What does the research say?
A 2025 meta-analysis that pooled 17 studies reported that cannabinoids slightly reduced itching; it found no clear difference in dryness, redness, quality of life or atopic dermatitis severity. This review is not in the table below: none of the 6 controlled studies that gave the itching result used a product containing CBD or THC. They were creams containing PEA, a cannabinoid-like substance also found in the body, hemp seed oil and synthetic research compounds; when the PEA creams were looked at separately, there was no clear difference in itching. The two CBD studies in the meta-analysis appear as separate rows in the table.
In a study of products applied to the skin in atopic dermatitis, clear improvement was seen only with the product in which CBD was combined with another substance; no clear difference was reported between the cream containing CBD alone and an inactive cream (placebo). In a study of 28 people with psoriasis, CBD oil taken by mouth did not improve the score that shows disease severity; itching decreased at week 8.
In a crossover study of 14 dialysis patients with itching, nabilone, a synthetic THC-like medicine, was taken by mouth for 4 weeks. Itching was no different from placebo; the worst-itch score was slightly higher with nabilone, and the impact of the skin problem on daily life was slightly worse with nabilone. Sleepiness and dizziness were similarly common in the two periods. This page previously described a 60-person study of a CBD-dominant cream in the same patient group; according to the article, the oil in the cream was obtained by pressing hemp seeds and contained only 1.27 mg of CBD and 0.13 mg of THC per kilogram. Because hemp seed oil is not considered a CBD or THC product, this study was removed from the table. The cream had reduced itching slightly more than placebo, but the reduction in itching from baseline was not significantly different between the two groups.
In 69 people with recurrent canker sores, a product containing 0.1% CBD was applied to the sores three times a day for 7 days. The sores shrank faster than with placebo, and pain was lower on day 5; a commonly used steroid product was similarly effective. This is a single, small study of sores inside the mouth.
The only controlled study in acne is a 368-person study of a CBD solution applied to the skin (below); as of October 2026, its results had not been published in a scientific journal. Apart from that, there are cell studies and small studies without a comparison group using products that contain CBD together with substances already used for acne, such as salicylic acid. A 2026 systematic review also found the evidence in acne to be preliminary; this review does not include the 368-person study. Effects seen with cosmetic products cannot be attributed directly to CBD because of the other ingredients they contain.
The results of a controlled study of a CBD solution applied to the skin in 368 people with acne were entered in the registry: after 84 days, the number of inflamed pimples had dropped by 8.5 to 12 lesions in the CBD groups and by 11.3 lesions with the same solution without CBD. In other words, the CBD solutions were not better; as of October 2026, these results had not been published in a peer-reviewed journal.
In a 200-person controlled study in atopic dermatitis whose results were entered only in the registry, a 4% CBD solution was applied twice a day for 12 weeks. The proportion of people whose skin became clear or almost clear was 12% with the CBD solution and 19% with the same solution without CBD; in other words, CBD was not better. The study was run by the product's manufacturer. Two studies of cosmetic moisturizers combining CBD with ginger extract were not included in the table: one had no group using CBD alone, and the product was found to be similar to a moisturizer on the market in skin measurements; the other had no comparison group. Both were run by the company that developed the product.
Epidermolysis bullosa is an inherited disease in which the skin blisters and wounds very easily. In a crossover study of 8 adults with pain, an oil containing THC and CBD dropped under the tongue did not change the emotional burden of pain, the study's main measure, differently from an inactive oil (placebo); on secondary measures, 6 of the 8 people reported at least a 30% reduction in pain, and painkiller use was halved. Mild side effects were common. In a study without a comparison group of 10 children (according to the registry) taking pure CBD by mouth, itching and pain decreased, and sleep and quality of life improved.
The only controlled study of CBD applied to the skin in psoriasis is from Thailand: in 51 people with mild psoriasis, a 2.5% CBD ointment was applied to one half of the body and an inactive ointment to the other half twice a day for 12 weeks. According to two reviews, the psoriasis severity score was clearly lower on the side treated with CBD. However, the study was published in the journal as a short letter, has no abstract, and its full text could not be obtained; because the result could not be verified from the source itself, it is not in the table. In 50 people with scalp psoriasis or seborrheic dermatitis, a CBD shampoo reduced redness, scaling and itching in 2 weeks; but this study had no comparison group.
In a placebo-controlled study of movement symptoms in people with Parkinson's disease, a CBD-rich extract that also contained a little THC was taken by mouth for about 16 days. In a secondary analysis in which photos of the face were also examined, seborrheic dermatitis (flaking and redness on the face) showed no clear difference between those taking CBD and those taking placebo. The disease was mild at the start, and the study was not large enough to answer this question.
In a study of 45 patients with painful finger ulcers due to systemic sclerosis (a connective tissue disease in which the skin and internal organs harden), pain and sleep improved more in the 25 people who applied CBD oil to their ulcers than in the 20 who received standard care alone, and complete healing of the ulcers was more common. However, the groups do not appear to have been formed at random, according to a table in the publication the CBD group also took CBD oil by mouth, and some numbers in the publication do not agree with one another. A 2024 systematic review pooling studies on skin wounds wrote that most human studies were case reports and small series and that the evidence was of low quality; the series on leg ulcers were done with gels that contained other active ingredients in addition to THC and CBD, together with compression bandaging.
For inherited hair loss (androgenetic alopecia) there are only two case series by the same doctor, without a comparison group: an increase in hair count over 6 months was reported with CBD-rich hemp extracts applied to the scalp once a day (35 and 31 people). The assessment was not blinded; these results do not show that CBD treats hair loss.
Strength of evidence: ●○○○ Very weak evidence. Finding: Mixed results. What does this mean?
Ongoing research on this topic
Trials testing CBD for this topic on ClinicalTrials.gov: 5 under way (2 recruiting). A trial being under way does not show that CBD works.
- NCT07245238: Phase 3, recruiting, target 36 participants
- NCT06435299: Phase 2, not yet recruiting, target 218 participants
- NCT06994520: Phase 2, enrolling by invitation, target 60 participants
Possible side effects and drug interactions
In the psoriasis study, side effects were mild or moderate and similar in the two groups. Cases of rash linked to CBD taken by mouth have been reported. In a 20-person series with a CBD ointment and no comparison group, no irritation or allergy was reported; no side effects were seen in the canker sore study either. In the 200-person eczema study, redness at the application site was slightly more common with the CBD solution (7% vs. 1%); in the 368-person acne study, more people dropped out in the CBD groups. In systemic sclerosis, 7 of the 25 people who applied CBD oil to their ulcers had mild itching or redness; in epidermolysis bullosa, mild side effects were common with the THC+CBD oil under the tongue. In dialysis patients, sleepiness and dizziness were seen with nabilone, but they were similarly common in the placebo period.
Drug interactions were not studied in these trials. CBD taken by mouth can slow the breakdown of some medicines in the body; people who take medicines regularly should keep this in mind.
Interactions with medicines used for this condition
Interactions reported for these medicines in peer-reviewed publications and official prescribing information. A medicine that is not listed can still interact with CBD; tell your doctor about every medicine you take.
| Medicine | Importance | Warning | Source |
|---|---|---|---|
| Cyclosporine | Information | Cyclosporine: If you take cyclosporine, tell your transplant team about CBD. In a small study, cyclosporine levels stayed stable; data are limited. | PMID 29579828 |
To check interactions with the medicines you take: CBD drug interaction checker.
How was CBD used in studies?
In the psoriasis study, CBD oil was taken by mouth at 60 mg a day (study dose; not a personal recommendation). In the canker sore study, the 0.1% CBD product was applied three times a day (study dose; not a personal recommendation). In other products applied to the skin, the CBD content ranged from 0.075% (shampoo) to 5% (solution), and the products were applied once or twice a day; according to the publication, the oil applied to systemic sclerosis ulcers contained 10% CBD (in acid form), while some cosmetic product studies did not report the amount of CBD. In children with epidermolysis bullosa, pure CBD by mouth was, according to the registry, started at 5 mg per kilogram per day and increased to 20 mg (study dose; not a personal recommendation). Because the products and doses differ so much, it is not possible to speak of a specific amount of CBD used for the skin.
Study details (for health professionals)
No effect on itching attributable to CBD or THC has been shown. Two sources previously described on this page as a signal of a small reduction were reread, found not to have been done with a cannabinoid product, and removed from the table. The small reduction (SMD −0.29) in a 2025 meta-analysis (Sermsaksasithorn 2025, PMID 41181596) came from 6 studies with products that contained no CBD or THC (PEA creams, hemp seed oil, and the synthetic research compounds HU-210 and lenabasum). The cream in the 60-person study that reduced itching in dialysis patients 1.1 points more than placebo (Anumas 2024, PMID 39328960) was made with hemp seed oil pressed from the seeds (1.27 mg of CBD and 0.13 mg of THC per kilogram of oil), and the difference in reduction from baseline was not significant. In the only RCT with a cannabinoid in dialysis itch, the THC-like nabilone was no better than placebo (14 people, crossover). In recurrent canker sores, a 0.1% CBD product shrank the sores faster than placebo (69 people, 7 days). In atopic dermatitis, in the manufacturer's 200-person RCT, a 4% synthetic CBD solution was no better than vehicle at 12 weeks (clear or almost clear: 12% vs. 19%; results only on ClinicalTrials.gov, no statistical comparison); in a small RCT completed by 57 people, the cream containing CBD alone was no different from placebo, and improvement was seen only with the combination of CBD and aspartame. In psoriasis, oral CBD oil did not improve PASI (28 people). Two reviews report that a placebo-controlled study comparing the two halves of the body in 51 people with a 2.5% CBD ointment applied to the skin (Puaratanaarunkon 2022, PMID 35536599) had a positive result; the publication has no abstract, its full text could not be obtained, and the result could not be verified. In a secondary analysis of an RCT with a CBD-rich, low-THC extract in people with Parkinson's disease, there was also no difference in seborrheic dermatitis (53 people, 16 days). In acne there is no peer-reviewed published RCT result; in the results of the only 368-person RCT (BTX 1503), available only on ClinicalTrials.gov, the CBD arms were no better than vehicle. The 2026 systematic review in acne (Komane) found the evidence preliminary; this review screened only 50 records and does not include the BTX 1503 registration. In epidermolysis bullosa, in an 8-person crossover RCT with THC+CBD oil under the tongue, the main measure (emotional burden of pain) did not change; in an uncontrolled study of 10 children taking pure CBD by mouth, itching and pain decreased. For finger ulcers in systemic sclerosis, CBD oil reduced pain in a non-randomized comparison; for scalp psoriasis, seborrheic dermatitis and hair loss there are only uncontrolled series. The studies are small and short; the effect of cosmetic products cannot be attributed to CBD because of the other ingredients they contain. October 2026 search (PubMed, Cochrane, ClinicalTrials.gov, Europe PMC): 9 controlled studies (8 randomized), 5 uncontrolled studies and 2 systematic reviews with cannabis products and cannabinoids on this topic were found; all are in the table below. There is no Cochrane review. Not included in the table: reviews examining cannabinoids in dermatology as a whole (most of the controlled studies they pooled were done with PEA creams, hemp seed oil or synthetic lenabasum; the CBD studies within them are separate rows in the table), studies with hemp seed oil (seed oil is not a CBD or THC product; it contains cannabinoids only in trace amounts), lenabasum, a synthetic drug candidate tested in systemic sclerosis and dermatomyositis (not a plant cannabinoid; in a 365-patient phase 3 study it was no different from placebo on the main measure), studies of cosmetic and wound products that contain CBD together with ginger, salicylic acid or other active ingredients and have no CBD-only group or no comparison group, and 2 randomized studies whose results could not be read or have not been released (see unverifiable points).
Evidence assessment: Very weak; conflicting (RCTs with CBD applied to the skin in acne and atopic dermatitis, with oral CBD in psoriasis, and with a CBD-rich extract in seborrheic dermatitis were negative; positive results come from a single small RCT in canker sores, a cream in which CBD was combined with another substance, a psoriasis RCT whose result could not be verified, and small non-randomized or uncontrolled studies)
CBD alone
Controlled evidence for CBD in the skin is very limited and conflicting: in canker sores, a 0.1% CBD product shrank the sores quickly; in psoriasis, 60 mg/day of oral CBD did not improve PASI; in atopic dermatitis, no significant difference was reported for the CBD-only arm. In the manufacturer's RCTs, a CBD solution applied to the skin was no better than the solution without CBD in acne (368 people) and atopic dermatitis (200 people) (results only in the registry, no statistical comparison). Two reviews report that a 51-person controlled study of a 2.5% CBD ointment applied to the skin in psoriasis was positive; the publication itself could not be read. Other positive results come from small, non-randomized or uncontrolled studies: finger ulcer pain in systemic sclerosis (45 people) and itching and pain in children with epidermolysis bullosa (10 people).
Products containing THC
There are two small randomized studies with THC-containing products, and neither found a difference from placebo: in epidermolysis bullosa, THC+CBD oil under the tongue did not change the emotional burden of pain, the main measure, differently from placebo (8 people; pain and painkiller use decreased on secondary measures); in people with Parkinson's disease, a CBD-rich extract that also contained a little THC did not change seborrheic dermatitis differently from placebo in a secondary analysis (53 people). The synthetic THC-like nabilone was also no better than placebo for itching in dialysis patients (14 people). For hair loss, there are only uncontrolled series with hemp extracts containing trace amounts of THC. The dialysis itch cream previously described in this box was made with hemp seed oil pressed from the seeds (0.13 mg of THC per kilogram of oil); because seed oil is not considered a THC product, it is no longer assessed here.
Evidence: Very weak (two small RCTs, both negative, one only a secondary analysis; a small RCT with the THC-like drug nabilone was also negative; the rest uncontrolled series)
Dose range in studies
Oral 60 mg/day (psoriasis); 0.075–5% CBD shampoo, solution or cream on the skin 1–2 times a day (acne, eczema, scalp); oral 5–20 mg/kg/day in children (epidermolysis bullosa, uncontrolled); 0.1% CBD 3 times a day (canker sores); concentration not reported for some cosmetic products
Starting dose and titration
Not reported in detail in the studies.
These are doses researchers gave under medical supervision; they are not a recommended dose for you. "mg/kg/day" means milligrams per kilogram of body weight per day.
| Study | Participants | Product | Dose and titration | Duration | Result | Source |
|---|---|---|---|---|---|---|
| Roongpisuthipong 2026 Double-blind, placebo-controlled RCT | 28 Psoriasis | Oral CBD oil | 60 mg/day not reported | Not reported (results up to week 8) | No significant improvement in PASI; itching decreased at week 8, and time to fall asleep was temporarily shortened Side effects: Mild to moderate, similar between groups | PMID 41459647, doi:10.1080/09546634.2025.2604448 |
| Umpreecha 2023 Randomized controlled trial, 3 arms (CBD / steroid / placebo) | 69 patients with canker sores (plus 100 and 50 healthy volunteers for safety) Recurrent canker sores (mouth ulcers) | 0.1% CBD oral paste applied to the sore (CBD from a 5% hemp extract donated by a company; purity and THC content not reported) | 3 times a day | 7 days | Sore size decreased more with CBD and with the steroid than with placebo on all measurement days; pain was lower with CBD than placebo only on day 5; no difference in quality of life Side effects: No allergy or side effects seen; blood values did not change | PMID 36803360, doi:10.1186/s12906-023-03886-0, TCTR20220802004 (retrospective registration) |
| Gao 2022 Randomized, double-blind, placebo-controlled, 3 groups (CBD+aspartame / CBD alone / placebo) | 57 (completed) Atopic dermatitis | Topical: CBD+aspartame (JW-100), CBD alone, placebo | not reported | 14 days | Significant improvement only with the CBD+aspartame combination; no significant difference reported for the CBD-alone arm (clear/almost clear: CBD 20%, placebo 15%; decrease in ISGA score 0.81 with CBD alone and 0.71 with placebo). 3 of the 13 authors were linked to a US company (Applied Biology) Side effects: Not reported (in the abstract) | PMID 34056830, doi:10.1111/jocd.14263 |
| Palmieri 2019 Retrospective, uncontrolled anecdotal series | 20 (5 psoriasis, 5 atopic dermatitis, 10 scars) Skin conditions | CBD ointment (THC-free) | Twice a day | 3 months | Improvement reported (no control) Side effects: No irritation/allergy reported | PMID 30993303, doi:10.7417/CT.2019.2116 |
| BTX 1503 (Botanix) Phase 2, randomized, quadruple-blind, vehicle-controlled (3 CBD arms + vehicle); no peer-reviewed publication (October 2026), results from the ClinicalTrials.gov record | 368 (92 in each arm) Moderate to severe acne | Topical CBD solution (2.5% or 5% CBD) | 5% twice a day, 5% once a day, or 2.5% once a day | 84 days | The primary outcome was safety only. At day 84, the reduction in inflammatory lesions was −8.5 / −12 / −11.7 in the CBD arms and −11.3 with vehicle; CBD not better than vehicle (statistical comparison not published) Side effects: No serious side effects; dropping out more common in the CBD arms (23, 15, 22 vs. 11 with vehicle) | NCT03573518 |
| BTX 1204 (Botanix) Phase 2, randomized, quadruple-blind, vehicle-controlled, multicenter (USA); sponsor Botanix, the owner of the product; no peer-reviewed publication (October 2026), results from the ClinicalTrials.gov record | 200 randomized (100 + 100); 99 and 95 in the primary analysis Ages 12–70, moderate atopic dermatitis (5–30% of body surface area) | BTX 1204: synthetic CBD solution applied to the skin (4%) | Twice a day | 84 days | Primary outcome (clear or almost clear on the investigator's global assessment at day 85 and at least a 2-grade improvement): 12 of 99 people (12%) with CBD, 18 of 95 (19%) with vehicle; CBD not better than vehicle (no statistical comparison in the registry). Completed the study: 65 vs. 55 of 100 people Side effects: Serious side effects in 2 of 99 people with CBD (postoperative complication, pneumonia), 0 of 98 with vehicle; application-site redness 7 vs. 1, eczema flare 11 vs. 15 | NCT03824405, results entered in the registry on September 27, 2022) |
| Collister 2026 (DISCO-POT) Double-blind, placebo-controlled, crossover RCT (2 × 4 weeks, with a 2-week break); two centers (Canada); investigator-initiated | 14 randomized; 12 completed Adults (aged 25 and over) on hemodialysis with at least moderate uremic pruritus | Nabilone (synthetic THC-like cannabinoid), oral capsule | 0.5 mg at night for 1 week, then 0.5 mg twice a day for 3 weeks From 1 capsule to 2 capsules a day after 1 week | 4 weeks nabilone, 4 weeks placebo | On the primary outcome (worst itch in the past 24 hours, 0–100 mm), nabilone was no different from placebo; if anything, 4.9 mm worse (95% CI −3.6 to 13.4; p = 0.26, 0.28 in the abstract). No difference in average itch, sleep or overall quality of life; the score for the impact of skin disease on life (DLQI) was 1.4 points worse with nabilone (95% CI 0.4 to 2.5; p = 0.01). Public funding (Canadian Institutes of Health Research); the authors reported no conflicts of interest Side effects: Sleepiness (4 vs. 5 of 14 people) and dizziness (3 vs. 2) similar in the two periods; no falls, fractures or serious side effects | PMID 42889824, doi:10.1016/j.xkme.2026.101451, NCT05180968 |
| Weber 2024 Secondary analysis of a double-blind, placebo-controlled, parallel RCT (the main study was on movement symptoms in Parkinson's disease; facial photos were assessed blind by two dermatologists) | 53 (26 CBD, 27 placebo); 12 and 15 people with seborrheic dermatitis at baseline Adults aged 40–85 with Parkinson's disease (USA) | Oral CBD-rich cannabis extract (100 mg/mL CBD and 3.3 mg/mL THC in sesame oil; supplied by the US National Institute on Drug Abuse) | 1.25 mg per kilogram per day (4 days), then 2.5 mg (about 10 days); on the last day, a mean of 188 mg CBD and 6.3 mg THC Single morning dose for 4 days, then twice a day | About 16 days | No significant difference from placebo in the presence and severity of seborrheic dermatitis (SEDASI) (treatment effect p = 0.12 and 0.38); people with seborrheic dermatitis fell from 12 to 9 in the CBD group and rose from 15 to 18 in the placebo group. The disease was mild at baseline, and the study was not large enough for this outcome. Funding: Colorado state medical marijuana research fund and NIH Side effects: No notable changes in blood pressure, pulse, temperature or blood tests (side-effect details in the main study) | PMID 38466972, doi:10.2196/49965, NCT03582137, results in the registry) |
| Schräder 2026 (C4EB) Double-blind, placebo-controlled, crossover RCT (2 × 14 days, with a 2-week break); single center (Netherlands); investigator-initiated | 8 adults Patients with inherited epidermolysis bullosa and chronic pain | Cannabis oil under the tongue (Transvamix: 10% THC, 5% CBD; a prescription product compounded in pharmacies in the Netherlands) | Dose increased until side effects appeared, divided into 4 doses a day; at most 0.75 mL/day (75 mg THC + 37.5 mg CBD) (according to the protocol) From 0.05 mL/day to 0.75 mL/day over 14 days; one step back if side effects appeared (according to the protocol) | 14 days THC+CBD, 14 days placebo | No significant difference from placebo on the primary outcome (emotional burden of pain, SF-MPQ2). 6 of 8 people (75%) reported at least a 30% reduction in overall pain, and painkiller use was halved during the active period (secondary outcomes). Numbers from the publication's short abstract; full text could not be read. Funding: DEBRA UK (patient association) Side effects: Mild side effects common (details in the full text; could not be read) | PMID 41742693, doi:10.1093/bjd/ljag072, Netherlands Trial Register NL9347, protocol: Schräder 2022, PMID 36508401 |
| Spinella 2023 Comparative, non-randomized (the publication says both "randomized" and "retrospective evaluation"); no blinding; single center (Italy) | 45 (25 CBD, 20 standard care) Painful, opioid-resistant finger ulcers in systemic sclerosis | CBD oil applied to the ulcer (according to the publication, 10% CBD in acid form + 90% hemp seed oil, THC-free); according to the table, the CBD group also took CBD oil by mouth | 4 drops once a day on the ulcer, with dressing | 1–2 months (CBD mean 1.2 months) | At 1 month, ulcer pain (0–10) fell from 8.4 to 6.0 with CBD and from 8.44 to 7.88 with standard care (between groups p < 0.0001); sleep duration and daily function (HAQ-DI) improved in the CBD group. Extra painkillers needed in 12 of 25 people vs. 20 of 20; ulcer completely healed in 18 of 25 vs. 6 of 20. Some numbers in the publication do not agree (final HAQ-DI given as 1.61 and 0.79 in two different places). The authors reported no financial relationships Side effects: Mild itching or redness around the ulcer in 7 of 25 people; no serious side effects | PMID 36537770, doi:10.1097/01.ASW.0000891856.08360.61 |
| Komane 2026 Systematic review (PRISMA, PROSPERO-registered; no meta-analysis) | 13 studies (mostly cell and laboratory studies; clinical studies few and mostly uncontrolled) Acne | CBD-containing hemp extracts and pure CBD | Found the clinical evidence preliminary: small, mostly non-randomized studies, different products and frequent use together with other active ingredients. Limitations: it screened only 50 records, does not mention the BTX 1503 registration, the only controlled study in acne, and one source it presents as a clinical finding is actually a cell study. No funding; the authors reported no conflicts of interest Side effects: Well tolerated in the short term; mild irritation in less than 10% (according to the review) | PMID 42357416, doi:10.3390/molecules31122017, PROSPERO CRD420251272093 | ||
| Niyangoda 2024 Systematic review (PROSPERO-registered; no meta-analysis) | 18 studies: 11 animal, 7 human (92 people) Skin wounds and skin infections (human studies: epidermolysis bullosa, pyoderma gangrenosum, calciphylaxis and venous leg ulcers, pressure ulcer, canker sores) | THC- and CBD-containing products applied to the skin or taken by mouth/under the tongue, crude extracts | Of the human studies, 1 was an RCT (canker sores) and the rest were case reports and small uncontrolled series; positive results were reported for pain and wound healing, but the evidence is of low quality and better studies are needed. 4 of the 10 authors reported company ties; one (Maida) is the chief scientific officer of a company developing cannabinoid products for wounds, and one works for a medical cannabis company | PMID 39204426, doi:10.3390/pharmaceutics16081081, PROSPERO CRD42021255413 | ||
| Mayrand 2026 Prospective, single-group, open-label pilot study (no comparison group); Paris | 10 children (according to the registry) Children with recessive dystrophic epidermolysis bullosa | Oral pure CBD solution (Epidyolex) | According to the registry, divided into 3 doses a day; started at 5 mg per kilogram per day and, if insufficient, increased to 10 and 20 mg (at most 800 mg/day) 5 mg/kg/day (days 0–4) → 10 (days 5–9) → 20 mg/kg/day (days 10–29), according to the registry | 30 days (according to the registry) | Mean chronic itch and pain, and itch and pain during dressing changes, decreased; quality of life and sleep clearly improved. Numbers could not be read (from the publication's short abstract and the registry; full text not accessible). No comparison group Side effects: No serious side effects reported | PMID 41105816, doi:10.1093/bjd/ljaf400, NCT05651607, September–November 2023, primary outcome itch at day 30, no results in the registry, the registry number could not be seen in the publication, and the link is based on matching institution, disease and design) |
| Vincenzi 2020 Prospective, single-group, open-label (no comparison group) | 50 Mild to moderate scalp psoriasis or seborrheic dermatitis | Shampoo containing 0.075% broad-spectrum CBD (THC content not in the abstract) | Used in place of the usual shampoo (frequency not in the abstract) | 14 days | Over 14 days, vascular findings and scaling scores on trichoscopy (0–5) fell from 2.3–3.6 to 0.5–0.8; itching fell from 6.9 to 1.5, burning from 4.5 to 1.0, redness from 5.5 to 1.3 and scaling from 7.0 to 1.6 (0–10) (all p < 0.0001); satisfaction was high. Because there was no comparison group, the effect cannot be attributed separately to the shampoo itself or to CBD. Manufacturer and funding not in the abstract; one of the authors is a consultant to various pharmaceutical and cosmetic companies (relationship to CBD not stated) Side effects: Tolerability reported as very good | PMID 33313051, doi:10.1159/000510896 |
| Smith 2021 Case series (single group, no comparison group); single center (USA) | 35 Adults with androgenetic alopecia (inherited hair loss) not using minoxidil or finasteride | Hemp oil applied to the scalp (from CBD-rich hemp flower powder; 10.78% CBD, 0.21% THC) | Once a day; mean 3–4 mg CBD/day | 6 months | Over 6 months, hair count in the area of greatest loss increased by a mean of 93.5%; some increase was reported in everyone. No comparison group, assessment not blinded; participants were recruited through a Facebook ad and received the product free of charge. No funding; the authors reported no conflicts of interest Side effects: Mild increase in shedding in one-third during the first month; no other side effects reported | PMID 37287996, doi:10.26828/cannabis/2021.01.003 |
| Smith 2023 Case series (single group, no comparison group); single center (USA) | 31 Adults with androgenetic alopecia (Norwood-Hamilton 3V and above) | Hemp extract foam applied to the scalp (CBD 60%, CBDV 12.6%, THCV 3.7%, THC 0.18%) + menthol and peppermint oil | Once a day; mean 33 mg extract/day | 6 months | Over 6 months, hair count per 1 cm² rose from 9.5 to 25.0 (mean increase 246% in men and 127% in women; p < 0.00001); an increase was reported in everyone. No comparison group; the sole author's institution is a company (Nex Therapeutics), and the publication states there is no conflict of interest; some numbers in the publication are inconsistent (e.g., the ethnic breakdown of participants differs between the abstract and the methods) Side effects: No side effects reported | PMID 37305187, doi:10.4103/ijt.ijt_34_22, NCT04842383, the publication describes a single group of 31, the link could not be verified, no results in the registry) |
Reported side effects
- Cases of rash linked to oral CBD use have been reported (Souza 2022, PMID 35662726, case series)
Who should be careful
- Most cosmetic products contain CBD together with other active ingredients (aspartame, ginger, CBG, retinol, salicylic acid, etc.); the effect of these products cannot be attributed to CBD
- In the canker sore study, CBD was applied to sores inside the mouth; this result cannot be carried over to skin diseases or to swallowed CBD
- Some products sold as "hemp oil" are oil pressed from the seeds and contain very little CBD or THC; the cream in the dialysis itch study was like this. Check whether the label and the certificate of analysis show the amount of CBD
Abbreviations: RCT, randomized controlled trial (a study in which participants are assigned to groups by chance); placebo, an inactive substance; SMD, OR and RR, effect measures used in analyses that pool studies; CYP, liver enzymes that break down medicines.
If you are considering CBD
- Do not stop the treatments you use for psoriasis, eczema or acne; there is no evidence that CBD can replace them.
- Check which ingredients a cream's label lists alongside CBD; the effect you see may come from those ingredients.
- If you are considering taking CBD by mouth and you take medicines regularly, talk with your doctor or pharmacist about drug interactions.
- If a rash appears or itching gets worse after a new product, stop using the product and see your doctor.
- Check the product's certificate of analysis (COA).
Choose your age group, medicines and situation to prepare a short summary you can take to your doctor. Nothing you enter is sent anywhere.
Bottom line
Research on CBD in skin conditions is still at an early stage. There are signs of improvement from small studies in canker sores inside the mouth and in finger ulcers in systemic sclerosis; results for itching are inconsistent. In acne and eczema, controlled studies did not show that CBD improves the condition; in psoriasis, CBD oil taken by mouth did not help either, and the only controlled study reported to show a positive result for a CBD ointment applied to the skin could not be verified. It is best to make treatment decisions about your skin together with your dermatologist.
Frequently asked questions
Does CBD oil work for skin conditions?
The evidence is very weak and the results are mixed. In a 200-person study in eczema, a CBD solution was no better than the same solution without CBD. Two results previously cited as positive for itching (a meta-analysis and a cream study in dialysis patients) turned out, on rereading, to come not from products containing CBD or THC but from cannabinoid-like substances such as PEA and from hemp seed oil. Positive results come from a small study of a CBD product in canker sores inside the mouth and from how reviews report a psoriasis ointment study whose results could not be obtained. In psoriasis, CBD oil taken by mouth did not improve disease severity; in acne, in the only controlled study of a CBD solution applied to the skin, the CBD solution was no better than the same solution without CBD.
Does CBD help with itching?
The evidence is very weak and the results are inconsistent. In a 14-person controlled study in dialysis patients with nabilone, a THC-like medicine, itching was no different from placebo. In a 28-person study of oral CBD oil in psoriasis, itching decreased at week 8, but itching was not the study's main measure; in 10 children with epidermolysis bullosa, itching decreased with oral CBD, but this study had no comparison group. The cream that slightly reduced itching in dialysis patients and was previously described as CBD-dominant was made with hemp seed oil pressed from the seeds; the oil contained only trace amounts of CBD and THC.
Does CBD cream help eczema?
In a study in atopic dermatitis, no clear difference was reported between a cream containing CBD alone and an inactive cream (placebo); improvement was seen only with the product in which CBD was combined with another substance. In another 200-person study whose results were entered only in the registry, the proportion of people whose skin became clear or almost clear after 12 weeks with a 4% CBD solution was not higher than with the same solution without CBD (12% vs. 19%).
Does CBD help acne?
A controlled study of a CBD solution applied to the skin was done in 368 people with acne; its results were entered only in the ClinicalTrials.gov registry and, as of October 2026, had not been published in a peer-reviewed journal. After 84 days, the number of inflamed pimples had dropped to a similar extent with the CBD solutions and the solution without CBD. Apart from that, there are cell studies and small studies without a comparison group using products that contain CBD together with substances already used for acne, such as salicylic acid; the effect in these cannot be attributed to CBD.
Does CBD oil taken by mouth help psoriasis?
In a 28-person study, CBD oil taken by mouth did not improve the score that shows psoriasis severity. Itching decreased at week 8, but this is a single, small study. Two reviews report that a controlled study of a 2.5% CBD ointment applied to the skin in 51 people had a positive result, but because the publication itself could not be obtained, this could not be verified. In a study without a comparison group in scalp psoriasis, symptoms decreased with a CBD shampoo.
Does CBD cause side effects on the skin?
Cases of rash linked to CBD taken by mouth have been reported. In the psoriasis study, side effects were mild or moderate and similar to the placebo group. With CBD products applied to the skin, redness or itching at the application site can occur: in the 200-person eczema study, redness was slightly more common with the CBD solution (7% vs. 1%).
Can CBD oil be obtained with a prescription for skin conditions in Turkey?
In Turkey, products derived from cannabis are regulated by the Kenevirden Elde Edilen Ürünlere Dair Yönetmelik (Regulation on Products Derived from Hemp) dated January 31, 2026. These products, including CBD, can be sold only in pharmacies. Those classed as medicines and health products are dispensed with a prescription; the regulation does not state a prescription requirement for supplement products and cosmetics with a THC content of no more than 0.3%. As of October 2026, there is no approved (licensed) CBD medicine in Turkey. If you are thinking of using it, talk with your doctor first.
References
Show 16 references
- Roongpisuthipong 2026. Double-blind, placebo-controlled RCT. PMID 41459647, doi:10.1080/09546634.2025.2604448
- Umpreecha 2023. Randomized controlled trial, 3 arms (CBD / steroid / placebo). PMID 36803360, doi:10.1186/s12906-023-03886-0, TCTR20220802004 (retrospective registration)
- Gao 2022. Randomized, double-blind, placebo-controlled, 3 groups (CBD+aspartame / CBD alone / placebo). PMID 34056830, doi:10.1111/jocd.14263
- Palmieri 2019. Retrospective, uncontrolled anecdotal series. PMID 30993303, doi:10.7417/CT.2019.2116
- BTX 1503 (Botanix). Phase 2, randomized, quadruple-blind, vehicle-controlled (3 CBD arms + vehicle); no peer-reviewed publication (October 2026), results from the ClinicalTrials.gov record. NCT03573518
- BTX 1204 (Botanix). Phase 2, randomized, quadruple-blind, vehicle-controlled, multicenter (USA); sponsor Botanix, the owner of the product; no peer-reviewed publication (October 2026), results from the ClinicalTrials.gov record. NCT03824405, results entered in the registry on September 27, 2022)
- Collister 2026 (DISCO-POT). Double-blind, placebo-controlled, crossover RCT (2 × 4 weeks, with a 2-week break); two centers (Canada); investigator-initiated. PMID 42889824, doi:10.1016/j.xkme.2026.101451, NCT05180968
- Weber 2024. Secondary analysis of a double-blind, placebo-controlled, parallel RCT (the main study was on movement symptoms in Parkinson's disease; facial photos were assessed blind by two dermatologists). PMID 38466972, doi:10.2196/49965, NCT03582137, results in the registry)
- Schräder 2026 (C4EB). Double-blind, placebo-controlled, crossover RCT (2 × 14 days, with a 2-week break); single center (Netherlands); investigator-initiated. PMID 41742693, doi:10.1093/bjd/ljag072, Netherlands Trial Register NL9347, protocol: Schräder 2022, PMID 36508401
- Spinella 2023. Comparative, non-randomized (the publication says both "randomized" and "retrospective evaluation"); no blinding; single center (Italy). PMID 36537770, doi:10.1097/01.ASW.0000891856.08360.61
- Komane 2026. Systematic review (PRISMA, PROSPERO-registered; no meta-analysis). PMID 42357416, doi:10.3390/molecules31122017, PROSPERO CRD420251272093
- Niyangoda 2024. Systematic review (PROSPERO-registered; no meta-analysis). PMID 39204426, doi:10.3390/pharmaceutics16081081, PROSPERO CRD42021255413
- Mayrand 2026. Prospective, single-group, open-label pilot study (no comparison group); Paris. PMID 41105816, doi:10.1093/bjd/ljaf400, NCT05651607, September–November 2023, primary outcome itch at day 30, no results in the registry, the registry number could not be seen in the publication, and the link is based on matching institution, disease and design)
- Vincenzi 2020. Prospective, single-group, open-label (no comparison group). PMID 33313051, doi:10.1159/000510896
- Smith 2021. Case series (single group, no comparison group); single center (USA). PMID 37287996, doi:10.26828/cannabis/2021.01.003
- Smith 2023. Case series (single group, no comparison group); single center (USA). PMID 37305187, doi:10.4103/ijt.ijt_34_22, NCT04842383, the publication describes a single group of 31, the link could not be verified, no results in the registry)
Points that could not be verified from the full text
- In Umpreecha 2023, the CBD was prepared from a 5% hemp extract donated by a company; the purity and THC content of the extract are not stated in the publication
- A two-center, placebo-controlled RCT of a 2.5% CBD ointment applied to the skin in psoriasis (Puaratanaarunkon 2022, Thailand; letter to the journal) could not be included in the table: it has no abstract and its full text could not be accessed. According to two reviews, in 51 people with mild psoriasis the ointment was applied to one half of the body and an inactive ointment to the other half twice a day for 12 weeks, and the psoriasis severity score was significantly lower on the side treated with CBD (p = 0.026); this result could not be verified from the source itself
- A 19-person randomized phase 2 study of a CBN cream (INM-755) in epidermolysis bullosa (NCT04908215) ended in April 2023; as of October 2026 there are no results in the registry and no peer-reviewed publication. There are also no results for a CBD study in chronic urticaria (NCT04439955, 15 people, ended in 2021)
- Full texts that could not be read (publisher access blocked or paywalled): Schräder 2026 and Mayrand 2026 (results from the publication's short abstract and the registry), Vincenzi 2020, Gao 2022 and Roongpisuthipong 2026 (from the abstract)
- Ongoing or not yet started studies (October 2026): 1% CBD lotion for itch in systemic sclerosis (NCT07245238), CBD compared with a steroid cream in eczema (NCT06994520), oral and topical CBD:CBG:THC in psoriasis (NCT07839455), THC:CBD oil for severe itch (NCT06435299, start January 2027 according to the registry); no results
Research history
Here we record what changed on this page, when and why, with dates. The newest change is at the top.
In October 2026, PubMed, Cochrane, ClinicalTrials.gov and Europe PMC were systematically searched; 11 studies were added: 4 randomized studies (the manufacturer's 200-person study in atopic dermatitis, nabilone for dialysis itch, seborrheic dermatitis in people with Parkinson's disease, THC+CBD oil in epidermolysis bullosa), a non-randomized comparison in systemic sclerosis ulcers, 2 systematic reviews (acne, skin wounds) and 4 uncontrolled studies (epidermolysis bullosa, scalp, hair loss). Two existing rows were reread and removed from the table: the itch reduction in the 2025 meta-analysis came from products containing no CBD or THC (PEA creams, hemp seed oil, synthetic research compounds), and the cream in the dialysis itch study was hemp seed oil pressed from the seeds and contained only trace amounts of CBD. The positive account, in two reviews, of a psoriasis study whose results could not be obtained (a CBD ointment applied to the skin) was added to the unverifiable points; product sources, company ties and publication dates were corrected. The conclusion and the strength of evidence did not change.
Related studies: BTX 1204 (Botanix) (NCT03824405, results entered in the registry on September 27, 2022)); Collister 2026 (DISCO-POT), published November 2026 (online June 23, 2026) (PMID 42889824, doi:10.1016/j.xkme.2026.101451, NCT05180968); Weber 2024, published March 11, 2024 (PMID 38466972, doi:10.2196/49965, NCT03582137, results in the registry)); Schräder 2026 (C4EB), published May 2026 (online February 25, 2026) (PMID 41742693, doi:10.1093/bjd/ljag072, Netherlands Trial Register NL9347, protocol: Schräder 2022, PMID 36508401); Spinella 2023, published January 2023 (PMID 36537770, doi:10.1097/01.ASW.0000891856.08360.61); Komane 2026, published June 9, 2026 (PMID 42357416, doi:10.3390/molecules31122017, PROSPERO CRD420251272093); Niyangoda 2024, published August 17, 2024 (PMID 39204426, doi:10.3390/pharmaceutics16081081, PROSPERO CRD42021255413); Mayrand 2026, published January 2026 (online October 17, 2025) (PMID 41105816, doi:10.1093/bjd/ljaf400, NCT05651607, September–November 2023, primary outcome itch at day 30, no results in the registry, the registry number could not be seen in the publication, and the link is based on matching institution, disease and design)); Vincenzi 2020, published November 2020 (PMID 33313051, doi:10.1159/000510896); Smith 2021, published April 2021 (PMID 37287996, doi:10.26828/cannabis/2021.01.003); Smith 2023, published 2023 (PMID 37305187, doi:10.4103/ijt.ijt_34_22, NCT04842383, the publication describes a single group of 31, the link could not be verified, no results in the registry))
PubMed and ClinicalTrials.gov were searched for newly published and previously missed clinical studies. 2 studies added: Anumas 2024 and Umpreecha 2023. With the new studies, the strength of evidence changed from "Insufficient" to "Very weak", and the conclusion from "Not studied enough" to "Mixed results". The strength of evidence for THC-containing products was written for the first time, as "Very weak".
Studies that led to the change: Anumas 2024, published October 2024 (PMID 39328960, doi:10.1016/j.xkme.2024.100894, NCT06159686); Umpreecha 2023, published February 20, 2023 (PMID 36803360, doi:10.1186/s12906-023-03886-0, TCTR20220802004 (retrospective registration))
The full texts of the studies were reread and the information on the page was checked against its sources. Corrected: registry number, study design, participant numbers and side-effect information in the study table; evidence summary. No points remain that could not be verified from full texts (previously 1 item). The results of the BTX 1503 study were added from its ClinicalTrials.gov record. The year of 2 publications named on the page was corrected to the journal's print year instead of the online publication year. The conclusion and the strength of evidence did not change.
Older changes (1)
Page published. In this first assessment, 5 studies were reviewed.