OCD, Hair Pulling and CBD Oil: What Does the Research Say?

What does the research say?

Researchers' interest in this topic comes largely from animal studies: in animals, substances that strengthen the body's own cannabinoid system (the endocannabinoid system) reduced compulsion-like behaviors, and substances that weaken it increased them. It has also been reported that some genes of this system are less active in the blood of people with OCD. These findings explain the search for new treatments, but they do not show a treatment effect in humans.

In OCD, the only study that compared cannabinoids with an inactive substance (placebo) and assigned participants to groups at random is a laboratory study. In the US, 12 adults with OCD who had used cannabis before smoked cannabis once on each of three separate days: THC-dominant, CBD-dominant (also containing a small amount of THC) or cannabis with no active ingredient; neither the participants nor the people assessing them knew who smoked what on which day. Obsessions and compulsions decreased over time on all three days, but there was no difference by type of cannabis. Momentary anxiety, on the other hand, decreased more after the inactive cannabis than after the THC-dominant or CBD-dominant cannabis. According to the researchers, this result shows how important placebo-controlled studies are for understanding the effect of cannabis. The study used a single dose; it does not show the effect of long-term use.

The same team also tested nabilone, a synthetic THC-like medicine, together with exposure and response prevention therapy, one of the most effective treatments for OCD. In all, 11 adults with OCD took nabilone for 4 weeks; 5 of them, chosen at random, also received therapy during the same period. OCD severity (the 0–40-point Yale-Brown scale, Y-BOCS) fell by an average of 2.5 points in those taking nabilone alone and by 11.2 points in those receiving therapy along with nabilone. However, this study had no group taking a dummy treatment or receiving therapy alone; so it is not known how much of the improvement came from the therapy and how much from nabilone. The first 3 participants left the study because of increased anxiety lasting 12–24 hours after the first dose; later participants were started on a lower dose.

In hair-pulling disorder, dronabinol, a synthetic form of THC, was first tried in a small study without a comparison group: in 14 women who pulled their hair, the hair-pulling score fell by about half over 12 weeks, and 9 people responded to treatment. The same team later tested it in 50 adults split at random into two groups; 34 of the participants pulled their hair and 16 picked their skin. At the end of 10 weeks, symptoms had clearly decreased in both groups, but dronabinol was not better than the dummy treatment on any measure: the treatment response was 67% with dronabinol and 50% with placebo, and this difference was not statistically significant. Of the 50 people, 20 (11 in the dronabinol group, 9 in the placebo group) did not complete the study; a study this small can miss a small difference. In other words, much of the improvement in the first study may have come from expectation and from change over time. A 2021 Cochrane review of medicine treatments for hair pulling also found no firm evidence for any medicine; there were no cannabinoid studies among the studies it reviewed.

The only trial of pure CBD in OCD and related disorders was carried out in the US with the approved (licensed) CBD medicine. This small study, open to adults diagnosed with OCD, hoarding, hair pulling or skin picking, or Tourette syndrome, was started by 4 people, and 2 of them completed it. The results were added to the ClinicalTrials.gov registry record in summer 2026; there is no peer-reviewed publication. In the 2 people who completed it, the average change in the OCD severity score was 2 points (0–40 scale; the registry record does not clearly state the direction of the change), and overall illness severity did not change; no one was assessed with the hair-pulling or skin-picking scale. No conclusion can be drawn from a study without a comparison group and with so few people.

In children, the only data come from pediatric acute-onset neuropsychiatric syndrome (PANS), a condition in which obsessions and compulsions start suddenly after infections. In Australia, 14 children took a cannabis extract containing mainly the acid form of CBD (CBDA, cannabidiolic acid) and a small amount of THC for 12 weeks; all of the children had OCD symptoms. Overall illness severity and a children's OCD scale improved clearly. However, there was no comparison group, the children kept taking their other medicines, PANS can get better and worse on its own, and the study was funded by the companies developing the product. The researchers write that a placebo-controlled study is being prepared.

A large 2026 review bringing together cannabinoid studies in mental health conditions did not find enough studies for a pooled analysis in OCD. A 2026 scoping review looking specifically at cannabinoids in OCD also wrote that most of the evidence comes from surveys and individual patient reports, that there are no well-controlled studies, and that the current evidence does not support using synthetic or natural cannabinoids to treat OCD. In THC studies with Tourette syndrome patients, obsessive-compulsive symptoms were examined only as an additional measure, and the results are mixed: in a placebo-controlled study in which 12 adults received a single dose of THC, these symptoms decreased more than with placebo on a symptom checklist filled in by the patients, but worsening was seen on another scale used in the same study; a study with 22 adults reported that these symptoms might decrease with an oil containing THC and CBD; in an uncontrolled study with THC-containing medical cannabis, obsessions worsened in one patient. The Cochrane review of cannabinoids in Tourette syndrome (2009) did not find enough evidence to support their use for tics and obsessive-compulsive behaviors. Details are on the site's Tourette Syndrome page.

Surveys show that many people with OCD who use cannabis report positive effects, but also that it carries significant risks. In a US survey answered by 601 adults with OCD who use cannabis, 29% of participants used cannabis specifically for OCD symptoms, and most said cannabis reduced their obsessions and compulsions; on the other hand, about 17% reported worsening obsessions, 42% met the criteria for cannabis use disorder (addiction), and the more cannabis people used, the less likely they were to receive proven OCD treatment. In data that 87 people who described themselves as having OCD logged in a phone app, users reported a clear drop in compulsions right after smoking cannabis; the reported drop was larger with products high in CBD. However, the diagnoses were not confirmed, there was no placebo group, and the effect on obsessions faded over time; because this study relied only on users' own records, it was not included in the table. At an OCD clinic in Italy, 43% of 70 patients had used cannabis at some point in their lives; about 10% of those who had used it reported improvement in OCD symptoms and 23% reported worsening anxiety. These findings rely on people's own reports and do not match the result of the placebo-controlled laboratory study.

In an online survey of people with hair-pulling or skin-picking disorder, about a third of participants had used cannabis in the past year; those who used it pulled their hair or picked their skin on more days per week. In a US study that followed 547 young people from age 17 to age 19, cannabis use at 17 was linked to slightly more obsessive-compulsive symptoms at 19. Individual patient reports have described both improvement with THC and obsessions that began with cannabis use. These are observational findings; they do not show cause and effect, but they suggest that cannabis does not always ease these symptoms.

Strength of evidence: ○○○○ Insufficient evidence. Finding: Not studied enough. What does this mean?

Possible side effects and drug interactions

In the controlled study in hair pulling and skin picking, mild or moderate side effects occurred in 16 of 25 people taking dronabinol and in 7 of 25 taking placebo; the most common were dizziness, sleepiness, a 'high' feeling, dulled attention and thinking, and dry mouth; there were no serious side effects. In people with OCD, smoked THC-containing cannabis raised heart rate and blood pressure; one participant who also had panic disorder had temporary panic symptoms. After the first dose of nabilone, 3 people stopped because of increased anxiety; once the dose was started lower, no one stopped. In the 4-person study with the approved CBD medicine, 1 person had diarrhea. With the CBDA-dominant extract used in children, a third of the children had nausea, diarrhea, tiredness or a delay in falling asleep.

Many people with OCD take antidepressants. CBD can slow the breakdown of some medicines in the body: in healthy volunteers, CBD increased the total amount of citalopram in the blood by 43%; low sodium and difficulty thinking were reported in a patient using CBD with sertraline. No human studies were found that looked at CBD together with fluvoxamine or clomipramine, which are commonly used in OCD. Details are below and on the site's Antidepressants page.

To check interactions with the medicines you take: CBD drug interaction checker.

How was CBD used in studies?

In the controlled study in hair pulling and skin picking, dronabinol (synthetic THC) was started at 5 mg a day, increased every two weeks up to 15 mg a day, and used for 10 weeks; in the uncontrolled study in hair pulling, 2.5–15 mg a day was used (average effective dose about 11.6 mg a day). In the nabilone study in OCD, 1 mg was given twice a day (2 mg a day) for 4 weeks; after some people dropped out, it was started at 0.25–0.5 mg twice a day and increased over about 1.5 weeks. In the laboratory study, participants smoked half of a cannabis cigarette in a single session (study doses; not a personal recommendation).

In the only study with pure CBD, the approved CBD medicine was given at 5 mg per kilogram of body weight a day in the first week, then 10 mg a day (milligrams per kilogram per day; for a person weighing 60 kg, that is 300 and 600 mg a day, respectively). In children with PANS, the CBDA-dominant extract was started at 5 mg per kilogram a day and increased to 20 mg over 4 weeks (study doses; not a personal recommendation). At none of these doses has a benefit for OCD or hair pulling been shown in a controlled study.

Bottom line

CBD oil has not yet been studied enough in OCD and hair pulling; the only data come from an uncontrolled study that 4 people started. In two small controlled studies with THC-containing products, no clear benefit was seen compared with a dummy treatment (very weak evidence), and use disorder is common among people with OCD who use cannabis. OCD has well-established treatments, and behavioral therapies are used for hair pulling too; talking with a specialist is a good place to start.

Frequently asked questions

References