Direct answer: Cannabis patches are marketed as a steady, smoke-free way to take cannabinoids, but no cannabis patch is a registered medicine in South Africa or anywhere else. The transdermal science behind them is real and narrow: human skin studies show cannabidiol and cannabinol cross skin roughly ten times more readily than THC, and even a purpose-built pharmaceutical transdermal cannabidiol gel missed its primary endpoint in a phase 3 trial. Judge a patch by what it has been shown to deliver, not by the format.
Last updated:
Key takeaways
- Transdermal means the drug reaches the bloodstream through skin. Topical means it acts where you put it. Most products sold as cannabis patches have never shown the first, and the two are not interchangeable.
- In human skin diffusion studies, the permeabilities of cannabidiol and cannabinol were ten-fold higher than for delta-8-tetrahydrocannabinol. Cannabinoids are greasy, large and slow, which is the core delivery problem.
- Ethanol at 30 to 33 percent significantly increased skin flux in those experiments, which is why a carrier matters as much as the cannabinoid.
- ZYN002, a pharmaceutical transdermal cannabidiol gel, enrolled 212 children and adolescents in the phase 3 CONNECT-FX trial and did not achieve statistical significance on its primary endpoint in the full cohort.
- A patch containing THC is a Schedule 6 substance in South Africa and is prescription-only. Selling an unregistered product with a therapeutic claim breaches the Medicines and Related Substances Act 101 of 1965 whatever the delivery format.
- Ask for the permeation data and the content per patch. A product that cannot produce either is selling a format, not a dose.
Researching cannabis patches because you want a route with a known dose behind it? Register for Professor Green store access to begin the doctor-led process. Registration does not guarantee eligibility, a prescription or supply.
What are cannabis patches meant to do?
A patch is a delivery system, not an ingredient. The idea borrowed from nicotine and hormone patches is that a reservoir of drug sits against the skin and releases at a roughly constant rate for hours, giving steadier blood levels than a dose you swallow or inhale. That is the promise being sold.
Whether it happens depends entirely on whether the molecule can get through the stratum corneum, the dead outer layer of skin that exists precisely to keep things out. Nicotine crosses it easily: small, reasonably water-loving, low dose required. Cannabinoids are the opposite. They are large, intensely fat-loving and poorly soluble in the watery layers underneath, so they tend to park in the upper skin and stay there.
That parking behaviour is not a failure if local action is what you want. It is a total failure if you were promised a systemic effect. The whole argument about cannabis patches turns on which of those two things a given product actually does.
What did the human skin studies actually measure?
The most cited work is a 2004 study in the Journal of Pharmacy and Pharmacology by Stinchcomb and colleagues at the University of Kentucky, which set out to quantify in-vitro human skin transdermal flux for three cannabinoids: delta-8-tetrahydrocannabinol, cannabidiol and cannabinol.
Their headline finding is the one worth carrying around. The permeabilities of cannabidiol and cannabinol were ten-fold higher than for delta-8-tetrahydrocannabinol. Cannabinol also had a lower heat of fusion and a correspondingly higher calculated relative thermodynamic activity than cannabidiol, and lag times for cannabidiol were significantly smaller than for cannabinol. Tissue concentrations of delta-8-THC were significantly higher than for cannabinol, which is the parking effect described above showing up in the numbers.
They also found that ethanol concentrations of 30 to 33 percent significantly increased the transdermal flux of delta-8-THC and cannabidiol. That is a formulation finding, not a plant finding. It means the vehicle a product is dissolved in can matter more to delivery than which cannabinoid is in it, and it is the reason a homemade oil smeared under a plaster is not the same thing as a designed transdermal system.
One caveat runs through all of this. These were diffusion-cell experiments on excised human tissue, not people wearing patches. They tell you what can cross skin under controlled conditions. They do not tell you that any particular commercial product achieves a useful blood level.
Has a transdermal cannabinoid been tested properly in people?
Yes, and the result is instructive. ZYN002 is a pharmaceutical-grade transdermal cannabidiol gel developed by Zynerba Pharmaceuticals. CONNECT-FX was a double-blind, randomised phase 3 trial of ZYN002 for behavioural symptoms in children and adolescents with fragile X syndrome, published in the Journal of Neurodevelopmental Disorders in 2022.
A total of 212 patients, mean age 9.7 years, were randomised to 12 weeks of ZYN002 at 250 mg or 500 mg daily on a weight basis, or placebo, added to standard care. Statistical significance was not achieved for the primary endpoint in the full cohort. A pre-specified subgroup with at least 90 percent methylation of the FMR1 promoter did show improvement on the social avoidance measure at a nominal P value of 0.020, which the authors correctly describe as evidence of efficacy in that subgroup rather than a positive trial.
Safety read well. All treatment-emergent adverse events were mild or moderate, and the most common treatment-related one was application site pain at 6.4 percent on ZYN002 against 1.0 percent on placebo.
Take the honest lesson from that. A company with a designed transdermal formulation, a defined daily milligram dose and a fully powered phase 3 trial still missed its primary endpoint. Retail cannabis patches have none of those three things.
What about the animal evidence people quote?
The study usually cited is Hammell and colleagues in the European Journal of Pain in 2016, which applied transdermal cannabidiol gels at 0.6, 3.1, 6.2 or 62.3 mg per day for four consecutive days in a rat model of induced arthritis of the knee.
The findings were genuinely encouraging within their limits. Plasma cannabidiol concentrations from transdermal absorption were linear across the 0.6 to 6.2 mg per day range. The gel reduced joint swelling, spontaneous pain scores, immune cell infiltration and synovial membrane thickening in a dose-dependent way, with 6.2 and 62 mg per day identified as the effective doses, and exploratory behaviour was unaffected.
That is a rat, four days, an induced arthritis model and a gel rather than a patch. It supports further research. It does not establish that a patch relieves arthritis pain in a person, and anyone quoting it as if it does has skipped several steps.
Before you start
In South Africa, medical cannabis containing THC is prescription-only through a registered doctor, usually via a SAHPRA Section 21 authorisation. This is general information for adults 18 and over, not medical advice. Speak to a doctor or pharmacist before starting, stopping or changing any treatment, especially about dosing, and tell them about every other medicine you take.
Transdermal or topical: how to tell which you are buying
This single distinction resolves most of the confusion, so it is worth setting out plainly.
| Question | Transdermal | Topical |
|---|---|---|
| Where it acts | Bloodstream, so effects can be body-wide | The tissue under the application site only |
| What must be proven | Measured plasma concentrations over time | Local effect, plasma levels may be near zero |
| Formulation | Designed system with permeation enhancers and a controlled release rate | A cream, balm or oil base |
| Drug test implications | A THC-containing system can produce detectable metabolites | Less likely, but never assume zero |
| Typical retail claim | Often claimed | Usually what is actually delivered |
Our guide to cannabis topicals in South Africa covers the local-action side in detail, including what a balm can reasonably be expected to do for a sore joint or skin complaint.
Where do cannabis patches sit in South African law?
Delivery format does not change scheduling. South African medicines law schedules substances, not shapes, so the question is what is in the patch.
Government Notice 586 in Government Gazette 43347 of 22 May 2020 placed tetrahydrocannabinol in Schedule 6 with narrow exceptions, and Government Notice R5181 in Government Gazette 51171 of 6 September 2024 amended that entry. The current entry reads tetrahydrocannabinol rather than naming only the delta-9 isomer, the previous 0,2 percent industrial and 0,001 percent processed-product exceptions were deleted, and the exception for raw plant material cultivated, possessed and consumed by an adult in private remains. A patch delivering THC therefore sits squarely in Schedule 6 and needs a prescription.
Cannabidiol is handled separately. It is Schedule 4, with a Schedule 0 carve-out for complementary medicines supplying no more than 600 mg of cannabidiol per sales pack at a maximum daily dose of 20 mg making only a low-risk claim, and for ingested processed products containing 0,0075 percent or less. Read those conditions carefully before assuming a cannabidiol patch qualifies. The carve-out is written around ingested and packaged complementary medicines, and a transdermal delivery system making a therapeutic claim is a different proposition.
The point that survives every variation: selling an unregistered product with a therapeutic claim breaches the Medicines and Related Substances Act 101 of 1965 regardless of the molecule or the format. Our explainer on how cannabis scheduling works in South Africa sets out the schedules in full, and the step-by-step guide to SAHPRA Section 21 covers the route for an unregistered medicine where a prescriber considers one appropriate.
What to ask before buying cannabis patches
Five questions separate a considered product from a printed box. None of them require you to be a pharmacologist.
First, how many milligrams of each cannabinoid does one patch contain, and over what stated wear time is it released. A product that answers only the first half is describing contents, not a dose. Second, what permeation data exists for this formulation, in what tissue, and is it published or available on request. Third, what is the vehicle, because the ethanol finding above shows the carrier drives flux.
Fourth, ask for the certificate of analysis for the batch in front of you, with the date and the laboratory named. Fifth, ask directly whether the seller claims a systemic or a local effect, and hold them to whichever they answer. Sellers who will not commit to one have usually chosen the ambiguity on purpose.
The same discipline applies to any cannabis product, not just patches. A named cultivar page such as Blue Zushi is only as informative as the batch certificate behind it, which is why the analysis document matters more than the name on the label.
If the product is intended to treat a diagnosed condition, that conversation belongs with a prescriber rather than a shop assistant, and the answers should be in writing.
How Professor Green fits into the process
Professor Green is a South African medical cannabis access platform and a marketing affiliate of the Dr Green network, which holds the medical cannabis licences in South Africa. Professor Green does not itself sell, handle or ship cannabis. Products are sold and fulfilled by the licensed operator.
Professor Green does not manufacture delivery systems, does not test formulations and does not decide what is clinically appropriate for you. A registered medical practitioner assesses and prescribes, a registered pharmacist dispenses from a licensed pharmacy, SAHPRA authorises unregistered products where section 21 applies, and a licensed operator supplies.
Registering creates an account and starts that doctor-led route. It is not a guarantee of eligibility, of a prescription, of stock or of supply, and it does not turn an unproven product into a proven one.
The bottom line on cannabis patches
The science underneath cannabis patches is legitimate and unfinished. Human tissue work shows that cannabidiol and cannabinol cross skin about ten times more readily than THC, that lag times differ between them, and that the solvent carrying them changes flux substantially. Animal work suggests a transdermal cannabidiol gel can reduce inflammation and pain behaviour at defined daily doses. Those are reasons to keep researching.
What does not follow is that a patch on a shelf delivers a therapeutic dose to your bloodstream. The one large randomised trial of a purpose-built transdermal cannabidiol product did not meet its primary endpoint, and no cannabis patch is a registered medicine here. Treat the format as a hypothesis with some support, not as a finished answer.
If you are looking at cannabis patches for a real medical problem, take the question to a prescriber, ask for the per-patch milligram content and the permeation evidence, and keep the legal position in view: a THC patch is a Schedule 6 medicine in South Africa and belongs on a prescription.
Frequently asked questions
Do transdermal patches deliver THC into the bloodstream?
They can in principle, but THC is the hardest of the three studied cannabinoids to get through skin. In human diffusion studies the permeabilities of cannabidiol and cannabinol were ten-fold higher than for delta-8-tetrahydrocannabinol, and tissue concentrations of delta-8-THC were significantly higher, meaning more of it stays in the skin.
What is the difference between transdermal and topical?
Transdermal means the drug is intended to reach the bloodstream through the skin and act body-wide. Topical means it acts in the tissue under the application site. Most retail products deliver the second while being marketed as the first.
Has a transdermal cannabidiol product passed a phase 3 trial?
Not on its primary endpoint. CONNECT-FX randomised 212 children and adolescents with fragile X syndrome to ZYN002 transdermal cannabidiol gel or placebo for 12 weeks and did not achieve statistical significance in the full cohort, though a subgroup with at least 90 percent FMR1 methylation improved at a nominal P value of 0.020.
Does the carrier in the formulation matter?
Substantially. Ethanol concentrations of 30 to 33 percent significantly increased the transdermal flux of delta-8-THC and cannabidiol in human skin experiments, which is why a designed transdermal system is not comparable to oil under a plaster.
Are these products legal to buy in South Africa?
A product delivering THC is a Schedule 6 medicine and is prescription-only. Cannabidiol sits in Schedule 4 with narrow Schedule 0 exceptions written around complementary medicines. Selling an unregistered product with a therapeutic claim breaches the Medicines and Related Substances Act 101 of 1965 whatever the format.
Could wearing one affect a workplace drug test?
If the product contains THC and delivers any systemically, detectable metabolites are possible. Never assume a skin product is undetectable, and declare it if you are subject to testing.
What should I ask a seller?
The milligram content of each cannabinoid per unit, the stated wear time, the permeation data for that specific formulation, the batch certificate of analysis with the laboratory named, and whether they are claiming a systemic or a local effect.
Is the rat arthritis study enough evidence to rely on?
No. It applied cannabidiol gel for four days in an induced arthritis model, found dose-dependent reductions in swelling and pain behaviour with 6.2 and 62 mg per day effective, and showed linear plasma levels from 0.6 to 6.2 mg per day. It supports further research rather than a clinical claim in people.
Sources and references
- Stinchcomb AL, Valiveti S, Hammell DC, Ramsey DR. Human skin permeation of delta-8-tetrahydrocannabinol, cannabidiol and cannabinol. Journal of Pharmacy and Pharmacology, March 2004, 56(3), 291 to 297
- Hammell DC et al. Transdermal cannabidiol reduces inflammation and pain-related behaviours in a rat model of arthritis. European Journal of Pain, July 2016, 20(6), 936 to 948
- Berry-Kravis E et al. A randomized, controlled trial of ZYN002 cannabidiol transdermal gel in children and adolescents with fragile X syndrome (CONNECT-FX). Journal of Neurodevelopmental Disorders, 25 November 2022, 14(1), 56
- SAHPRA, Communication to industry on the scheduling status of THC and CBD, covering Government Notice 586, Government Gazette 43347, 22 May 2020
- Government Notice R5181, Government Gazette 51171, 6 September 2024, amending the Schedules to the Medicines and Related Substances Act
- Medicines and Related Substances Act 101 of 1965, including sections 14, 21 and 22A
- SAHPRA, Licences and permits, including section 21 applications for unregistered products
Study findings above were read from the published abstracts and the regulatory position from the notices themselves on 20 September 2026. Schedules change by notice, so verify any legal point against the current instrument before relying on it. General information for adults aged 18 and over, not medical or legal advice.
Want a route where the dose is actually known?
If reading about cannabis patches has convinced you that content per dose and real evidence matter, the next step is to create a Professor Green account and begin the doctor-led route. A registered doctor assesses and prescribes where appropriate, a registered pharmacist dispenses, and a licensed operator supplies. Registration does not guarantee eligibility, a prescription or supply.
Create Your Professor Green Account

