CBD (cannabidiol) is sold almost everywhere as a wellness product, but the clinical evidence is much narrower than the marketing. This guide explains how CBD acts in the body, how much of a dose reaches your blood, and which uses the trials support as of mid-2026.
CBD is now one of the most widely sold supplement ingredients in the world. Pharmacies, health food stores and online sellers carry it in dozens of countries, including Thailand, the Czech Republic and the rest of Southeast Asia. Since 2019, large trials have tested several popular claims, and some of them did not hold up. The evidence is not empty either, so it is worth knowing exactly what it shows.
Part 1 covers what CBD is, the evidence for each claimed use, the other non-intoxicating cannabinoids and terpenes. Part 2 covers the extract types named on labels, lab certificates, doses, drug interactions and the rules in Thailand and the European Union.
Translator’s note: This article is educational material, not medical advice. Growers hear CBD claims all the time from customers, buyers and seed catalogues. If you know which claims have been tested, you can avoid repeating the ones that have not. Any decision to use CBD for a health problem should be made with a doctor.
1. What CBD Is
1.1 Where CBD comes from
CBD is a phytocannabinoid, which means a cannabinoid made by a plant. It occurs in Cannabis sativa, both in drug-type cannabis and in hemp. Here, hemp means varieties high in CBD and low in THC (delta-9-tetrahydrocannabinol), the compound that causes intoxication.
THC causes intoxication by switching on the CB1 cannabinoid receptor. A receptor is a protein on a cell that responds when a specific compound binds to it. A compound that switches a receptor on is called an agonist. CBD does not directly switch on CB1, so at typical doses it does not make you high.
1.2 How CBD acts in the body
Researchers have not yet found every way CBD acts. Four mechanisms are established:
- TRPV1 receptor. At high concentrations, CBD switches on TRPV1, a receptor of the vanilloid type. This is linked to changes in pain and inflammation.
- 5-HT1A receptor. This is a receptor for serotonin, one of the brain’s signalling chemicals. CBD partly switches it on (it is a “partial agonist”). This may explain a possible anti-anxiety effect.
- GPR55 receptor. Scientists call GPR55 an “orphan receptor” because its natural trigger in the body is not settled. It responds to cannabinoids, and CBD blocks it. This may affect bone density and inflammation.
- Anandamide recycling. Anandamide is an endocannabinoid, which is a cannabinoid the body makes itself. Cells normally take it back in to end its signal, a step called reuptake. CBD slows this step, so anandamide keeps signalling longer even though CBD does not act on the cannabinoid receptors directly.
Because CBD acts on several targets, its effects are real, but they depend on the dose and on which target is involved. One effect may need a different dose from another.
What you can do: when you read a claim about CBD, find the dose used in the study behind it. An effect seen at one dose tells you little about a product that delivers a different dose.
1.3 How much of a dose reaches the blood
Bioavailability is the share of a dose that reaches the bloodstream. For CBD it varies widely with the route, meaning the way you take it. So the route decides how much the dose on a label is actually worth.
| Route | Share that reaches the blood | Time until it starts working | Notes |
|---|---|---|---|
| Swallowed (oil or capsule) | 6–19% | 1–2 hours | Varies widely between people; food increases absorption |
| Under the tongue (sublingual) | 12–35% | 15–45 minutes | Part is absorbed in the mouth; the rest is swallowed |
| Inhaled (vapour) | 31–45% | 1–5 minutes | Fastest route; lung health has to be considered |
| On the skin (topical) | Less than 1% | 15–60 minutes | Acts locally; very little enters the blood |
| Nanoemulsion products | 35–80% (company claims) | 15–30 minutes | Each company’s own figure; independent checks vary |
These figures are approximate estimates taken from different studies, and the nanoemulsion figures are company claims. Few human studies have measured how much CBD reaches the blood directly. A systematic review (Millar et al., 2018) found only one such study, which reported 31% after smoking.
A nanoemulsion is a product in which the oil is broken into very fine droplets so that it mixes into a water-based liquid. The higher figures for these products come from the companies themselves.
Here is a worked example. A 25 mg CBD capsule that you swallow puts only about 1.5–5 mg of CBD into your blood, because only an estimated 6–19% is absorbed. Many wellness products are dosed so low that blood levels stay below those in the trials that showed an effect. The marketing does not always say so.
What you can do: to compare two products, multiply the dose on each label by the absorbed share for its route. A 25 mg capsule and a 25 mg inhaled dose are not the same dose.
2. What the Evidence Shows
How much a result counts depends on the type of study. These terms appear throughout this section:
- Randomised controlled trial: people are assigned by chance to the product or to a placebo, a dummy that contains no active ingredient. Any difference between the groups can then be credited to the product.
- Meta-analysis: a study that pools the results of several earlier studies to get a bigger sample.
- Case series: a report on a group of patients with no comparison group. A retrospective case series looks back at existing records.
- Preclinical study: research in cells or animals, done before testing in people.
- Observational study: a study that records what people already do, without assigning any treatment.
2.1 Proven: two rare childhood epilepsies
Epidiolex is a pharmaceutical CBD medicine. It is approved by the US Food and Drug Administration (US FDA), the American medicines regulator. Several randomised controlled trials show that high doses work in Dravet syndrome and Lennox-Gastaut syndrome. These are rare, severe epilepsies that begin in childhood. The doses in those trials were 10–20 mg per kg of body weight per day.
This evidence is strong. It led to approval in the US, the EU and several other countries.
It says little about wellness products, though. Epidiolex delivers a steady, high dose of pharmaceutical-grade CBD, and its absorption is known. Wellness products deliver variable and usually much lower doses, made in less standardised ways.
To see the size of the gap, take a child who weighs 20 kg. At 10–20 mg/kg, that child receives 200–400 mg a day. Common wellness doses are 10–50 mg a day (see Part 2). So the epilepsy results do not support claims that low-dose CBD helps healthy adults with anxiety or sleep.
Translator’s note: Ask a neurologist whether a registered pharmaceutical CBD medicine can be obtained in Thailand. A wellness CBD oil cannot replace a registered epilepsy medicine prescribed by a neurologist.
What you can do: if someone asks you about CBD for a child with epilepsy, send them to a neurologist. This is the one use where a tested, approved medicine exists.
2.2 Mixed evidence: anxiety, sleep and joint pain
Anxiety. The two most recent reviews disagree. A 2024 meta-analysis pooled 8 studies with 316 people who had anxiety disorders and found that CBD reduced anxiety (Wang et al., 2024). The authors warned that the samples were small.
A 2026 meta-analysis of 54 randomised trials of cannabinoids for mental disorders found no significant effect on anxiety (Wilson et al., 2026). Neither review supports any specific effective dose.
Sleep. Shannon et al. (2019) reported better sleep scores in patients who took 25–175 mg of CBD. This was a retrospective case series, so there was no placebo group to compare with. Later randomised trials gave inconsistent results.
Some of those trials found better sleep quality and fewer anxiety-related sleep problems. Others found no difference from placebo. As of 2026, the sleep evidence is promising but not conclusive.
Inflammation and joint pain. Animal studies strongly support an anti-inflammatory effect. In people, a trial in rheumatoid arthritis (Blake et al., 2006) tested nabiximols (Sativex®, a mouth spray containing both THC and CBD), not CBD alone. In 58 patients over five weeks, it found modest but significant improvements in pain during movement and at rest compared with placebo. It is still one of the few rigorous arthritis trials.
Later trials in osteoarthritis and chronic pain gave mixed results. One reason is that the doses varied from trial to trial.
What you can do: when a seller says “studies show”, ask three questions. Was there a placebo group? How many people took part? What dose did they take? A small study, or one with no placebo group, is a lead and not proof.
2.3 Weak or inconsistent evidence
- General wellbeing and mood. No strong randomised trial shows that CBD improves wellbeing, energy or mood in healthy people who have no medical condition.
- CBD on the skin for skin problems and pain. There are very few human trials. Preclinical work supports use against skin inflammation. Very little CBD passes through healthy skin into the blood, so any effect is probably local at best.
- Sports recovery. Observational studies and surveys of athletes show that many use CBD for recovery and pain. Randomised trials are few, and their results are mixed.
What you can do: treat these uses as unproven. If you sell, recommend or post about CBD, do not present them as established benefits.
3. Other Non-Intoxicating Cannabinoids
These compounds show up more and more on product labels and in seed and clone catalogues. For each one, ask the same question: has it been tested in people?
3.1 Cannabigerol (CBG)
The plant first makes cannabigerolic acid (CBGA) and then converts it into the other cannabinoids. That is why CBG is marketed as the “stem cell” cannabinoid. Most cannabis and hemp contain only small amounts of CBG, but breeders have developed CBG-dominant varieties.
Most of the evidence comes from cells and animals. These studies suggest antibacterial activity, including against MRSA, a staphylococcus bacterium that resists common antibiotics (Appendino et al., 2008). They also suggest protection of nerve cells, anti-inflammatory effects and appetite stimulation.
In 2026, human trial data were still very limited. CBG products are being sold well ahead of the evidence.
Translator’s note: Before you buy CBG-dominant seed or clones, confirm which licence covers growing them in Thailand. If you are considering such a variety, grow a small trial lot first and send samples for a cannabinoid test. The test shows whether the plant really produces CBG in your climate before you commit a whole crop. Clones from other farms can bring in pests, hop latent viroid (HLVd) and the wrong variety, so keep new plants away from your main crop until you have checked them.
3.2 Cannabinol (CBN)
CBN forms when THC breaks down by reacting with oxygen, so old cannabis contains more CBN. CBN switches on CB1 only weakly, at about 10% of the strength of THC. It is often sold as a sleep aid.
There are few human trials. In a randomised trial of 293 adults who slept poorly, 20 mg of CBN did not significantly improve overall sleep quality compared with placebo (Bonn-Miller et al., 2024). It did reduce waking during the night and overall sleep disturbance. Adding 10–100 mg of CBD gave no extra benefit over CBN alone.
A 2026 crossover trial tested CBN in 20 adults with insomnia (Lavender et al., 2026). In a crossover trial, each person takes both the product and the placebo, at different times. Neither 30 mg nor 300 mg of CBN reduced the time spent awake during the night. At 300 mg, people fell asleep faster and rated their sleep better.
This means the popular “CBN for sleep” claim still has limited support.
Translator’s note: For growers, CBN is also a storage warning. CBN comes from broken-down THC, so a rising CBN result when you test stored flower means the THC is degrading. Heat, light and air speed this up, and an uncooled storeroom in Thailand often reaches 30–35 °C. Keep dried flower cool, dark and sealed; see storing dried flower.
3.3 Cannabichromene (CBC)
Animal studies suggest CBC reduces inflammation and pain. Rodent studies also suggest antidepressant-like effects. Human evidence is essentially absent.
CBC is one of the compounds in full-spectrum hemp extracts. Nobody has yet isolated it and tested it in a meaningful human trial.
3.4 Tetrahydrocannabivarin (THCV)
THCV is a close chemical relative of THC, but it acts differently at different doses. At low doses it partly blocks CB1 (it is a “partial antagonist”). At high doses it partly switches CB1 on (a “partial agonist”).
Animal studies suggest THCV may suppress appetite, help control blood sugar and reduce seizures. One randomised trial (Jadoon et al., 2016) found better blood sugar measures in people with type 2 diabetes. More human data are needed before strong claims are justified.
What you can do: treat any health claim for CBG, CBN, CBC or THCV as unproven in people. Ask the seller for the human trial behind the claim. For most of these claims there is none yet.
4. Terpenes and the Entourage Effect
4.1 What terpenes are
Terpenes are the aroma compounds that give cannabis its smell. They are volatile, which means they evaporate easily. Common ones are limonene (citrus), myrcene (earthy, musky), linalool (floral, like lavender), beta-caryophyllene (spicy, like black pepper) and pinene (pine). There are dozens of others.
The plant makes terpenes in the trichomes, the tiny resin glands that also produce cannabinoids. Many other plants used in traditional medicine contain the same terpenes.
At high enough concentrations, terpenes have effects of their own. Beta-caryophyllene is unusual because it directly switches on the CB2 cannabinoid receptor. By the receptor definition, that makes it technically a cannabinoid, even though its structure has nothing in common with THC or CBD. So full-spectrum extracts, which contain terpenes, may act differently from pure CBD.
4.2 What terpene research shows
Most terpene research has been done in cells and animals:
- Myrcene: sedative in rodents. It also makes the sleep caused by barbiturates (sedative drugs) last longer. It is often used to explain why “indica” strains feel sedating, but there is little direct human evidence.
- Limonene: reduces anxiety in rodents and may have antidepressant activity. Evidence in people is limited. A small 2024 trial in 20 healthy adults found that vaporised limonene taken with THC reduced the anxiety that THC caused.
- Linalool: sedative and anti-anxiety effects in several animal models. There are more human data for lavender oil, which is rich in linalool, than for linalool on its own.
- Beta-caryophyllene: reduces inflammation through CB2 and relieves pain in rodents.
The practical problem is that terpene amounts in commercial CBD products vary widely. Labels rarely give them as numbers. “Full-spectrum” names a type of product. It does not tell you how much of each compound is in the bottle.
Translator’s note: Terpenes evaporate easily, so heat during drying and storage costs you terpenes. Without air conditioning, rooms in Thailand often reach 30–35 °C. If you grow plants for full-spectrum extraction, ask the lab to report terpenes as numbers (for example, mg per gram). Your buyer can then see what the material actually contains. See drying without losing terpenes.
4.3 The entourage effect: a hypothesis, not a proven fact
The “entourage effect” is the idea that whole-plant extracts work better than single cannabinoids. The suggested reason is that cannabinoids, terpenes and other plant compounds reinforce each other. CBD marketing cites it often.
The idea comes mainly from Russo (2011), an influential paper that proposed how terpenes might change cannabinoid receptor activity. Cell and animal studies partly support it. Few human trials compare pure CBD (isolate) with full-spectrum extract at the same CBD dose, and the ones that exist give mixed results.
The hypothesis is plausible and has some support from cell and animal studies. It should not be presented as proven, as it often is. The terpenes and minor cannabinoids in full-spectrum extracts may add to the effect, or CBD may do most of the work. The answer may also differ from one condition to another and from one person to another.
What you can do: when a label promises an entourage benefit, look for terpene and minor cannabinoid amounts on its lab certificate. If those numbers are missing, compare products by their CBD dose alone.
Part 1 in Brief
- Proven: pharmaceutical CBD at 10–20 mg/kg/day for two rare childhood epilepsies, Dravet syndrome and Lennox-Gastaut syndrome.
- Mixed: anxiety, sleep and joint pain. No specific dose has been shown to work for anxiety.
- Weak: general wellbeing, mood, skin use and sports recovery.
- Absorption: only 6–19% of a swallowed dose reaches the blood. Compare products by dose multiplied by the absorbed share for each route.
- CBG, CBN, CBC and THCV: mostly cell and animal data. CBN trials in people show only partial sleep effects.
- Entourage effect: a plausible hypothesis, not a proven fact.
If you have questions, or lab results from CBD or CBG crops grown in Thailand, post them below this topic.
References
- Millar, S. A., Stone, N. L., Yates, A. S., & O’Sullivan, S. E. (2018). A systematic review on the pharmacokinetics of cannabidiol in humans. Frontiers in Pharmacology, 9, 1365. Frontiers | A Systematic Review on the Pharmacokinetics of Cannabidiol in Humans
- Wang, J.-Y., Wang, P.-Y., Han, K., & Peng, Y.-C.-H. (2024). Therapeutic potential of cannabidiol (CBD) in anxiety disorders: a systematic review and meta-analysis. Psychiatry Research, 339, 116049. Redirecting
- Wilson, J., et al. (2026). The efficacy and safety of cannabinoids for the treatment of mental disorders and substance use disorders: a systematic review and meta-analysis. The Lancet Psychiatry, 13(4), 304–315. Redirecting
- Shannon, S., et al. (2019). Cannabidiol in anxiety and sleep: a large case series. Permanente Journal, 23, 18–041.
- Russo, E. B. (2011). Taming THC: potential cannabis synergy and phytocannabinoid-terpenoid entourage effects. British Journal of Pharmacology, 163(7), 1344–1364.
- Appendino, G., Gibbons, S., Giana, A., et al. (2008). Antibacterial cannabinoids from Cannabis sativa: a structure−activity study. Journal of Natural Products, 71(8), 1427–1430. https://doi.org/10.1021/np8002673
- Bonn-Miller, M. O., et al. (2024). A double-blind, randomized, placebo-controlled study of the safety and effects of CBN with and without CBD on sleep quality. Experimental and Clinical Psychopharmacology, 32(3), 277–284. APA PsycNet
- Lavender, I. G., et al. (2026). Cannabinol for acute treatment of insomnia disorder in a randomized placebo-controlled crossover trial. Journal of Sleep Research, 35(4), e70284. https://doi.org/10.1111/jsr.70284
- Blake, D. R., Robson, P., Ho, M., Jubb, R. W., & McCabe, C. S. (2006). Preliminary assessment of the efficacy, tolerability and safety of a cannabis-based medicine (Sativex) in the treatment of pain caused by rheumatoid arthritis. Rheumatology, 45(1), 50–52. https://doi.org/10.1093/rheumatology/kei183
- Jadoon, K. A., Ratcliffe, S. H., Barrett, D. A., et al. (2016). Efficacy and safety of cannabidiol and tetrahydrocannabivarin on glycemic and lipid parameters in patients with type 2 diabetes: a randomized, double-blind, placebo-controlled, parallel group pilot study. Diabetes Care, 39(10), 1777–1786. https://doi.org/10.2337/dc16-0650
Continued in Part 2: CBD and Cannabinoid Wellness: Extract Types, Lab Certificates, Doses, Drug Interactions and the Rules (Part 2 of 2)