When was cannabis first used medicinally?

Written records of cannabis in medicine date back at least to ancient Chinese pharmacopoeias around 2700 BCE, and it appears in Egyptian, Indian, Greek and Islamic medical texts across subsequent millennia. That said, historical documentation of use is not the same as modern clinical evidence of efficacy — the two are worth keeping separate.

Why did medicinal cannabis largely disappear from Western pharmacopoeias in the 20th century?

The decline was driven more by politics and international drug control frameworks than by scientific findings. The US Marihuana Tax Act of 1937 and the UN Single Convention on Narcotic Drugs in 1961 effectively ended legal research and medical access in most Western countries for several decades. Practical problems with dosing consistency had also diminished medical enthusiasm before that.

How is medicinal cannabis legally accessed in Australia today?

In Australia, medicinal cannabis is a prescription-only product regulated by the TGA. Patients access it through the Special Access Scheme or through an Authorised Prescriber. Products containing THC above certain thresholds are listed as Schedule 8 controlled drugs under the Poisons Standard, requiring additional prescribing controls. Only products assessed and approved through the TGA's regulatory process are legally available.

Who discovered THC and the endocannabinoid system?

THC was isolated and its structure identified by Raphael Mechoulam and Yechiel Gaoni at the Hebrew University of Jerusalem in 1964. Mechoulam's team later contributed to identifying anandamide in 1992, a naturally occurring cannabinoid in the human body, which helped establish the broader endocannabinoid system — the receptor network through which cannabinoids act.

Does a long history of traditional use mean a substance is safe or effective?

Not automatically. Longevity of use is useful context and can point researchers toward biological mechanisms worth investigating — but it is not clinical evidence. Traditional preparations varied enormously in composition and dose. Modern assessment requires controlled trials, standardised preparations, and peer-reviewed methodology. Both things can be true: a plant can have a long history of use AND require rigorous modern evaluation.

A History of Cannabis in Medicine

By Hannah Bui · 24 May 2026 · 7 min read

Delicate orange flowers on a vine with leaves

Five thousand years of medicinal use, and we still argue about the evidence

The oldest surviving pharmacopoeia to record cannabis use dates to around 2700 BCE , the Shennong Bencao Jing, a Chinese herbal compendium attributed (almost certainly mythologically) to the emperor Shennong. That's not a minor footnote. That's a span of documented human interest in this plant so long that most of our other "traditional" medicines barely register by comparison.

I find that genuinely fascinating. And, as someone who spends a lot of time teaching people how to read research critically, I also find it a useful reminder that longevity of use is not the same as evidence of efficacy. These are different things. Holding both thoughts at once is exactly what this history demands.

Ancient and medieval records

Beyond the Shennong text, cannabis appears in the Ebers Papyrus , an Egyptian medical document from roughly 1550 BCE , where preparations derived from the plant were recorded for a range of uses. Indian Ayurvedic texts from around the first millennium BCE describe bhang, a preparation of cannabis leaves, in ritual and medicinal contexts. Greek physician Dioscorides, writing in the first century CE, catalogued cannabis in his De Materia Medica, a text that remained a foundational European reference for well over a thousand years.

Medieval Islamic physicians, including Ibn Sina (Avicenna), also documented cannabis in their medical writings. The plant crossed trade routes, pharmacopoeias and cultures with remarkable consistency. It is worth being careful, though, about reading modern pharmacological intent into ancient records. Historical usage was embedded in cosmological, spiritual and humoral frameworks very different from contemporary biomedicine. Noting that a 10th-century physician used cannabis is not the same as saying the application maps cleanly onto a 2024 clinical trial.

Cannabis in Western medicine: the 19th century peak

The period between roughly 1840 and 1900 was probably the high-water mark of cannabis in mainstream Western medicine. The key figure here is William Brooke O'Shaughnessy, an Irish physician working in Calcutta for the East India Company, who published detailed observations on cannabis preparations in 1843 after studying their use in India. His work introduced cannabis tinctures to Victorian-era medicine, and they were subsequently listed in the British Pharmacopoeia.

By the late 1800s, cannabis-based tinctures were sold through pharmacies across Britain, the United States and colonial Australia. The problem, from a pharmacological standpoint, was reproducibility. Without standardised extraction, dosing was wildly inconsistent. Two bottles from different batches could vary enormously in their active constituents. This is a problem that standardised extracts attempt to solve today, but in the 19th century there was no reliable solution. Physicians grew frustrated, and the arrival of more chemically consistent synthetic drugs; aspirin in 1899, barbiturates in the early 1900s, gave medicine options that were at least measurably consistent.

Prohibition and the 20th-century gap

Cannabis didn't gradually fade from medicine on scientific grounds. The real story is messier and more political than that. The US Marihuana Tax Act of 1937 effectively ended legal medical cannabis use in America, driven by a confluence of racial politics, industry lobbying and moral panic that historians have documented thoroughly. International prohibition followed through the United Nations Single Convention on Narcotic Drugs in 1961, which classified cannabis as a Schedule IV substance, its most restrictive category; alongside heroin.

Australia followed this international framework. Cannabis was progressively prohibited across state and territory jurisdictions through the mid-20th century, and research ground to a near halt. Decades of potential scientific inquiry were simply lost. I'll admit that when I first sat with that properly, not as a political statement, but as a research-history fact, it struck me as a significant cost to the scientific record. Whatever one's view on prohibition as policy, the gap in the evidence base is real.

The chemistry catches up: isolating the key compounds

The modern scientific chapter begins with chemistry. In 1940, Roger Adams at the University of Illinois isolated cannabidiol (CBD). In 1964, Raphael Mechoulam and Yechiel Gaoni at the Hebrew University of Jerusalem isolated and identified tetrahydrocannabinol (THC); the primary psychoactive constituent of cannabis. Mechoulam, who continued cannabinoid research for over five decades, also contributed to the identification of anandamide in 1992, an endogenous cannabinoid ligand, which helped map what we now call the endocannabinoid system.

That discovery mattered enormously. Identifying a system of receptors, including CB1 receptors, concentrated in the central nervous system, and CB2 receptors, more prevalent in immune tissues, gave researchers a biological framework for understanding how cannabinoids interact with the body at a molecular level. It transformed cannabis from folk remedy to subject of serious receptor pharmacology. Hundreds of other compounds have since been catalogued in the plant, including cannabinoids like cannabigerol (CBG) and various terpenes that contribute to the plant's overall chemical profile.

Re-medicalisation: late 20th century to the present

California passed Proposition 215 in 1996, permitting medical cannabis use under state law; the first such move in the US since prohibition. The Netherlands had already been operating a de facto tolerance policy for years. Through the 2000s, a growing number of jurisdictions began separating medical access from criminal prohibition, and clinical research, though still hampered by regulatory hurdles, began accumulating.

Australia's formal re-engagement came through a series of legislative steps. The Narcotic Drugs Amendment Act 2016 established a federal licensing framework for the cultivation and production of medicinal cannabis. The TGA then established pathways for patients to access cannabis-based medicines through the Special Access Scheme and the Authorised Prescriber pathway. Products listed on the Australian Register of Therapeutic Goods (ARTG) are available under Schedule 4, while products containing THC above certain thresholds are generally scheduled as Schedule 8 controlled drugs under the Poisons Standard.

Access has grown substantially. The TGA reported over 300,000 approvals for medicinal cannabis products through 2022, and prescribing has continued to expand. That's a remarkable turnaround from the near-total prohibition of the mid-20th century.

What the history actually tells us about the evidence

Here's where I'll be honest about my sceptical streak, because I think it matters. A long history of use is genuinely interesting context. It tells us humans have found repeated reasons to return to this plant, and it guided researchers toward the biological systems worth investigating. But it is not a substitute for controlled trials. The 19th-century enthusiasm for cannabis tinctures coexisted with genuine inconsistency in both preparation and effect.

Modern research is working through that backlog in real time. Some cannabinoid applications now have regulated, approved pharmaceutical forms; these have passed through formal trial processes and received TGA approval. Others are the subject of ongoing trials with preliminary findings but no regulatory conclusion yet. Reading the difference between those two categories carefully is, honestly, the most important skill anyone engaging with this space can bring. Between sea swims last winter I went back through a cluster of systematic reviews on cannabinoids, and the variation in methodology across studies is exactly why I keep saying: look at the design, not just the headline.

History gives us the "why this is worth studying." The trial data, properly scrutinised, tells us what we actually know so far.

Sources

, Hannah Bui, Evidence & research-literacy writer

Common questions

When was cannabis first used medicinally?: Written records of cannabis in medicine date back at least to ancient Chinese pharmacopoeias around 2700 BCE, and it appears in Egyptian, Indian, Greek and Islamic medical texts across subsequent millennia. That said, historical documentation of use is not the same as modern clinical evidence of efficacy — the two are worth keeping separate.
Why did medicinal cannabis largely disappear from Western pharmacopoeias in the 20th century?: The decline was driven more by politics and international drug control frameworks than by scientific findings. The US Marihuana Tax Act of 1937 and the UN Single Convention on Narcotic Drugs in 1961 effectively ended legal research and medical access in most Western countries for several decades. Practical problems with dosing consistency had also diminished medical enthusiasm before that.
How is medicinal cannabis legally accessed in Australia today?: In Australia, medicinal cannabis is a prescription-only product regulated by the TGA. Patients access it through the Special Access Scheme or through an Authorised Prescriber. Products containing THC above certain thresholds are listed as Schedule 8 controlled drugs under the Poisons Standard, requiring additional prescribing controls. Only products assessed and approved through the TGA's regulatory process are legally available.
Who discovered THC and the endocannabinoid system?: THC was isolated and its structure identified by Raphael Mechoulam and Yechiel Gaoni at the Hebrew University of Jerusalem in 1964. Mechoulam's team later contributed to identifying anandamide in 1992, a naturally occurring cannabinoid in the human body, which helped establish the broader endocannabinoid system — the receptor network through which cannabinoids act.
Does a long history of traditional use mean a substance is safe or effective?: Not automatically. Longevity of use is useful context and can point researchers toward biological mechanisms worth investigating — but it is not clinical evidence. Traditional preparations varied enormously in composition and dose. Modern assessment requires controlled trials, standardised preparations, and peer-reviewed methodology. Both things can be true: a plant can have a long history of use AND require rigorous modern evaluation.