CBD LoveThe independent Australian guideSearch

CBD dosage: what trials used and how prescribers decide

There is no standard CBD dose. See the doses clinical trials used for each condition, what happened, how mg and mL relate, and how prescribers set a dose.

Facts checked 13 November 20269 min readHow we research

There is no single correct dose of CBD, and this page does not recommend one. In Australia the dose is set by the doctor or nurse practitioner who prescribes it, for one person and one condition. What we can do is show the doses that clinical trials used, what happened at each, and how to read a dose once it has been prescribed.

Who decides the dose?

Your prescriber does. Only two cannabis medicines are registered in Australia: cannabidiol oral solution (Epidyolex) and nabiximols (Sativex), which contains THC. Every other CBD product is an “unapproved” medicine that a prescriber must judge case by case.1 That means there is no official dose chart for most uses. The prescriber weighs the condition, your other medicines, your liver health and how you respond.

The law also allows a low-dose pharmacist-only category (Schedule 3), capped at 150 mg a day for adults, although as of November 2026 no product has been approved for it. See Does low-dose CBD work? and Can you buy CBD over the counter in Australia? for how that works in practice.2

What doses have trials used, and what happened?

The table lists human trials of CBD on its own. None of these products contained THC unless stated. “Placebo” means a dummy treatment.

Condition Dose and form Who and how long Result
Dravet syndrome (a severe childhood epilepsy) 20 mg/kg/day oral solution, added to existing medicines 120 children and young adults, 14 weeks Median monthly convulsive seizures fell from 12.4 to 5.9 on CBD and from 14.9 to 14.1 on placebo3
Tuberous sclerosis complex seizures 25 or 50 mg/kg/day, oral 224 people Seizures fell 48.6% (25 mg/kg), 47.5% (50 mg/kg) and 26.5% (placebo). The lower dose was better tolerated4
Social anxiety, public-speaking test Single 600 mg dose 24 people with untreated social anxiety disorder Less anxiety during the speech than placebo5
Treatment-resistant anxiety in young people Stepped up to as much as 800 mg/day 31 people aged 12 to 25, 12 weeks, no placebo group Anxiety scores fell 42.6%. With no placebo group, the authors could not rule out a placebo effect6
Insomnia 150 mg under the tongue each night 30 adults, 2 weeks No difference from placebo in insomnia severity or time to fall asleep. One objective measure and self-rated wellbeing favoured CBD7
Acute low back pain Single 400 mg oral dose 100 adults in a Melbourne emergency department No better than placebo at 2 hours8
Hand osteoarthritis and psoriatic arthritis 20 to 30 mg/day, synthetic CBD 129 adults analysed, 12 weeks No difference in pain from placebo9
Knee osteoarthritis 600 mg/day, added to paracetamol 86 adults, 8 weeks No difference in pain. More liver enzyme rises on CBD (15 people versus 5)10
Schizophrenia 1,000 mg/day, added to antipsychotics 88 adults, 6 weeks Modest improvement in some symptoms. Across five trials, results are mixed1112
Cannabis use disorder 200, 400 or 800 mg/day 82 people, 4 weeks 200 mg was dropped as ineffective. 400 and 800 mg reduced cannabis use13
Heroin craving 400 or 800 mg/day for 3 days 42 abstinent people with heroin use disorder Less craving and anxiety when shown drug cues14
Cocaine use disorder 800 mg/day 78 people Did not reduce craving or relapse15
Parkinson’s disease 75 or 300 mg/day 21 people No difference in movement scores. A quality-of-life score was better at 300 mg16
Irritable bowel syndrome 50 mg chewing gum, as needed 32 women No significant difference in pain from placebo17
COVID-19 (mild to moderate) 300 mg/day 14 days Did not change the course of the illness18

Three things stand out.

  • The strongest evidence is at very high doses. Epilepsy trials used 10 to 25 mg for every kilogram of body weight, every day. For a 70 kg adult that is roughly 700 to 1,750 mg a day.
  • A bigger dose did not rescue a failed use. Pain trials found no benefit at 20 to 30 mg, at 400 mg and at 600 mg.8910
  • Many positive results are from single doses or small, short studies. Each condition page explains how much weight they can bear. Start with how we grade the evidence.

Why are consumer-sized doses so different from trial doses?

Trial doses are mostly counted in hundreds of milligrams. A review by Australian researchers of oral CBD at 400 mg a day or less found that benefits “became more clearly evident at doses ≥300 mg”.19 The same group wrote in Australian Prescriber that very low doses, about 5 to 25 mg, have little evidence of health benefit.20

Consumer advice overseas points the other way, for safety reasons. The UK Food Standards Agency advises healthy adults to limit CBD from food supplements to 10 mg a day, mainly because of liver effects and sleepiness with long-term use.21

So there is a gap between the doses in consumer products sold overseas and the doses tested in most trials. Evidence at 600 mg is not evidence at 20 mg.

How do milligrams, millilitres and strength fit together?

A dose is the amount of CBD, in milligrams (mg). An oil is measured out in millilitres (mL). The link between them is the strength, written as mg/mL. Epidyolex, for example, is 100 mg of CBD in each mL.22

The sum is: mL to measure = dose in mg ÷ strength in mg/mL.

  • For an oil labelled 100 mg/mL, a prescribed 50 mg dose is 0.5 mL.
  • If a label gives only the total, divide by the bottle size. A 30 mL bottle holding 3,000 mg is 100 mg/mL.

Two oils can look identical and differ ten-fold in strength, so the mL from one bottle never carries over to another. Our CBD dose calculator does this arithmetic for a dose your prescriber has set. See also how to read a CBD label.

Does body weight change the dose?

For the registered epilepsy medicine, yes: the dose is worked out per kilogram.23 Most adult trials for other conditions used one fixed dose for everyone, as the table shows. Whether weight matters for you is a question for your prescriber.

What is titration?

Titration means starting at a low dose and raising it in steps while watching for benefit and side effects. The US prescribing information for CBD oral solution in epilepsy starts at 5 mg/kg/day and moves to 10 mg/kg/day after one week, with a maximum of 20 mg/kg/day for Dravet and Lennox-Gastaut syndromes. It sets lower starting and maintenance doses for people with moderate or severe liver impairment.23 Australian prescribing information may differ, and your prescriber sets the schedule.

Titration works both ways. TGA guidance says treatment should be stopped if benefit has not been demonstrated.24 See starting CBD and stopping CBD.

Why can the same dose act differently?

The milligrams on the label are not the milligrams that reach your blood.

  • Food. A high-fat meal raised total CBD exposure about 4-fold compared with an empty stomach in studies of the oral solution.23 See taking CBD with food.
  • Route. Swallowed CBD is poorly absorbed, about 6% on an empty stomach.25 See bioavailability.
  • Other medicines. CBD changes the blood levels of some medicines, and some medicines change CBD levels.23 See CBD drug interactions.

Do higher doses carry more risk?

Yes, for the liver in particular. In epilepsy trials of prescription doses, liver enzymes (blood markers of liver stress) rose above three times the normal limit in 13% of people on CBD and 1% on placebo.23 In a US Food and Drug Administration study, 5.6% of healthy adults taking 5 mg/kg/day (about 350 to 400 mg a day) for 28 days had the same rise, against none on placebo.26 The Australian review of oral doses up to 400 mg a day did not see side effects increase between 60 and 400 mg, so the picture at lower doses is less clear.19 See CBD and the liver and CBD side effects.

Questions to ask your prescriber

  • What dose am I starting on, and what dose are we working towards?
  • What strength is this product, and how many mL is each dose?
  • Should I take it with food, and at the same time each day?
  • Which of my other medicines could interact with it?
  • Do I need liver blood tests?
  • When will we review whether it is working, and what happens if it is not?

Footnotes

  1. NPS MedicineWise. Medicinal cannabis: what you need to know. https://www.nps.org.au/professionals/medicinal-cannabis-what-you-need-to-know ↩

  2. Therapeutic Goods Administration. Notice of final decision to amend the Poisons Standard: cannabidiol. https://tga.gov.au/node/284859 ↩

  3. Devinsky O, et al. Trial of cannabidiol for drug-resistant seizures in the Dravet syndrome. New England Journal of Medicine, 2017. https://doi.org/10.1056/NEJMoa1611618 ↩

  4. American Epilepsy Society. Abstract: efficacy and safety of cannabidiol for seizures associated with tuberous sclerosis complex, GWPCARE6 phase 3 trial (later published by Thiele et al., JAMA Neurology, 2021). https://aesnet.org/abstractslisting/efficacy-and-safety-of-cannabidiol-cbd-for-seizures-associated-with-tuberous-sclerosis-complex-tsc-in-pediatric-and-adult-patients-from-gwpcare6-a-phase-3-trial-with-open-label-extension-ole ↩

  5. Bergamaschi MM, et al. Cannabidiol reduces the anxiety induced by simulated public speaking in treatment-naive social phobia patients, 2011. https://pubmed.ncbi.nlm.nih.gov/21307846/ ↩

  6. Berger M, et al. Cannabidiol for treatment-resistant anxiety disorders in young people: an open-label trial. Journal of Clinical Psychiatry, 2022. https://pubmed.ncbi.nlm.nih.gov/35921510/ ↩

  7. Narayan AJ, et al. Cannabidiol for moderate-severe insomnia: a randomized controlled pilot trial of 150 mg of nightly dosing. Journal of Clinical Sleep Medicine, 2024. https://pubmed.ncbi.nlm.nih.gov/38174873/ ↩

  8. Bebee B, et al. The CANBACK trial: a randomised, controlled clinical trial of oral cannabidiol for people presenting to the emergency department with acute low back pain. Medical Journal of Australia, 2021. https://www.mja.com.au/journal/2021/cannabidiol-no-better-placebo-low-back-pain ↩ ↩2

  9. Vela J, et al. Cannabidiol treatment in hand osteoarthritis and psoriatic arthritis: a randomized, double-blind, placebo-controlled trial (conference abstract; full paper in Pain, 2022). https://acrabstracts.org/abstract/cannabidiol-treatment-in-hand-osteoarthritis-and-psoriatic-arthritis-a-randomized-double-blind-placebo-controlled-trial/ ↩ ↩2

  10. Pramhas S, et al. Oral cannabidiol as add-on to paracetamol for painful chronic osteoarthritis of the knee. Lancet Regional Health Europe, 2023. https://pmc.ncbi.nlm.nih.gov/articles/PMC10682664/ ↩ ↩2

  11. Review summarising McGuire et al., 2018 (cannabidiol 1,000 mg/day as add-on therapy in schizophrenia). https://pmc.ncbi.nlm.nih.gov/articles/PMC8005219/ ↩

  12. Meta-analysis of five randomised trials of add-on cannabidiol in schizophrenia, 2026. https://pubmed.ncbi.nlm.nih.gov/42047775/ ↩

  13. Freeman TP, et al. Cannabidiol for the treatment of cannabis use disorder: a phase 2a, double-blind, placebo-controlled, randomised, adaptive Bayesian trial. Lancet Psychiatry, 2020. https://pmc.ncbi.nlm.nih.gov/articles/PMC7116091/ ↩

  14. Review summarising Hurd et al., American Journal of Psychiatry, 2019 (cannabidiol and cue-induced craving in heroin use disorder). https://pmc.ncbi.nlm.nih.gov/articles/PMC12104536/ ↩

  15. Mongeau-Pérusse V, et al. Cannabidiol as a treatment for craving and relapse in individuals with cocaine use disorder: a randomized placebo-controlled trial, 2021. https://pubmed.ncbi.nlm.nih.gov/33464660/ ↩

  16. Chagas MHN, et al. Effects of cannabidiol in the treatment of patients with Parkinson’s disease: an exploratory double-blind trial, 2014. https://pubmed.ncbi.nlm.nih.gov/25237116/ ↩

  17. Randomised crossover trial of cannabidiol chewing gum in irritable bowel syndrome (PubMed record). https://pubmed.ncbi.nlm.nih.gov/33998882/ ↩

  18. ClinicalTrials.gov. Cannabidiol for COVID-19 patients with mild to moderate symptoms (CANDIDATE), NCT04467918. https://clinicaltrials.gov/study/NCT04467918 ↩

  19. Arnold JC, McCartney D, Suraev A, McGregor IS. The safety and efficacy of low oral doses of cannabidiol: an evaluation of the evidence. Clinical and Translational Science, 2023. https://pubmed.ncbi.nlm.nih.gov/36259271/ ↩ ↩2

  20. Arnold JC, et al. Prescribing medicinal cannabis. Australian Prescriber, 2020. https://www.nps.org.au/assets/p152-Arnold-et-al.pdf ↩

  21. UK Government. Provisional cannabidiol (CBD) acceptable daily intake: lay summary. https://www.gov.uk/government/publications/provisional-cannabidiol-cbd-acceptable-daily-intake-adi/lay-summary ↩

  22. Therapeutic Goods Administration. Epidyolex (Australian prescription medicine decision summary). https://www.tga.gov.au/resources/auspmd/epidyolex ↩

  23. US National Library of Medicine, DailyMed. Epidiolex (cannabidiol) oral solution: prescribing information, revised 2026. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=8bf27097-4870-43fb-94f0-f3d0871d1eec ↩ ↩2 ↩3 ↩4 ↩5

  24. Therapeutic Goods Administration. Guidance for the use of medicinal cannabis in Australia: overview. https://www.tga.gov.au/resources/resource/guidance/guidance-use-medicinal-cannabis-australia-overview ↩

  25. World Health Organization, Expert Committee on Drug Dependence. Cannabidiol (CBD) critical review report, 2018. https://cdn.who.int/media/docs/default-source/controlled-substances/whocbdreportmay2018-2.pdf ↩

  26. EurekAlert news release on Florian J, et al., JAMA Internal Medicine, 2025 (cannabidiol and liver enzyme elevations in healthy adults). https://www.eurekalert.org/news-releases/1089832 ↩

Common questions

How much CBD should I take?

There is no standard dose, and this site does not recommend one. In Australia CBD is a prescription medicine, so the prescriber sets the dose for the person and the condition, usually starting low and adjusting. Clinical trials have used doses from about 20 mg a day up to more than 1,000 mg a day, with very different results.

Is the CBD dose different for anxiety, pain and sleep?

Trials have used different doses for different conditions, but a bigger dose did not reliably mean a better result. Single doses of 300 to 600 mg reduced anxiety in some laboratory studies, 150 mg a night was similar to placebo for insomnia on most measures, and pain trials from 20 mg to 600 mg a day found no benefit.

Does body weight change the CBD dose?

For the registered epilepsy medicine, yes. Its dose is calculated in milligrams per kilogram of body weight per day. Most adult trials for other conditions used a fixed dose in milligrams regardless of weight. Your prescriber decides which approach applies.

How do I convert mg of CBD to mL of oil?

Divide the dose in milligrams by the strength in milligrams per millilitre. For an oil labelled 100 mg/mL, a 50 mg dose is 0.5 mL. This is arithmetic for a dose your prescriber has already set, not a way to choose a dose.

What is a high dose of CBD?

There is no official definition. For context, the Australian pharmacist-only category is capped at 150 mg a day, most trials that reported a benefit used 300 mg a day or more, and epilepsy doses of 10 to 25 mg per kilogram work out at roughly 700 to 1,750 mg a day for a 70 kg adult. Liver enzyme rises have been clearly documented from about 300 mg a day upward.

This page is general information, not medical or legal advice. Laws and evidence change; we show the date we last checked the facts at the top of every page. Talk to your doctor or pharmacist before starting, stopping or changing any medicine.

Didn't find what you needed? Ask us a question.

Keep reading