Objective It is important to know the prevalence and source of medicinal cannabis use in the population because non-prescribed medicinal use of cannabis products places individuals at higher risk of harms. We estimated the prevalence and correlates of the use of cannabis for medicinal purposes in Australia, three years after Australians were given legal access.

Design Cross-sectional.

Setting The 2019 Australian National Drug Strategy Household Survey.

Participants Participants were 22,015 Australians aged 14 or above.

Outcome measure Self-reported cannabis use in the last 12 months for medicinal purposes only, both medicinal/recreational reasons, or recreationally only. Those who reported medicinal use were asked if it had been prescribed by a doctor. Prevalence estimates were weighted to the population and multinomial logistic regression examined the correlates.

Results The prevalence of any medicinal cannabis use in the past year was 2.6%. Only 0.8% of the sample reported using cannabis solely for medicinal reasons, 95.9% of whom did not have a prescription. A self-reported diagnosis of cancer was associated with medicinal use only. Self-reported chronic pain was associated with both medicinal only and medicinal/recreational use. Medicinal cannabis use was associated with opioids use.

Conclusions In 2019, the prevalence of cannabis use solely for medicinal reasons remains under 1%, was more common among people with specific medical conditions, but most individuals do not have a prescription. The prevalence of self-reported medicinal cannabis use in Australia is low and there is limited use of the legal pathway for medicinal cannabis.