CBD is the non-intoxicating cannabinoid from hemp — same plant as the one that gets you high, none of the high. It touches many targets weakly, which is why trials needed large doses. The part that matters clinically is separate: CBD slows the cytochrome P450 enzymes that clear roughly 60% of prescription drugs. In one epilepsy trial, adding it raised blood levels of the seizure drug clobazam about five-fold, enough to force a dose cut 1. The same effect reaches statins, blood thinners, antidepressants, transplant and chemotherapy drugs 2. "Natural" is not the same as inert.
The one clean signal is the one nobody's selling. Purified CBD cut convulsive seizures about 40% in children with rare epilepsies, at 10–20 mg per kg per day — for an adult, 700–1,400 mg daily under a neurologist 3. That's a prescription drug the FDA approved on that evidence.
For anxiety, the studied dose runs in hundreds of milligrams. 600 mg before a public-speaking test lowered anxiety scores 4; a follow-up found 300 mg helped while 100 and 900 didn't 5. A systematic review still rated the anxiety evidence insufficient to recommend 6. For chronic pain, the trials showing anything mostly used products containing THC, not CBD alone 7.
If you're going to try it, run it like a trial, not a habit. A 10 mg gummy every night, forever, told to no one is the pattern least able to give you an answer.
The honest arc is short. A four-week trial at a real dose gives a clean answer in a month: it worked and you continue with eyes open, or it didn't and you stop before the bottle sells itself to you for two more years. Skip it and a few hundred to a few thousand dollars a year stays in your account, along with the habit of checking dose against evidence before you buy — which pays off on every wellness product invented after this one.
The fine print — when to skip it, and what people get wrong
The label is often wrong. Of 84 products tested, only 31% matched their label and 21% held THC despite being sold as hemp 9; most topicals were mislabeled too 10. Nobody at the FDA is checking 11.
Clear it with a doctor first if you take warfarin 12, anti-seizure drugs, transplant immunosuppressants, statins, or chemotherapy. Skip it in pregnancy and with liver disease; therapeutic doses raised liver enzymes in healthy adults 13.
Most "it didn't work" cases are one thing: the dose was a fraction of what studies used, or the bottle held less than the label claimed 9. A real dose from a third-party-tested product rules both out.
- 1Geffrey AL, Pollack SF, Bruno PL, Thiele EA (2015). Drug-drug interaction between clobazam and cannabidiol in children with refractory epilepsy. Epilepsia. link
- 2Brown JD, Winterstein AG (2019). Potential adverse drug events and drug-drug interactions with medical and consumer cannabidiol (CBD) use. Journal of Clinical Medicine. link
- 3Devinsky O, Cross JH, Laux L, et al. (2017). Trial of cannabidiol for drug-resistant seizures in the Dravet syndrome. New England Journal of Medicine. link
- 4Bergamaschi MM, Queiroz RH, Chagas MH, et al. (2011). Cannabidiol reduces the anxiety induced by simulated public speaking in treatment-naïve social phobia patients. Neuropsychopharmacology. link
- 5Linares IM, Zuardi AW, Pereira LC, et al. (2019). Cannabidiol presents an inverted U-shaped dose-response curve in a simulated public speaking test. Brazilian Journal of Psychiatry. link
- 6Black N, Stockings E, Campbell G, et al. (2019). Cannabinoids for the treatment of mental disorders and symptoms of mental disorders: a systematic review and meta-analysis. The Lancet Psychiatry. link
- 7Mücke M, Phillips T, Radbruch L, Petzke F, Häuser W (2018). Cannabis-based medicines for chronic neuropathic pain in adults. Cochrane Database of Systematic Reviews. link
- 8Birnbaum AK, Karanam A, Marino SE, et al. (2019). Food effect on pharmacokinetics of cannabidiol oral capsules in adult patients with refractory epilepsy. Epilepsia. link
- 9Bonn-Miller MO, Loflin MJE, Thomas BF, et al. (2017). Labeling accuracy of cannabidiol extracts sold online. JAMA. link
- 10Spindle TR, Sholler DJ, Cone EJ, et al. (2022). Cannabinoid content and label accuracy of hemp-derived topical products available online and at national retail stores. JAMA Network Open. link
- 11FDA (2023). FDA concludes that existing regulatory frameworks for foods and supplements are not appropriate for cannabidiol, will work with Congress on a new way forward. link
- 12Damkier P, Lassen D, Christensen MMH, et al. (2019). Interaction between warfarin and cannabis. Basic & Clinical Pharmacology & Toxicology. link
- 13Watkins PB, Church RJ, Li J, Knappertz V (2021). Cannabidiol and abnormal liver chemistries in healthy adults: results of a phase I clinical trial. Clinical Pharmacology & Therapeutics. link
CBD (Cannabidiol) Products
Swallow a gummy or drop a tincture. Trivial daily effort.
Typical 25 mg gummy regimen $200–$800/year; reaching a studied 300 mg/day anxiety dose runs ~$450–$2,700/year at $0.05–$0.30 per mg consumer pricing. Not insurance-covered (Epidiolex aside).
Strong RCT evidence exists only for paediatric epilepsy at 10–20 mg/kg/day (Devinsky 2017, 2018). For the consumer indications (anxiety, sleep, pain), evidence is sparse, dose-mismatched, or uncontrolled; Cochrane and Lancet Psychiatry both find it insufficient at scale (Mücke 2018, Black 2019).
Acute single-dose anxiolysis at 300–600 mg is reproducible in provocation paradigms (Bergamaschi 2011, Linares 2019), and a small clinical case series reports anxiety/sleep improvement at 25–175 mg/day (Shannon 2019); chronic, controlled effects at retail consumer doses (5–25 mg) are essentially unestablished. Real but small for the typical buyer.
Sleep evidence at consumer doses rests on an uncontrolled retrospective case series (Shannon 2019, 67% short-term improvement that fluctuated) and a 2019 Lancet Psychiatry systematic review concluding the evidence is insufficient. RCT-grade sleep data at 5–25 mg/serving is absent.
Bergamaschi 2011 showed CBD 600 mg cut simulated-public-speaking anxiety in 24 social-phobia patients; Linares 2019 found an inverted-U dose-response (300 mg helped; 100 and 900 did not). Black 2019 (Lancet Psychiatry) concluded evidence for anxiolysis in primary anxiety disorders is insufficient for recommendation. Real acute signal at hundreds of mg, near-absent at retail doses.