids (excluding over-the-counter) are the most common pharmaceuticals used for non-medical purposes (2.7%), followed by tranquilisers/sleeping pills (1.8%) (AIHW 2020, Table 4.6).

Between 2016 and 2019, there was a significant decrease in the use of pharmaceuticals for non-medical purposes (from 4.8% to 4.2%) (AIHW 2020, Table 4.6). This was driven by a decrease in the use of pain-killers/pain-relievers and opioids (excluding over-the-counter) (from 3.6% to 2.7%)–in particular, the proportion of people using codeine for non-medical reasons halved in this period (decreasing from 3.0% in 2016 to 1.5% in 2019) (AIHW 2020). This aligns with restrictions to codeine being introduced in 2018 (see Policy context: Codeine rescheduling).

The 2019 NDSHS showed that people who use pharmaceutical drugs for non-medical purposes were older than people who use illicit drugs (excluding pharmaceuticals) (median ages of 40.7 and 30.7 years, respectively). However, the recent non-medical use of pharmaceuticals was most common among those aged 20–29 (6.4%) (Figure PHARMS2; AIHW 2020).

Figure PHARMS3: Recentᵃ use of pharmaceuticals, by age, sex and drug used, 2019 (percent)

View data tables >

Data from the 2019 NDSHS showed that a higher proportion of Australians aged 14 and over approved the regular adult non-medical use of prescription pain-killers/pain-relievers (12.4%) and tranquilisers/sleeping pills (9.3%) than all illicit drugs except for cannabis (19.6%). However, support for the non-medical use of pharmaceuticals was lower than approval for the use of tobacco (15.4%) and alcohol (45%) (AIHW 2020).

Geographic trends

Findings from the 2019 NDSHS (AIHW 2020) showed:

• The significant decrease between 2016 and 2019 in the use of pain-killers/pain-relievers and opioids (excluding over-the-counter) for non-medical purposes that was observed at the national level also occurred in Victoria (from 3.4% to 2.6%), Queensland (from 4.1% to 2.7%), South Australia (from 4.3% to 2.9%) and the Northern Territory (from 4.2% to 1.9%; note, the 2019 estimate for the Northern Territory has a relative standard error of 25% to 50% and should be used with caution). As noted previously, these decreases align with restrictions to codeine being introduced in 2018 (see Policy context: Codeine rescheduling).
• There was a significant increase in the proportion of people living in Victoria who had recently used tranquillisers/sleeping pills for non-medical purposes (from 1.7% in 2016 to 2.4% in 2019).
• Between 2016 and 2019, there were significant decreases in the use of pain-killers/pain-relievers and opioids (excluding over-the-counter) for non-medical purposes in Major cities (from 3.3% to 2.6%) and Inner regional areas (from 3.6% to 2.5%) (AIHW 2020, Table 7.15).
• People living in Remote and very remote areas were about 1.5 times as likely as those from Major cities to have recently used pain-killers/pain-relievers and opioids (excluding over-the-counter) for non-medical purposes in 2019 (4.1% compared with 2.6%).
• The proportion of people who recently used pharmaceuticals for non-medical purposes was similar for those living in Major cities (4.3%) and Remote and very remote areas (4.8%) (Figure PHARMS3). Note, the estimates for Remote and very remote areas have a relative standard error of 25% to 50% and should be used with caution. These findings were still apparent after adjusting for differences in age (AIHW 2020).
• Between 2016 and 2019, there were significant decreases in the non-medical use of pain-killers/pain-relievers and opioids (excluding over-the-counter) for people living in both the most disadvantaged (from 4.8% to 3.0%) and the most advantaged (from 2.6% to 1.8%) socioeconomic areas (AIHW 2020, Table 7.18). This is consistent with findings at the national level and aligns with restrictions to codeine being introduced in 2018 (see Policy context: Codeine rescheduling).
• People living in the most disadvantaged socioeconomic areas were 1.6 times as likely as those from the most advantaged socioeconomic areas to have used pain-killers/opioids for non-medical purposes (3.0% compared with 1.8%). Conversely, people living in the most advantaged areas were twice as likely as those from the most disadvantaged socioeconomic areas to have recently used tranquilisers for non-medical purposes (2.6% compared with 1.2%). These findings were still apparent after adjusting for differences in age (AIHW 2020, Table 7.18).
• There was a significant decrease in the non-medical use of pharmaceuticals for people living in the most disadvantaged socioeconomic areas (from 6.0% in 2016 to 4.0% in 2019). This resulted in the proportion in 2019 being similar to that for people in the most advantaged socioeconomic areas, even after adjusting for differences in age (AIHW 2020, Table 7.18).

Figure PHARMS4: Recentᵃ use of painkillers, pain-relievers and opioidsᵇ, by remoteness area or socioeconomic area, people aged 14 and over, 2016 and 2019 (per cent)

View data tables >

The National Wastewater Drug Monitoring Program (NWDMP) measures the presence of substances in sewerage treatment plants across Australia. This includes estimated population-weighted average consumption of oxycodone and fentanyl. It should be noted that wastewater analysis cannot differentiate between prescribed and illicit use.  Estimated consumption of oxycodone and fentanyl have plateaued at historically low levels (ACIC 2023).

Data from Report 19 of the NWDMP indicate that nationally:

• Between August and December 2022, oxycodone consumption decreased in both capital city and regional sites.
• In December 2022, oxycodone population-weighted average consumption continued to be higher in regional areas than capital cities.
• Nationally, consumption of oxycodone has been decreasing over the last 2 years and appears to have plateaued at historically low levels (ACIC 2023).

Between August and December 2022, nationally the estimated population-weighted average consumption of fentanyl increased in both capital cities and regional areas. Data from Report 19 of the NWDMP indicate:

• In December 2022, consumption levels in regional areas continued to be higher than in capital cities.
•  Fentanyl consumption has increased since April 2022 in both capital cities and regional areas (ACIC 2023).

For state and territory data, see the National Wastewater Drug Monitoring Program reports.

Poly drug use

Poly drug use is defined as the use of mixing or taking another illicit or licit drug whilst under the influence of another drug. In 2019, the NDSHS showed that just over half (52%) of people who reported the non-medical use of pharmaceuticals in the last 12 months said they did not use any other illicit drug. The most common other drugs concurrently used were alcohol (50% exceeded the single occasion risk guideline at least monthly), cannabis (40%) and cocaine (27%) (AIHW 2020, Table 1.3).

Data on alcohol and other drug-related ambulance attendances are sourced from the National Ambulance Surveillance System (NASS). Monthly data for 2021 are currently available for New South Wales, Victoria, Queensland, Tasmania and the Australian Capital Territory It should be noted that some data for Tasmania and the Australian Capital Territory have been suppressed due to low numbers. Please see the data quality statement for further information.

In 2021, multiple drugs (excluding alcohol) were involved in at least half of ambulance attendances related to benzodiazepines and opioid analgesics. For benzodiazepines, this ranged from 68% of attendances in New South Wales to 75% of attendances in Tasmania; for opioid analgesics the range was from 73% in New South Wales to 81% in Victoria (Table S1.10).

For related content on multiple drug involvement see Impacts: Ambulance attendances.

Concomitant dispensing of prescription drugs

Due to the compounding effects of respiratory depression, patients using opioids and benzodiazepines at the same time may be at higher risk of harm (Boon et al. 2020; RACGP 2022). Data from the Pharmaceutical Benefits Scheme (PBS) indicate that most opioid scripts are dispensed to patients who have not received a recent supply of a benzodiazepine, and vice versa. However, concomitant dispensing is not uncommon (that is, dispensing multiple medicines at the same time or almost the same time). In 2021–22:

• Around 2.4 million opioid scripts were dispensed to patients who had already received a PBS supply of a benzodiazepine in the past 30 days, and 358,000 patients who were dispensed an opioid had also received a benzodiazepine (a rate of 9,400 scripts and 1,400 patients per 100,000 population) (tables PBS37–40).
• Around 1.5 million benzodiazepine scripts were dispensed to patients who had received a PBS supply of an opioid in the past 30 days, and 358,000 patients who received a benzodiazepine had received an opioid (a rate of 6,000 scripts and 1,400 patients per 100,000) (tables PBS41–44).

Rates of concomitant dispensing of opioids and benzodiazepines declined overall between 2013–14 and 2021–22 but fluctuated over time (tables PBS37–44). For more information on dispensing of prescription drugs under the PBS, see Prescription drug dispensing.

Harms

There are a range of short- and long-term health, social and economic harms associated with the non-medical use of pharmaceutical drugs (Table PHARMS1). People who use opioids for chronic pain are more likely than the general population to use pharmaceutical drugs for non-medical purposes (Currow, Phillips & Clark 2016; Vowels et al. 2015). Iatrogenic dependence occurs when patients become dependent on medications that they were medically prescribed for legitimate purposes. Iatrogenic dependence is an increasing concern among people living with chronic non-cancer pain.

Source: Adapted from Currow, Phillips & Clark 2016; DCPC 2007; Nicholas, Lee & Roche 2011; NSW Ministry of Health 2017.

Ambulance attendances

Data on alcohol and other drug-related ambulance attendances are sourced from the National Ambulance Surveillance System (NASS). Monthly data are presented in 2021 for people aged 15 years and over for New South Wales, Victoria, Queensland, Tasmania and the Australian Capital Territory.

Any pharmaceutical drug

In New South Wales, Victoria, Queensland, Tasmania and the Australian Capital Territory, in 2021, for ambulance attendances where any pharmaceutical drug was reported:

• Rates of attendances ranged from 168.4 per 100,000 population in New South Wales to 295.1 per 100,000 population in Queensland.
• Unlike most other drug types, most attendances were for females, 3 in 5 (60%) of total attendances.
• The highest rates of attendances were in people aged 15–24, ranging from:
  • 344.5 per 100,000 population in New South Wales (3,371 attendances), to 662.0 per 100,000 population in the Australian Capital Territory (363 attendances) (Table S1.10; PHARMS5)

Figure PHARMS5: Ambulance attendances for pharmaceuticals, by age, sex and selected states and territories, 2021

Benzodiazepines

In 2021, for benzodiazepine-related ambulance attendances in these jurisdictions:

• Rates of attendances ranged from 50.1 per 100,000 population in New South Wales to 100.4 per 100,000 population in Queensland.
• Over half (55%) of total attendances were for females.
• The highest rates of attendances were in people aged:
  • 15–24 in Queensland (1,058 attendances, 162.5 per 100,000 population) and Victoria (1,284 attendances, 160.1 per 100,000 population) and New South Wales (736 attendances, 75.2 per 100,000 population), 25–34 in the Australian Capital Territory (95 attendances, 138.8 per 100,000 population) and Tasmania (62 attendances, 92.9 per 100,000 population) and (Table S1.10; Figure PHARMS5).

Opioid analgesics

In 2021, for opioid analgesic-related ambulance attendances in New South Wales, Victoria, Queensland, Tasmania and the Australian Capital Territory:

• The rate of attendances ranged from 23.4 per 100,000 population in New South Wales to 45.9 per 100,000 population in Queensland.
• 3 in 5 (56%) of total attendances were for females.
• The highest rates of attendances varied across jurisdictions and were higher in people aged 15–24, in:
  • The Australian Capital Territory (35 attendances, 63.9 per 100,000 population).
  • Tasmania (29 attendances, 46.8 per 100,000 population).
  • Victoria (351 attendances, 43.8 per 100,000 population).
• Opioid analgesic-related attendances were higher for people aged:
  • 45–54 in New South Wales (297 attendances, 29.2 per 100,000 population).
  • 35–44 in Queensland (403 attendances, 58.3 per 100,000 population) (Table S1.10; Figure PHARMS5).

Hospitalisations

Drug-related hospitalisations are defined as hospitalisations with a principal diagnosis relating to a substance use disorder or direct harm relating to use of selected substances (AIHW 2018).

AIHW analysis of the National Hospital Morbidity Database (NHMD) showed that, among all drug-related hospitalisations in 2021–22:

• Around 1 in 20 were for antiepileptic, sedative-hypnotic and antiparkinsonism drugs (excluding alcohol) (6.1% or 8,200 hospitalisations). Within this category, over 2 in 5 hospitalisations were for benzodiazepines (46% or 3,800 hospitalisations).
• Around 1 in 20 were for non-opioid analgesics (5.4% or 7,300 hospitalisations) or opioids (4.3% or 5,800) (Table S1.12c).

This represents a rate of 14.7 hospitalisations per 100,000 population for benzodiazepines and 22.6 per 100,000 for opioids (Table S1.13). 

In 2021–22:

• Almost 2 in 3 hospitalisations for benzodiazepines (67% or 2,500 hospitalisations) or opioids (64% or 3,700) involved an overnight stay.
• More females than males were hospitalised for benzodiazepines (60% or 2,300 hospitalisations).
• Over 1 in 4 hospitalisations for benzodiazepines (27% or 1,000 hospitalisations) were for people aged 55 and over.
• More males than females were hospitalised for opioids (58% or 3400 hospitalisations).
• 1 in 4 hospitalisations for opioids (25% or 1,400 hospitalisations) were people aged 55 and over (Table S1.12a–12c).
• Around 3 in 4 hospitalisations occurred in Major cities for both benzodiazepines (73% or 2,800 hospitalisations) and opioids (74% or 4,300 hospitalisations).
• The rate of hospitalisations for benzodiazepines was highest in Outer regional and Major city areas (16.5 and 15.0 hospitalisations per 100,000 population, respectively, compared with 10.3 per 100,000 in Remote and very remote areas).
• The rate of hospitalisations for opioids was highest in Major cities (23.3 per 100,000, compared with 15.6 per 100,000 in Remote and very remote areas) (Table S1.14).

In the 7 years to 2021–22:

• The number of benzodiazepine-related hospitalisations declined from 6,300 in 2015–16 to 3,800 in 2021–22 (or from 26.1 to 14.7 per 100,000 population)
• The rate of benzodiazepine-related hospitalisations was highest in Outer regional areas from 2015–16 (29.3 per 100,000) until 2018–19 (23.9 per 100,000). In 2019–20 and 2020–21 rates in Major cities (19.7 and 18.5 per 100,000 respectively) were higher than Outer regional (18.5 and 17.7 per 100,000 respectively), before returning to pre-pandemic trends in 2021–22 (Outer regional 16.5 per 100,000 and Major cities 15.0 per 100,000).
• The number of opioid-related hospitalisations fell from 9,200 in 2015–16 to 5,800 in 2021–22 (or from 38.3 to 22.7 per 100,000).

The rate of opioid-related hospitalisations decreased in Major cities from 2015–16 (39.1 per 100,000 to 2021–22 (23.3 per 100,000. (Table S1.14; Figure IMPACT5).

Deaths

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

Detailed analysis of the National Mortality Database by the AIHW showed that drug-induced deaths are more likely to be due to pharmaceutical drugs than illegal drugs, with benzodiazepines being the most commonly involved single drug type in drug-induced deaths (744 deaths in 2021). The rate of deaths where benzodiazepines were present rose from 1.9 per 100,000 population in 1997 to 2.9 deaths per 100,000 population in 2021.

Over the past decade, there has been a rise in deaths with a prescription drug present. For synthetic opioids (including fentanyl and tramadol) in particular, the rate has increased from 0.3 per 100,000 (75 deaths) in 2011 to 0.7 per 100,000 (172 deaths) in 2021 (Figure PHARMS6; Table S1.1).

Figure PHARMS6: Number or age-standardised rate (per 100,000 population) of drug-induced deathsᵃ for benzodiazepines, pharmaceutical opioids and heroin, 1997 to 2021

View data tables >

The National Drug and Alcohol Research Centre (NDARC) reported that over 2 in 3 (67% or 674 deaths) of all opioid-induced deaths in 2021 were attributed to pharmaceutical opioids only. This is compared to 25% (256 deaths) for heroin only and a further 6.9% (70 deaths) that were attributed to both pharmaceutical opioids and heroin (Chrzanowska et al. 2023). 

This is consistent with previous findings from 2018 indicating that pharmaceutical and prescription opioids were present in over 70% of opioid-induced deaths. Pharmaceutical opioids were also the most common opioid present in intentional overdose deaths (suicide) (ABS 2019a).

Involvement of other drugs in pharmaceutical drug-induced deaths is common. For example, preliminary estimates from NDARC show that benzodiazepines were involved in 58% (592 deaths) of opioid-induced deaths in 2021 (Chrzanowska et al. 2023).

Geographic trends

The rate of drug-induced deaths involving pharmaceuticals varied between Major cities and Regional and Remote areas. Specifically:

• Benzodiazepine induced deaths in 2021 were higher in Major cities (3.0 per 100,000 population) compared with Regional and remote areas (2.4 per 100,000 population).
• Drug-induced deaths involving synthetic opioids were slightly higher in Major cities (0.7 deaths per 100,000 population) than Regional and remote areas (0.6 per 100,000 population)
• Deaths involving natural and semi-synthetic opioids (1.7 deaths per 100,000 population) were slightly higher in Regional and remote areas than in Major cities (1.6 per 100,000 population) (Table 8).

Treatment

In 2021–22, where pharmaceuticals were the principal drug of concern:

• Around 3 in 4 (76%) episodes were for opioids or benzodiazepines.
• Over 3 in 5 (61%) clients were male and around 1 in 7 (15%) were Indigenous Australians (AIHW 2023, tables SC.30 and SC.32). 
• Almost 3 in 10 (29%) clients were aged 30–39, 25% were aged 20–29 (AIHW 2023).
• The most common source of referral was self/family (40%) followed by health services (39%) (AIHW 2023, Table Pharmctcl.4).
• Assessment only was the most common main treatment type (30% of treatment episodes), followed by counselling (22%) (AIHW 2023, Table Pharmctcl.3). 

Figure PHARMS7: Treatment provided for own non-medical use of pharmaceuticals, 2021–22

Source: AIHW 2023, tables Drg.87, SC.32 and Drg.91.

At-risk groups

The available evidence indicates that the non-medical use of pharmaceuticals among vulnerable populations is at substantially higher levels than in the general population.

• The non-medical use of pharmaceuticals was higher for Aboriginal and Torres Strait Islander people than for non-Indigenous Australians.
• People with a mental health condition were twice as likely to report the non-medical use of pharmaceuticals than people who had not been diagnosed or treated for a mental health condition.
• People who inject drugs consume pharmaceutical drugs, particularly prescription opioids, at higher rates than the general population and may substitute illicit drugs for pharmaceutical drugs depending on availability.

Policy context

Real-time prescription monitoring

In July 2017, the Australian Government announced $16 million in funding to implement a national real-time monitoring system of prescription drugs. The system will provide an instant alert to pharmacists and doctors if patients are receiving multiple supplies of prescription only medicines (also referred to as ‘doctor or pharmacy shopping’). The program will initially include the monitoring of controlled medicines that are particularly susceptible to non-medical use including morphine, oxycodone, dexamphetamine and alprazolam. The system aims to assist doctors and pharmacists to identify patients who are at risk of harm due to dependency or non-medical use of pharmaceutical drugs and patients that are diverting these medicines.

Alprazolam rescheduling

As of 1 February 2014, the benzodiazepine classed medication, alprazolam was rescheduled from schedule 4 to schedule 8; a controlled medication. This decision was made by the Therapeutics Goods of Administration (TGA) on the basis that substantial evidence demonstrated alprazolam has increased morbidity and mortality in overdose with the possibility of increased toxicity with no additional therapeutic benefits in comparison to other drugs classed as benzodiazepines (TGA 2013).

From February 2017, higher-strength alprazolam products and larger alprazolam pack sizes were delisted from the PBS, and are not captured in the data reported here.

Codeine rescheduling

As of 1 February 2018, medicines containing codeine were reclassified to schedule 4 drugs, meaning they could no longer be sold over-the-counter in pharmacies and were available by prescription only. This decision was made by the Therapeutic Goods Administration (TGA) following substantial evidence of harm from the use of low dose codeine-containing medicines including analgesic preparations combined with other pain relief medicines such as aspirin, paracetamol and ibuprofen (TGA 2016).

Many over-the-counter combination medicines containing codeine were not listed on the PBS prior to February 2018 and are now only able to be obtained with a private prescription. As a result, these medicines are not represented in PBS claim data. However, the scheduling change may account for some of the reductions observed in the 2019 NDSHS in the non-medical use of pain-killers and opioids since 2016 (AIHW 2020).

Pharmaceutical opioid regulatory changes

Regulatory changes are being implemented in Australia to reduce harm from opioid prescription medicines. These include smaller pack sizes for immediate-release prescription opioid products; the inclusion of boxed warnings and class statements in the Product Information documents regarding their potential for harmful and hazardous use; and reinforcing the indications, or appropriate circumstances for the use of, immediate release and modified release products and for fentanyl patches. Smaller pack sizes and fentanyl indication changes came into effect in the first half of 2020. Other changes will be phased in subsequently (TGA 2021).