What does the research say?
A total of 44 people with gastroparesis (stomach emptying measured as delayed) were randomly split into two groups and took either pharmaceutical-grade CBD or an identical-looking inactive solution (placebo) for 4 weeks; most participants were women, and in 12 of them gastroparesis was caused by diabetes. People taking CBD had a clearly lower total symptom score than the placebo group; inability to finish a normal meal, the number of vomiting episodes a day and overall symptom severity were also better, and they could drink more in a nutrient drink test. On the other hand, CBD slowed the emptying of solid food from the stomach slightly more than placebo (the difference in the study's main emptying measure was not significant), and diarrhea was much more common among people taking CBD (12 of 21; 2 of 23 on placebo). The study was done at a single center and was small and short; at the start, symptoms in the CBD group were numerically somewhat more severe (the analysis was adjusted for this). The total symptom score, presented in the publication as a main result, is not among the measures in the study's ClinicalTrials.gov registration; the main measures in the registration are stomach emptying, stomach volume and the nutrient drink test. In the unadjusted week-4 values in the registration, overall symptom severity was similar in the two groups, and daily vomiting was lower in the CBD group. There is not yet a second study repeating this result.
In a second study run at the same center, in the same period and under the same registration number, 48 people with functional dyspepsia whose stomach emptied normally took the same dose of CBD or placebo. After four weeks, there was no clear difference between CBD and placebo in upper abdominal pain, nausea, bloating or quality of life; stomach emptying, stomach volume and the satiation test did not differ either. In the weekly assessment, those reporting adequate relief were 12 of 25 people in the CBD group and 11 of 23 in the placebo group (results in the registry). The authors discussed, as an exploratory finding, a possible benefit for after-meal complaints in people carrying a particular gene variant; this finding was borderline.
The only controlled study of CBD in irritable bowel syndrome comes from the Netherlands: in this crossover study (each participant tried both products in turn), which started with 40 people and in which 32 women were assessed to the end, participants chewed either a gum containing 50 mg CBD or the same gum without CBD when their pain started, for 3 weeks each (at most 6 a day). Both gums reduced pain a little, but there was no clear difference between the CBD gum and placebo; quality of life did not differ either. Participants used the gum much less than expected (less than 1 a day). The manufacturer supplied the gums and covered some of the study's costs; according to the authors, the company was not involved in the design or writing. The publication does not state the purity of the CBD in the gum or whether it contained THC. A 2022 meta-analysis of cannabinoids in chronic primary pain, based on this single study, wrote that CBD may not reduce pain in irritable bowel syndrome and rated the evidence as low certainty.
The three controlled studies of THC in irritable bowel syndrome did not treat symptoms; they measured the movement and sensitivity of the bowel in the laboratory. In 75 patients, a single dose of synthetic THC (dronabinol) reduced contractions of the colon on an empty stomach; this effect was more marked in the diarrhea-predominant or alternating diarrhea-constipation type, and sensitivity to pain and gas did not change. In 36 patients with the diarrhea type, dronabinol taken for 2 days did not change transit time through the stomach, small intestine and colon differently from placebo. In 10 patients and 12 healthy volunteers, THC did not change sensitivity to stretching of the rectum with a balloon; at the highest dose, all participants reported effects such as light-headedness and sleepiness. A 2022 meta-analysis pooling the two single-dose studies found the effect to be zero and rated the evidence as very low certainty; the 2017 report of the US National Academies also wrote that the evidence was insufficient to say whether dronabinol works for the symptoms of this syndrome.
Controlled studies in healthy volunteers show that THC can slow the digestive tract. In a 13-person study, THC slowed the emptying of solid food from the stomach in every participant: 2 hours after the meal, the amount left in the stomach was 46% on the placebo day and 74% on the THC day. In a 30-person study, dronabinol delayed stomach emptying, especially in women; in another 52-person study, a single dose of dronabinol relaxed the colon, reduced after-meal contractions and raised pain scores in response to stretching. Low-dose THC given into a vein, however, did not change the emptying of liquid from the stomach in 7 volunteers. Because the stomach already empties slowly in gastroparesis, these findings are an important safety note.
There is no controlled study of THC-containing products in gastroparesis. In 24 patients who had not responded to treatment, symptom scores fell after at least 60 days of dronabinol or medical marijuana (variable THC and CBD ratio); but this study had no comparison group and no blinding, and side effects were not reported. In a multicenter patient registry in the US, 12% of 506 people with gastroparesis symptoms used marijuana; most users said they benefited (user reports), but their nausea-vomiting and pain were also more severe than in non-users, and about one in ten reported feeling worse. Analyses of large records conflict: in 406 patients from the same consortium and in the electronic records of more than 41,000 patients, cannabis use was linked to more emergency visits and hospital admissions; in coded data on people admitted to hospital in the US, however, length of stay and in-hospital death were lower. These observational studies do not show cause and effect.
In irritable bowel syndrome, observational data are also limited and come only from US hospital records: a diagnosis of cannabis use disorder was linked to an approximately 40% higher chance of hospital admission for irritable bowel syndrome; among patients admitted for irritable bowel syndrome whose cannabis use was recorded, readmission for any reason within 30 days was lower, and there was no difference in readmission related to irritable bowel syndrome. In Italy, in 60 people with fibromyalgia, THC-containing medical marijuana used for 6 months reduced scores for stomach pain, heartburn and bloating; three quarters of the patients also had irritable bowel syndrome or functional dyspepsia, but this study had no comparison group.
Most guidelines do not recommend cannabinoids for treating these three conditions. The American Gastroenterological Association's 2025 gastroparesis guideline suggests that CBD not be used as a first-line treatment (conditional recommendation). The Gastroenterological Society of Australia's 2026 position statement writes that cannabinoids slow the stomach but can surprisingly improve symptoms, yet the evidence is insufficient to recommend them (low-quality evidence). Germany's functional dyspepsia guideline (2026) recommends that cannabinoids not be used routinely in this condition and reminds readers that recreational use can cause indigestion-like complaints and even persistent vomiting. The 2025 guideline of the European and North American pediatric gastroenterology societies also recommends that cannabis not be used for irritable bowel syndrome and functional abdominal pain in children aged 4–18. The Canadian Association of Gastroenterology's 2019 position statement is more flexible: it writes that short-term use may help with persistent nausea, vomiting, diarrhea or abdominal pain when standard treatments are not enough; but it adds that long-term effectiveness and safety in functional disorders such as irritable bowel syndrome and functional dyspepsia are unknown, and that cannabinoids should not be considered first-line treatment (3 of the 7 experts who wrote the statement have ties to companies that make cannabis or cannabinoid products).
Strength of evidence: ●●○○ Weak evidence. Finding: Mixed results. What does this mean?
Ongoing research on this topic
No ongoing CBD trial on this topic appears on ClinicalTrials.gov right now. All ongoing trials
Possible side effects and drug interactions
In the two studies in which high-dose pharmaceutical-grade CBD was used for 4 weeks, the most notable side effect was diarrhea: in gastroparesis, in 12 of 21 people taking CBD and 2 of 23 taking placebo; in functional dyspepsia, in 8 of 25 and 1 of 23 (results in the registry). Tiredness and nausea were similar in the two groups; in the gastroparesis study, no changes were seen in liver and kidney tests after 4 weeks, and the two serious events recorded (an injury and an appendectomy) were not attributed to the drug. In the CBD gum study, two people stopped because swallowing air while chewing the gum bothered them. In short studies with THC and dronabinol, mental fogginess, dizziness, sleepiness and an increased heart rate were seen.
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. As a precaution, people taking certain medicines that could be affected in this way were not enrolled in the CBD gum study.
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 |
|---|---|---|---|
| Amitriptyline (tricyclic) (Laroxyl) | Caution | Amitriptyline: In healthy volunteers, a single 30 mg dose of CBD oil raised the blood level (AUC) of amitriptyline by 13% and its peak level by 17%; the effect is said to possibly be larger with ongoing use. Tell your doctor or pharmacist. | PMID 41398552 |
| PPI stomach medicines (omeprazole, lansoprazole, etc.) | Information | Stomach protectors (PPIs): CBD can raise the blood level of omeprazole; no clinical importance has been reported. If you take it long term, tell your pharmacist. | PMID 37313955, EMA Epidyolex SmPC 4.5 |
To check interactions with the medicines you take: CBD drug interaction checker.
How was CBD used in studies?
In the gastroparesis and functional dyspepsia studies, an approved (licensed) pharmaceutical-grade CBD solution was taken by mouth, split into two equal doses a day: it was started at 2.5 mg per kilogram of body weight a day and raised every two days up to 20 mg per kilogram a day (study dose; not a personal recommendation), and continued for 4 weeks; in the gastroparesis study, a few patients could not reach the full dose. This dose is much higher than the daily amounts suggested for over-the-counter CBD products (mostly 10–70 mg). In the irritable bowel syndrome study, a gum containing 50 mg CBD per piece was chewed when pain occurred, at most 6 a day (study dose; not a personal recommendation); participants used an average of 18 pieces of gum over 3 weeks. In the THC studies, synthetic THC (dronabinol) was given as a single dose of 2.5–7.5 mg or as 2.5–5 mg twice a day for 2 days (study dose; not a personal recommendation). None of these studies lasted longer than 4 weeks; there is no controlled study of long-term use.
Study details (for health professionals)
In gastroparesis (44 patients with gastric emptying measured as delayed; 32 idiopathic, 12 diabetic), pharmaceutical-grade CBD used at up to 20 mg per kilogram a day for 4 weeks clearly lowered the total symptom score compared with placebo (p = 0.008); inability to finish a normal meal (p = 0.029), number of vomiting episodes a day (p = 0.006) and overall severity (p = 0.034) were also better, and in the nutrient drink test the volume to comfortable fullness and the maximum tolerated volume increased. In the same study, CBD slowed the emptying of solid food (time to 10% emptying 101 versus 77 minutes, p = 0.030; emptied at hour 4, 52% versus 63%, p = 0.045; for the main measure, half-emptying time, p = 0.088), and diarrhea was common (12/21 versus 2/23). The study was single-center and supported by the US National Institutes of Health (the authors declared no relevant conflicts of interest); at the start, the symptom score in the CBD group was numerically higher, and the result has not been repeated. In the publication, the total symptom score appears as one of the main outcomes, but it is not in the study's ClinicalTrials.gov registration (February 2024 version): the main measures in the registration are gastric emptying, gastric volume and the nutrient drink test, and among symptoms only overall severity and number of vomiting episodes are secondary measures. In the unadjusted week-4 medians in the registration, overall severity was similar in the two groups (1.8 versus 1.6), and daily vomiting was lower in the CBD group (0 versus 0.2). After an interim analysis, the target sample size was lowered from 48 to 44. In a study run at the same time under the same registration (NCT03941288) in 48 people with functional dyspepsia and normal gastric emptying, CBD did not change symptoms, gastric emptying, gastric volumes or the satiation test differently from placebo. In a crossover RCT of 50 mg CBD gum in irritable bowel syndrome (40 randomized, 32 women analyzed), the difference in pain reduction was 0.1 points (95% CI −0.3 to 0.5; p = 0.61), and there was no difference in quality of life either; the gums were used little, and the publication does not state the purity of the CBD in the gum. A meta-analysis of cannabinoids in chronic primary pain (2022), based on this study, wrote that CBD may not reduce pain in irritable bowel syndrome (low certainty). In irritable bowel syndrome, there are 3 short physiology RCTs with dronabinol and THC: a single dose of dronabinol reduced colon contractions on an empty stomach (75 patients; marked in the diarrhea and alternating types), 2 days of dronabinol did not change bowel transit (36 patients with diarrhea), and THC did not change rectal sensitivity (10 patients, 12 healthy volunteers); in a 2022 meta-analysis, the effect was zero (very low certainty), and the US National Academies (2017) found the evidence insufficient. In healthy volunteers, THC (13 people) and dronabinol (30 people) slowed the emptying of solid food from the stomach, dronabinol (52 people) relaxed the colon and raised pain scores in response to stretching; low-dose THC given into a vein did not change liquid emptying (7 people). In gastroparesis, there is no controlled study of THC-containing products: in an uncontrolled series of 24 patients, symptom scores fell with dronabinol or medical marijuana; in a multicenter patient registry, 12% of patients used marijuana and most users reported benefit (user reports). Observational studies conflict: cannabis use was linked to more emergency visits and admissions in a patient registry (406 people) and in a matched analysis of electronic records of 41,374 people, while in US hospital data length of stay and in-hospital death were lower; in irritable bowel syndrome, hospital records showed more frequent admission and less readmission with cannabis use disorder. These do not show cause and effect. In people with type 1 diabetes, all 11 patients diagnosed with gastroparesis by gastric emptying test while using cannabis every day met the criteria for cannabinoid hyperemesis syndrome. The American Gastroenterological Association (2025) conditionally suggests that CBD not be used as a first-line treatment for gastroparesis; the Gastroenterological Society of Australia (2026) found the evidence insufficient to recommend; Germany's functional dyspepsia guideline (2026) recommends that cannabinoids not be used routinely, and ESPGHAN/NASPGHAN (2025) recommends that cannabis not be used for irritable bowel syndrome and functional abdominal pain in children. The Canadian Association of Gastroenterology's 2019 position statement, on the other hand, writes that short-term use may help with persistent nausea, vomiting, diarrhea or abdominal pain when standard treatments are not enough, but that long-term effectiveness and safety in functional bowel disorders are unknown and cannabinoids should not be considered first-line treatment (3 of the 7 authors have ties to companies that make cannabis or cannabinoid products). October 2026 search (PubMed, Cochrane, ClinicalTrials.gov, Europe PMC): 6 controlled studies with cannabinoids were found in these three conditions: 3 with pharmaceutical-grade or pure CBD (one each in gastroparesis, functional dyspepsia and irritable bowel syndrome; the longest 4 weeks) and 3 short physiology studies with dronabinol or THC in irritable bowel syndrome (single dose or 2 days). There are also 4 controlled studies measuring the effect of THC on stomach and colon movement in healthy volunteers, 3 systematic reviews, 1 expert committee report, 5 guidelines or position statements, 2 uncontrolled studies and 7 observational studies; all are in the table below. Six of the controlled studies (the gastroparesis and dyspepsia studies and the four dronabinol studies in patients and healthy volunteers) come from the same research group (Mayo Clinic, US). There is no Cochrane review; all 3 studies with results posted on ClinicalTrials.gov have been published; in October 2026, no ongoing cannabinoid study was found in these conditions; there are no studies from Turkey. Not included in the table: an irritable bowel syndrome study with the selective synthetic CB2 drug olorinab and stomach studies with the CB1 blocker rimonabant (site rule: not a cannabis or CBD product), substances that are not cannabinoid products such as PEA and hemp husk fiber, other US hospital database analyses with similar results, a retracted nabilone series, studies of chronic pancreatitis or postoperative abdominal pain, a small dronabinol study in functional chest pain, a single-center survey asking about use and perceived benefit in gastroparesis, and a review that examines the effect of CBD on bowel movement mostly through animal studies.
Evidence assessment: Weak; differs by condition (in each condition, a single small RCT with pure or pharmaceutical-grade CBD, the longest 4 weeks: in gastroparesis, improvement in symptoms, but on a measure not defined in the registration and not repeated; in functional dyspepsia and irritable bowel syndrome, no difference from placebo)
CBD alone
In gastroparesis, pharmaceutical-grade CBD (up to 20 mg per kilogram a day, 4 weeks, 44 people) clearly lowered the total symptom score compared with placebo (p = 0.008; this score is not among the measures in the study's ClinicalTrials.gov registration) and reduced vomiting; but it slowed the emptying of solid food from the stomach slightly more, and diarrhea was common (12 of 21 people, 2 of 23 on placebo); the result is from a single center and has not been repeated. The same dose was no different from placebo in symptoms, stomach measurements or quality of life in functional dyspepsia with normal stomach emptying (48 people). In irritable bowel syndrome, gum containing 50 mg CBD (32 women, crossover design, 3 weeks each) was no different from placebo gum in pain or quality of life. The American Gastroenterological Association (2025) suggests that CBD not be used as a first-line treatment for gastroparesis (conditional recommendation).
Products containing THC
In irritable bowel syndrome, there are only short physiology studies with THC: a single dose of dronabinol reduced colon contractions on an empty stomach in 75 patients, dronabinol taken for 2 days did not change bowel transit time in 36 patients with diarrhea, and THC did not change bowel sensitivity in 10 patients; in a meta-analysis pooling the single-dose studies, the effect was zero (very low certainty). There is no controlled THC study following symptoms over weeks; no THC study was found in functional dyspepsia. In gastroparesis, there is only an uncontrolled series (24 patients, improvement in symptoms) and user reports in a patient registry; in healthy volunteers, THC and dronabinol slowed the emptying of solid food from the stomach. In a patient registry and in a large analysis of electronic records, cannabis users with gastroparesis had more frequent emergency visits and hospital admissions (in coded in-hospital data, however, length of stay was shorter; these do not show cause and effect). Long-term heavy cannabis use can cause recurring bouts of vomiting that can be mistaken for gastroparesis.
Evidence: Insufficient (no controlled THC study following symptoms in patients over weeks; in irritable bowel syndrome, only 3 single-dose or 2-day physiology RCTs, with negative or unclear results; in gastroparesis, only an uncontrolled series and user reports; in healthy volunteers, THC slowed stomach emptying)
Dose range in studies
Pharmaceutical-grade pure CBD: from 2.5 mg up to 20 mg per kilogram a day, two doses a day, 4 weeks (gastroparesis, functional dyspepsia); 50 mg CBD gum as needed, at most 6 a day (300 mg), 3 weeks (irritable bowel syndrome; use less than 1 a day); THC-containing: dronabinol single dose 2.5–7.5 mg or 2.5–5 mg twice a day for 2 days, THC single dose 5–10 mg and 10 mg/m² (physiology studies); in the uncontrolled gastroparesis series, dronabinol 2–10 mg 2–4 times a day and medical marijuana of variable ratio
Starting dose and titration
In the gastroparesis and functional dyspepsia studies, started at 2.5 mg/kg/day and raised by 2.5–5 mg/kg every two days up to 20 mg/kg/day; at the end of 4 weeks, tapered off over 10 days (study dose; not a personal recommendation)
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 |
|---|---|---|---|---|---|---|
| Zheng 2023 Phase 2, double-blind, placebo-controlled, parallel RCT (single center) | 44 randomized (21 CBD / 23 placebo); 42 completed Idiopathic (32) or diabetic (12; 6 type 1, 6 type 2) gastroparesis with gastric emptying delayed on scintigraphy; 89% women (US, Mayo Clinic) | Approved (licensed), pharmaceutical-grade pure CBD oral solution (Epidiolex, 100 mg/mL) | Target 20 mg/kg/day, two equal doses a day; 4 patients (5 in the abstract) could not reach the full dose and continued at the highest dose they tolerated; according to the publication, the average dose was 9.43 mg/kg (whether daily or per dose is not clear) Started at 2.5 mg/kg/day and raised by 2.5–5 mg/kg every two days; at the end of 4 weeks, tapered off over 10 days | 4 weeks | On the primary symptom outcome (GCSI daily symptom diary total score), CBD was better than placebo (p = 0.008); inability to finish a normal meal (p = 0.029), number of vomiting episodes a day (p = 0.006) and overall symptom severity (p = 0.034) were also better, and upper abdominal pain was borderline (p = 0.076); effect size for the total score 0.46. In the nutrient drink test, comfortable fullness (817 versus 665 kcal; p = 0.028) and maximum volume drunk (1115 versus 889 kcal; p = 0.018) increased. CBD slowed the emptying of solid food: time to 10% emptying 101 versus 77 min (p = 0.030), emptied at hour 2, 19% versus 28% (p = 0.032), at hour 4, 52% versus 63% (p = 0.045); on the main emptying measure, half-emptying time, the difference was not significant (268 versus 227 min; p = 0.088); gastric volumes did not change. At the start, the total symptom score was numerically higher in the CBD group (median 2.0 versus 1.4; p = 0.37); the analysis was adjusted for baseline. In the publication, the GCSI total score is one of the primary outcomes; in the ClinicalTrials.gov registration (February 2024 version), the primary measures are only gastric measurements (half-emptying and lag time, gastric volumes, nutrient drink test), and the GCSI total score is not among the registered measures (overall severity and vomiting are secondary). Unadjusted week-4 medians in the registration: overall severity 1.8 versus 1.6, daily vomiting 0 versus 0.2. After an interim analysis following 24 patients, the target sample was lowered from 48 to 44. Single-center, short, not repeated Side effects: Diarrhea 12/21 with CBD, 2/23 with placebo (p < 0.001); tiredness 4 versus 4, nausea 3 versus 4; no change in liver enzymes, creatinine or HbA1c over 4 weeks; 2 serious events in the CBD group (abdominal injury in a domestic fight, appendectomy), not attributed to the drug | PMID 37482172, doi:10.1016/j.cgh.2023.07.008, NCT03941288, results in the registry, February 13, 2024), PMC10800684. NIH support (R01-DK122280), the authors declared no relevant conflicts of interest |
| Atieh 2022 Phase 2, double-blind, placebo-controlled, parallel RCT (single center; run at the same time as the gastroparesis study, same registration) | 48 randomized (25 CBD / 23 placebo); 43 completed (in the registry) Functional dyspepsia of the postprandial distress type (Rome IV) with normal gastric emptying (US, Mayo Clinic) | Pharmaceutical-grade pure CBD oral solution (according to the registration, the same product and schedule as the gastroparesis study) | Target 20 mg/kg/day, two doses a day (following the FDA dose-escalation guidance) According to the registration, started at 2.5 mg/kg/day and raised by 2.5–5 mg/kg every two days | 4 weeks | No clear difference between CBD and placebo in symptoms (daily pain, nausea, bloating), weekly adequate relief, the Leuven Postprandial Distress Scale or quality of life; gastric emptying, gastric volumes and the nutrient drink satiation test did not differ either. According to the results in the registry, adequate relief was reported by 12/25 versus 11/23; median daily symptom score 1.2 versus 1.1; median volume drunk to comfortable fullness 373 versus 622 mL (no statistical test in the registry). In carriers of a CNR1 gene variant, a borderline treatment-gene interaction for after-meal complaints (p = 0.06; exploratory) Side effects: According to the registration, diarrhea 8/25 with CBD, 1/23 with placebo; tiredness 8 versus 8, nausea 8 versus 8; no serious side effects | PMID 35537858, doi:10.14309/ajg.0000000000001805, NCT03941288, February 13, 2024), full text could not be read |
| van Orten-Luiten 2022 Double-blind, placebo-controlled, crossover RCT (exploratory; use as needed for pain) | 40 randomized; 32 analyzed (all women) Irritable bowel syndrome by Rome III/IV criteria, with at least 3 moderate-to-severe pain attacks a week (Netherlands) | Gum containing 50 mg CBD (CBD in purified food-grade hemp seed oil; the placebo gum contained only the seed oil; the purity of the CBD and THC content not stated in the publication); manufacturer Axim Biotechnologies | 1 piece of gum when pain started (chewed for at least 30 min), at most 6 a day (300 mg CBD); on average 18.2 CBD gums over 3 weeks (range 1–71) None (use according to symptoms) | 2 × 3 weeks (1-week washout in between; 8 weeks in total) | No difference between CBD and placebo on the primary outcome (pain reduction during a 30-minute pain attack): difference 0.1 points (0–10 scale; 95% CI −0.3 to 0.5; p = 0.61); both gums reduced pain a little. No difference in quality of life (−1.0; 95% CI −6.8 to 4.9; p = 0.74). The gum was used less than expected (on average 5.9 a week). The manufacturer supplied the gums and covered logistics costs; according to the authors, it was not involved in the design or writing; one of the authors sat on the company's scientific board until 2018 (unpaid) Side effects: 2 people stopped because swallowing air while chewing bothered them; otherwise well tolerated | PMID 33998882, doi:10.1089/can.2020.0087, NCT03003260, no results in the registry, the registration also listed daily adequate relief and stool frequency among its measures, but the publication gives no results for them), PMC9418368 |
| Wong 2012 Double-blind, placebo-controlled, parallel RCT (pharmacodynamic study measuring bowel transit) | 36 randomized (13 placebo / 10 dronabinol 2.5 mg / 13 dronabinol 5 mg); 1 person stopped because of side effects Diarrhea-predominant irritable bowel syndrome (Rome III); 34 women, cannabinoid-naive (US, Mayo Clinic) | Dronabinol (synthetic THC), by mouth | 2.5 mg or 5 mg twice a day None | 2 days | No overall effect of dronabinol on the primary outcome (colonic transit at 24 hours by scintigraphy) or on gastric and small bowel transit. In carriers of a variant of the CB1 receptor gene (CNR1 rs806378 CT/TT), a trend toward slightly delayed colonic transit (p = 0.13; exploratory). Symptoms were not measured Side effects: Side effects did not differ between groups; according to the registration, dizziness 4/13 at 5 mg, 1/13 on placebo; 1 person in the 5 mg group stopped because of side effects | PMID 22288893, doi:10.1111/j.1365-2982.2011.01874.x, NCT01253408, PMC3775711 |
| Wong 2011 Double-blind, placebo-controlled, parallel RCT (single-dose pharmacodynamic and pharmacogenetic study) | 75 randomized (27 placebo / 24 dronabinol 2.5 mg / 24 dronabinol 5 mg); 72 completed Irritable bowel syndrome: 35 constipation-predominant, 35 diarrhea-predominant, 5 alternating; more than 90% women, cannabinoid-naive (US, Mayo Clinic) | Dronabinol (synthetic THC), by mouth | Single dose of 2.5 mg or 5 mg None | Single dose | On the primary measure in the registration opened later (colonic wall compliance), dronabinol showed only a borderline increase (p = 0.058). It reduced the fasting motility index in the upper part of the left colon compared with placebo (overall p = 0.05; p = 0.046 for 5 mg); both effects were more marked in the diarrhea and alternating types (contractions p = 0.022; compliance p = 0.03). Sensitivity to pain and gas, colonic tone and the after-meal response did not change. Symptoms were not measured; single dose Side effects: Most common were sleepiness (23%), hot flushes (19%), headache (13%) and dizziness (11%); only mental fogginess was more common with dronabinol (p = 0.009; according to the registration, 6/24 at 5 mg, 0/27 on placebo) | PMID 21803011, doi:10.1053/j.gastro.2011.07.036, NCT01786109, after publication, the 75-person single-dose results in the registration belong to this study, NCT01253408, but the results of that registration belong to Wong 2012), PMC3202649 |
| Klooker 2011 Double-blind, placebo-controlled, randomized crossover study (rectal barostat test) | 22 (10 patients with irritable bowel syndrome, 12 healthy volunteers) Irritable bowel syndrome and healthy volunteers (Netherlands) | Δ9-THC (dronabinol); route of administration not in the abstract | 5 and 10 mg in healthy volunteers, 10 mg in patients None | Single dose (3 separate days in healthy volunteers, 2 in patients) | THC did not change baseline sensitivity to balloon distension of the rectum or the discomfort threshold after sigmoid colon stimulation differently from placebo (in healthy volunteers and in patients). According to the authors, the findings argue against using centrally acting cannabinoids to reduce visceral hypersensitivity Side effects: At the highest dose, central side effects in all participants (heightened awareness of surroundings, light-headedness, sleepiness); none with placebo. Heart rate increased, most in patients; blood pressure did not change | PMID 20718944, doi:10.1111/j.1365-2982.2010.01587.x, Full text could not be read (abstract) |
| Esfandyari 2007 Double-blind, placebo-controlled, parallel RCT (colonic barostat-manometry) | 52 Healthy volunteers (US, Mayo Clinic) | Dronabinol (synthetic THC), by mouth | Single dose of 7.5 mg None | Single dose | Dronabinol relaxed the colon: wall compliance increased (p = 0.045), borderline relaxation of fasting tone (p = 0.096), and after-meal tone (p = 0.048) and fasting and fed contraction pressures (p = 0.008 and p = 0.030) decreased. First gas and pain thresholds did not change, but pain scores during random distensions increased (p = 0.024). Physiology study in healthy volunteers Side effects: Not reported (in the abstract) | PMID 17395895, doi:10.1152/ajpgi.00565.2006, Full text could not be read (abstract) |
| Esfandyari 2006 Double-blind, placebo-controlled, parallel RCT | 30 Healthy volunteers (US, Mayo Clinic) | Dronabinol (synthetic THC), by mouth | 5 mg twice a day, 3 doses in total (according to the abstract) None | 3 doses (twice a day) | Dronabinol delayed gastric emptying overall (p = 0.018); the effect was marked in women (p = 0.011) and not significant in men (p = 0.184). Gastric volumes, maximum tolerated nutrient drink volume, post-drink symptoms and small and large bowel transit did not change; in men, fasting gastric volume was larger with dronabinol (238 versus 185; p = 0.04). Physiology study in healthy volunteers Side effects: Not reported (in the abstract) | PMID 16918762, doi:10.1111/j.1365-2982.2006.00834.x, Full text could not be read (abstract), the full text of Wong 2012 reports the dose in this study as 7.5 mg twice a day, while the abstract says 5 mg |
| McCallum 1999 Double-blind, randomized, placebo-controlled crossover study | 13 Healthy volunteers (US) | Δ9-THC; route of administration not in the abstract | 10 mg/m² body surface area (the dose used for chemotherapy nausea) None | Single dose (2 separate days) | THC slowed the emptying of solid food from the stomach compared with placebo in every participant: the proportion remaining in the stomach was higher from minute 30 onward (94.2% versus 85.5%; p < 0.05), and at minute 120 it was 73.9% versus 45.6% (p < 0.001). The delay was not related to blood THC levels Side effects: Not reported (in the abstract) | PMID 9892882, doi:10.1046/j.1365-2036.1999.00441.x, Full text could not be read (abstract) |
| Bateman 1983 Placebo-controlled study (randomization not stated in the abstract) | 7 Healthy volunteers | Δ9-THC, intravenous | 0.5 and 1 mg None | Single dose | Despite clear changes in pulse and psychological measures consistent with a cannabis effect, no clear change versus placebo in the emptying of liquid from the stomach Side effects: Increased pulse and psychological changes (abstract) | PMID 6307330, doi:10.1111/j.1365-2125.1983.tb01561.x, PMC1427915 (scanned page images only, text could not be read) |
| Bilbao 2022 Systematic review and meta-analysis of RCTs, separated by drug type (up to October 2021; GRADE) | 152 RCTs (all indications, 12,123 participants); in irritable bowel syndrome, 2 RCTs with dronabinol (Wong 2011, Klooker 2011) All medical indications (including an irritable bowel syndrome subgroup) | Dronabinol, nabilone, pure CBD and nabiximols (natural products such as smoked cannabis excluded) | Variable (single-dose dronabinol in irritable bowel syndrome) | In irritable bowel syndrome, the results of the two single-dose dronabinol studies pointed in opposite directions; the pooled effect was zero (SMD 0), with very low certainty of evidence. Wong 2012 and the CBD gum study were not included in this analysis; the gastroparesis and functional dyspepsia RCTs were published after the search date Side effects: Across all indications, the chance of any side effect with CBD was higher than with placebo (OR 1.82; 95% CI 1.08–3.07; 22 RCTs); no side effect analysis specific to irritable bowel syndrome | PMID 35982439, doi:10.1186/s12916-022-02459-1, PROSPERO CRD42021229932, PMC9389720 | |
| Giossi 2022 Systematic review and meta-analysis (RCTs and observational studies in chronic primary pain searched; PubMed, EMBASE, Cochrane, up to October 31, 2021; GRADE) | 8 RCTs, 240 people (all chronic primary pain); in irritable bowel syndrome, 2 RCTs, 68 people (Wong 2012 and the CBD gum study) Chronic primary pain: fibromyalgia, irritable bowel syndrome, complex regional pain syndrome type I, functional chest pain | Cannabinoids (nabilone, dronabinol, THC-rich oil, vaporized cannabis, CBD gum) | Variable | In irritable bowel syndrome, pain could be analyzed only from the CBD gum study: difference versus placebo 0.34 points (on a 0–10 scale; 95% CI −1.06 to 1.73; 1 RCT, 32 people; low certainty); difference in quality of life −1.0 (−6.8 to 4.9; 1 RCT, 30 people; low certainty). According to the authors, CBD may not reduce pain in irritable bowel syndrome. Across all chronic primary pain, pain reduction with cannabinoids was not significantly different from placebo (−0.64; −1.30 to 0.02) Side effects: No significant difference from placebo in stopping because of side effects (OR 2.15; 0.44–10.65; all chronic primary pain); no side effect analysis specific to irritable bowel syndrome | PMID 36129666, doi:10.1007/s40122-022-00434-5, PMC9633894, also a row in the fibromyalgia topic. One of the authors reported receiving conference support from Jazz Pharmaceuticals (the maker of Epidiolex) | |
| Volz 2016 Systematic review (up to March 2015; CENTRAL, PubMed, cannabis-med.org, ClinicalTrials.gov; at least 4 weeks and at least 10 people per arm; Cochrane risk-of-bias tool) | 1 RCT in IBD (21 people); no RCT meeting the criteria in irritable bowel syndrome Gastroenterology: inflammatory bowel disease, irritable bowel syndrome, chronic pancreatitis | Herbal cannabis and pharmaceutical cannabinoids | Not applicable | No RCT lasting at least 4 weeks and enrolling at least 10 people per arm was found in irritable bowel syndrome. The only suitable RCT was in Crohn's disease (21 people); the authors wrote that methodologically sound, sufficiently large and long studies are needed Side effects: Not applicable | PMID 26809974, doi:10.1007/s00482-015-0087-0, In German, full text could not be read (English abstract) | |
| NASEM 2017 (US National Academies report) Expert committee report (assessment based on good- and fair-quality systematic reviews and subsequent primary studies) | No suitable systematic review in irritable bowel syndrome; 1 RCT (Wong 2012, 36 patients) Irritable bowel syndrome | Cannabinoids (only dronabinol in irritable bowel syndrome) | The committee found no good- or fair-quality systematic review on medical cannabis in irritable bowel syndrome; in the single suitable study (Wong 2012), 2 days of dronabinol did not change bowel transit; the evidence is insufficient because of the short duration, small sample and lack of patient-reported outcomes. Conclusion 4-5: there is insufficient evidence to say whether dronabinol is effective for the symptoms of irritable bowel syndrome | PMID 28182367, doi:10.17226/24625, National Academies Press, chapter 4 (nap.nationalacademies.org/read/24625/chapter/6, read on October 28, 2026), the NCBI Bookshelf copy could not be opened because of bot verification | ||
| AGA 2025 (Staller) Clinical guideline (American Gastroenterological Association; GRADE) | Idiopathic and diabetic gastroparesis | CBD | One of 12 recommendations: a conditional recommendation against using CBD (together with domperidone, prucalopride, aprepitant, nortriptyline and buspirone) as first-line treatment. A conditional recommendation for metoclopramide or erythromycin as first-line. The certainty of evidence and the rationale are not in the abstract | PMID 40976635, doi:10.1053/j.gastro.2025.08.004, Full text could not be read (the publisher page redirects with a script, text-mining rights reserved) | ||
| GESA 2026 (Kellar) Professional society position statement (Gastroenterological Society of Australia; literature review and modified Delphi consensus; 20 statements, with level of evidence and strength of recommendation) | Idiopathic gastroparesis | Cannabinoids | Statement 17: cannabinoids slow gastric emptying but can paradoxically improve gastroparesis symptoms, including fullness; the evidence is insufficient to recommend their use (low-quality evidence; conditional recommendation; 100% consensus). Basis: the single 44-patient placebo-controlled CBD study (Zheng 2023) and large epidemiological studies showing greater healthcare use among cannabinoid users | PMID 42235947, doi:10.1111/jgh.70456, PMC13326989 | ||
| DGNM 2026 (Storr) Clinical guideline, S1 level (expert consensus; German Society of Neurogastroenterology and Motility), English edition | Functional dyspepsia | Cannabinoids | Pain relievers such as nonsteroidal anti-inflammatory drugs, acetaminophen (paracetamol), cannabinoids and opioids should not be used routinely to treat functional dyspepsia. It notes that cannabinoids used recreationally can cause dyspepsia-like complaints with nausea and, in more severe cases, cannabinoid hyperemesis syndrome or cyclic vomiting | PMID 42095009, doi:10.1155/grp/2610765, PMC13140302, original German text: Storr 2025, PMID 40199346 | ||
| ESPGHAN/NASPGHAN 2025 (Groen) Clinical guideline (European and North American pediatric gastroenterology, hepatology and nutrition societies; GRADE; 86 RCTs reviewed) | Irritable bowel syndrome and functional abdominal pain in children aged 4–18 | Cannabis and cannabinoids | The guideline group recommends that cannabis not be used to treat functional disorders with abdominal pain in children: there is no high-quality evidence supporting safety and effectiveness in children, the developing nervous system may be harmed, and the quality and content of products are inconsistent | PMID 40444524, doi:10.1002/jpn3.70070, PMC12314588 | ||
| CAG 2019 (Andrews) Professional society position statement (Canadian Association of Gastroenterology; non-comprehensive literature review and expert opinion; no formal grading of evidence in the statements) | Digestive and liver diseases (including functional disorders such as irritable bowel syndrome and functional dyspepsia) | Cannabis and cannabinoids | Statement on digestive symptoms: short-term use in particular may help with persistent nausea, vomiting, diarrhea or abdominal pain when standard treatments have not worked. Comment: effectiveness in other chronic pain cannot be transferred to functional disorders such as irritable bowel syndrome or functional dyspepsia; long-term effectiveness and safety in these disorders are unknown; cannabinoids should not be considered first-line treatment for non-chemotherapy nausea-vomiting or other off-label digestive symptoms. They should not be used in place of approved treatment; in cannabinoid hyperemesis syndrome, the main treatment is stopping cannabis | PMID 31294362, doi:10.1093/jcag/gwy064, PMC6507278. According to the declaration of interests, 3 of the 7 authors have ties to companies that make cannabis or cannabinoid products (advisory board and management roles, research support and product donations, speaker fees) | ||
| Barbash 2019 Prospective, uncontrolled before-and-after study (single center) | 24 Gastroparesis not responding to standard treatment (idiopathic 11, diabetic 8, other 5; 20 women; US, New York) | Dronabinol (6 patients), medical marijuana (10; vaporized or sublingual drops, variable THC:CBD ratio) or dronabinol first and then marijuana (8) | Dronabinol 2–10 mg 2–4 times a day; marijuana as needed, ratio decided by the dispensary Not reported | At least 60 days of use | The GCSI total symptom score fell by 12.8 points after treatment (95% CI 10.4–15.2; p < 0.001); the nausea-vomiting, fullness and bloating subscores and the abdominal pain score also fell; marijuana was linked to more improvement than dronabinol. No control group, blinding or standard product; treatment durations varied; the effect of expectation cannot be separated Side effects: Not reported | PMID 31993268, doi:10.7759/cureus.6430, PMC6970440 |
| Santonicola 2025 Prospective observational study; no control group | 60 Fibromyalgia; 46 (77%) with a disorder of gut-brain interaction according to Rome IV (10 irritable bowel syndrome, 17 functional dyspepsia, 19 both) (Italy) | Medical marijuana (Bedrocan; contains THC; route of use and dose not in the abstract) | Not reported (in the abstract) Not reported (in the abstract) | 6 months | Over 6 months, the fibromyalgia severity score fell (p < 0.001); scores for stomach pain, heartburn, abdominal pain, abdominal tension and bloating decreased (p < 0.01). No control group, small sample; the authors also stress these limitations Side effects: Not reported (in the abstract) | PMID 40556630, doi:10.55563/clinexprheumatol/o5ck22, Full text could not be read (abstract) |
| Parkman 2025 Multicenter patient registry, cross-sectional analysis (NIDDK Gastroparesis Clinical Research Consortium, GpR3) | 406 Patients with gastroparesis symptoms (gastric emptying delayed in 72%; US, 7 centers) | Cannabis products (for any purpose; patient-reported) | Not reported | 1 year before registry enrollment | 24% of patients (96 people) used cannabis products. In multivariable analysis, cannabis use was independently linked to emergency visits (OR 2.31; 95% CI 1.33–4.03) and hospital admission (OR 2.01; 95% CI 1.03–3.94) in the past year. Cross-sectional; people with more severe symptoms using more cannabis could explain this link; does not show cause and effect Side effects: Not applicable | PMID 40286861, doi:10.1016/j.cgh.2025.01.033, NCT01696747, PMC12353374 |
| Kilani 2026 Retrospective cohort, 1:1 propensity score matched (TriNetX electronic health records) | 41,374 (20,687 cannabis users, 20,687 non-users) Adults with a diagnosis of gastroparesis (US) | Cannabis (use according to records; no product information) | Not reported | Not reported (in the abstract) | After matching, cannabis users had more frequent emergency visits (adjusted OR 1.73; 95% CI 1.66–1.80) and hospital admissions (OR 1.44; 1.39–1.50), and slightly less frequent upper endoscopy (OR 0.93; 0.88–0.98). Users were younger; diabetes, mood and anxiety disorders and opioid use were more common. Observational; does not show cause and effect Side effects: Not applicable | PMID 40197432, doi:10.14309/ajg.0000000000003479, Full text could not be read (abstract) |
| McCarty 2022 Retrospective cross-sectional analysis (US National Inpatient Sample, 2008–2014) | 1,473,363 admissions with a diagnosis of gastroparesis; cannabis use disorder in 33,085 (2.25%) Hospitalized patients with gastroparesis (US) | Cannabis (cannabis use disorder diagnosis codes) | Not reported | Length of hospital stay | After adjusting for age, sex, coexisting illnesses and socioeconomic factors, those with cannabis use disorder had less in-hospital death (OR 0.45; 95% CI 0.33–0.60) and fewer long stays (OR 0.36; 0.33–0.39), and were more often discharged home (OR 1.26; 1.22–1.30) (Table 5; the numbers in the text differ from the table). Cannabis users were younger and mostly male; mental illnesses were more common. In-hospital data only, based on diagnosis codes; does not show cause and effect Side effects: Not applicable | PMID 33780213, doi:10.1097/MCG.0000000000001526, PMC8435035 |
| Akturk 2025 Retrospective chart review, case-control comparison (single center) | 813 charts reviewed; 86 with cannabinoid hyperemesis syndrome, 94 with diabetic gastroparesis Adults with type 1 diabetes (US, Colorado) | Cannabis (use confirmed by urine test or chart notes) | Not reported | Records from January 2019–January 2024 | People with cannabinoid hyperemesis syndrome were younger than those with diabetic gastroparesis (29 versus 54 years), had a shorter duration of diabetes (16 versus 41 years) and a higher HbA1c (9.2% versus 7.4%); the proportions of women and of neuropathy were lower. All 11 patients diagnosed with gastroparesis by gastric emptying test while using cannabis every day met the criteria for the hyperemesis syndrome. A model built with age, duration of diabetes, sex and HbA1c separated the two conditions well (area under the curve 0.98). According to the authors, the gastric emptying test cannot tell the two conditions apart Side effects: Not applicable | PMID 40741458, doi:10.2337/cd24-0096, PMC12304556 (read from the PMC page), short paper by the same group: Karakus 2025, PMID 39984123, no abstract, not read) |
| Parkman 2020 Multicenter patient registry, cross-sectional analysis (NIDDK Gastroparesis Clinical Research Consortium, GpR2) | 506 Patients with gastroparesis symptoms (gastric emptying delayed in 72%; 86% women; US, 7 centers) | Marijuana (patient-reported use) | Not reported | Registry enrollment visit | 11.7% of patients (59 people) used marijuana; the rate was similar in those with delayed and normal gastric emptying and in idiopathic and diabetic patients. Users had higher nausea-vomiting (2.7 versus 2.1; p = 0.002) and upper abdominal pain scores (3.5 versus 2.9; p = 0.003). 81% of users reported that marijuana improved their symptoms and 10% (6 people) that it worsened them; 81% of ondansetron users also reported similar benefit. User reports; no control Side effects: Not applicable | PMID 31758430, doi:10.1007/s10620-019-05963-2, NCT01696747, PMC7242137 |
| Patel 2020 Retrospective cross-sectional analysis (US National Inpatient Sample, 2010–2014) | 31,272 irritable bowel syndrome admissions (ages 15–54; out of 6.8 million admissions in total) Hospitalized patients with irritable bowel syndrome (US) | Cannabis (cannabis use disorder diagnosis codes) | Not reported | Hospital stay | Cannabis use disorder was present in 3.7% of irritable bowel syndrome admissions; after adjusting for demographics, mental and medical illnesses and other substance use, cannabis use disorder was linked to admission for irritable bowel syndrome (OR 1.41; 95% CI 1.32–1.50). Irritable bowel syndrome admissions with cannabis use disorder rose by 32.8% between 2010 and 2014. Based on diagnosis codes; does not show cause and effect Side effects: Not applicable | PMID 31573379, doi:10.1080/10826084.2019.1664591, Full text could not be read (abstract) |
| Choi 2022 Retrospective cohort (US Nationwide Readmissions Database, 2016) | 7,163 irritable bowel syndrome admissions; cannabis use recorded in 357 Inpatients with a principal diagnosis of irritable bowel syndrome (US) | Cannabis (use according to diagnosis codes) | Not reported | 30 days | No difference in the primary outcome (readmission related to irritable bowel syndrome within 30 days): 1.5% versus 1.1% (p = 0.53). Readmission for any reason was lower in cannabis users (8.1% versus 12.7%; adjusted OR 0.53; 95% CI 0.28–0.99). The most common diagnoses in cannabis users' readmissions were cyclic vomiting and diarrhea-predominant irritable bowel syndrome. Observational; does not show cause and effect Side effects: Not applicable | PMID 33471483, doi:10.1097/MCG.0000000000001498, Full text could not be read (abstract) |
Reported side effects
- Diarrhea with high-dose pharmaceutical-grade CBD: in gastroparesis, 12/21 versus 2/23 (Zheng 2023); in functional dyspepsia, 8/25 versus 1/23 (NCT03941288 registry results); tiredness and nausea similar between groups
- With dronabinol, mental fogginess (at 5 mg, 6 of 24 people; on placebo, none of 27) and dizziness (Wong 2011; NCT01786109); with 10 mg THC, central effects such as light-headedness and sleepiness in all participants, and increased heart rate (Klooker 2011)
- Cannabinoid hyperemesis syndrome: recurring bouts of nausea and vomiting with long-term regular cannabis use; it goes away when cannabis is stopped (Sorensen 2017, systematic review, PMID 28000146)
Monitoring done in studies
- Cannabis and cannabinoid use should be reported to the physician before a gastric emptying test; the American College of Gastroenterology's 2022 gastroparesis guideline writes that, among medicines that may affect the result, stopping cannabinoids 48 hours before the test is usual practice (Camilleri 2022, PMID 35926490)
- In the gastroparesis study in which high-dose pharmaceutical-grade CBD was tested, liver tests were measured at the start and at week 4, and no change was seen; with longer use, monitoring liver tests is the physician's decision (Zheng 2023)
Who should be careful
- THC slows stomach emptying: in healthy volunteers, THC and dronabinol delayed the emptying of solid food from the stomach (McCallum 1999; Esfandyari 2006). In people with gastroparesis, whose stomach already empties slowly, THC-containing products may make symptoms worse; in the gastroparesis study, CBD also slowed emptying slightly (Zheng 2023)
- Cannabinoid hyperemesis syndrome: people who use cannabis for a long time, usually every day or every week, can have recurring bouts of severe nausea, vomiting and abdominal pain; relief from a hot shower or bath is typical, and the complaints go away when cannabis is stopped (Sorensen 2017, systematic review, PMID 28000146). This picture can be mistaken for gastroparesis: in a series cited by the review, emptying was delayed in 30% of 61 patients who had a gastric emptying test; in a review of people with type 1 diabetes, all 11 patients diagnosed with gastroparesis while using cannabis every day met the criteria for this syndrome (Akturk 2025)
- High doses cannot be compared with market doses: the gastroparesis study used pharmaceutical-grade CBD at up to 20 mg per kilogram a day; in low-dose (≤175 mg/day) studies, blood CBD levels were found to be 5–100 times lower than at this dose (Warren 2026, PMID 42473010)
- Diarrhea: diarrhea was the most common side effect of high-dose CBD in these studies; in diarrhea-predominant irritable bowel syndrome, this side effect could make complaints worse (this is a precautionary note; high-dose CBD has not been tested in irritable bowel syndrome)
- Children: in their 2025 guideline, the European and North American pediatric gastroenterology societies recommend that cannabis not be used for irritable bowel syndrome and functional abdominal pain in children aged 4–18 (Groen 2025)
Drug interactions
- The studies on this page did not systematically assess drug interactions; details are in the site's drug interactions section
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 and dietary advice you use for irritable bowel syndrome, indigestion or gastroparesis on your own; there is no evidence that CBD can replace them.
- If you have gastroparesis or frequent nausea and vomiting, talk to your doctor about THC-containing products; THC can slow stomach emptying.
- If you are going to have a gastric emptying test, say in advance that you use cannabis or CBD products; this can affect the result.
- If you take medicines regularly, talk to your doctor or pharmacist about drug interactions; high-dose CBD can make diarrhea worse.
- Check the product's label and certificate of analysis (COA); find out whether it contains THC.
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 these three digestive problems is still at an early stage, and the results vary by condition: in gastroparesis, a single small study is promising; in functional dyspepsia and irritable bowel syndrome, one small study each found no difference. Because THC-containing products can slow the stomach and, with long-term heavy use, can cause bouts of vomiting, they call for caution, especially in gastroparesis. It is best to make treatment decisions together with your gastroenterologist.
Frequently asked questions
Does CBD oil work for irritable bowel syndrome?
There is no evidence that it does. In the only controlled study of CBD in irritable bowel syndrome, 32 women chewed either a gum containing 50 mg CBD or the same gum without CBD when their pain started, for 3 weeks each; pain fell a little with both gums, but there was no clear difference between them. Studies with synthetic THC (dronabinol) were only laboratory measurements lasting a single dose or two days, and they did not change bowel sensitivity. The US National Academies (2017) wrote that the evidence was insufficient to say whether dronabinol works for the symptoms of this syndrome.
Does CBD work for gastroparesis (lazy stomach)?
A single small study is promising, but it is not enough. In 44 patients, high-dose pharmaceutical-grade CBD for 4 weeks reduced the total symptom score and vomiting more than placebo; on the other hand, it slowed stomach emptying slightly more, and more than half of the people taking CBD had diarrhea. This result has not yet been repeated in another study; also, the symptom score was not among the measures in the study's registration. The American Gastroenterological Association's 2025 guideline suggests that CBD not be used as a first-line treatment; the Gastroenterological Society of Australia (2026) also writes that the evidence is insufficient to recommend it.
Does CBD work for indigestion (functional dyspepsia)?
In a study of 48 people with functional dyspepsia whose stomach emptied normally, high-dose pharmaceutical-grade CBD used for 4 weeks was no different from placebo in pain, nausea, bloating or quality of life; stomach measurements did not change either. This is a single study. Germany's functional dyspepsia guideline (2026) recommends that cannabinoids not be used routinely in this condition.
Does cannabis slow the stomach?
This has been shown for THC. In controlled studies in healthy volunteers, THC and synthetic THC (dronabinol) slowed the emptying of solid food from the stomach; in one study, the amount left in the stomach 2 hours after a meal was 46% on the placebo day and 74% on the THC day. Low-dose THC given into a vein, however, did not change the emptying of liquids. In people with gastroparesis, high-dose pharmaceutical-grade CBD was also seen to slow emptying slightly. So someone with gastroparesis should talk to their doctor about cannabis products and say they use them before a gastric emptying test.
What is cannabinoid hyperemesis syndrome?
It is recurring bouts of severe nausea, vomiting and abdominal pain seen in some people who use cannabis frequently over a long time. According to a systematic review, almost all patients used cannabis at least once a week, most described relief from a hot shower, and the complaints went away when cannabis was stopped. Because the gastric emptying test can also come back delayed in these patients, it can be mistaken for gastroparesis; in a review of people with type 1 diabetes, all 11 patients diagnosed with gastroparesis while using cannabis every day were found to meet the criteria for this syndrome. If you have bouts of vomiting and use cannabis, telling your doctor openly is important for the right diagnosis.
Does CBD cause diarrhea?
At high doses, it can. In the gastroparesis study, diarrhea occurred in 12 of 21 patients taking pharmaceutical-grade CBD and 2 of 23 taking placebo; in the functional dyspepsia study, the rate was 8 of 25 people versus 1 of 23. These doses were much higher than the amounts written on over-the-counter products. Someone with diarrhea-predominant irritable bowel syndrome should keep this in mind.
Can CBD oil be obtained with a prescription for irritable bowel syndrome or stomach problems 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 classified 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 at most 0.3%. As of October 2026, there is no approved CBD medicine in Turkey. If you are thinking of using it, talk to your doctor first.
References
Show 28 references
- Zheng 2023. Phase 2, double-blind, placebo-controlled, parallel RCT (single center). PMID 37482172, doi:10.1016/j.cgh.2023.07.008, NCT03941288, results in the registry, February 13, 2024), PMC10800684. NIH support (R01-DK122280), the authors declared no relevant conflicts of interest
- Atieh 2022. Phase 2, double-blind, placebo-controlled, parallel RCT (single center; run at the same time as the gastroparesis study, same registration). PMID 35537858, doi:10.14309/ajg.0000000000001805, NCT03941288, February 13, 2024), full text could not be read
- van Orten-Luiten 2022. Double-blind, placebo-controlled, crossover RCT (exploratory; use as needed for pain). PMID 33998882, doi:10.1089/can.2020.0087, NCT03003260, no results in the registry, the registration also listed daily adequate relief and stool frequency among its measures, but the publication gives no results for them), PMC9418368
- Wong 2012. Double-blind, placebo-controlled, parallel RCT (pharmacodynamic study measuring bowel transit). PMID 22288893, doi:10.1111/j.1365-2982.2011.01874.x, NCT01253408, PMC3775711
- Wong 2011. Double-blind, placebo-controlled, parallel RCT (single-dose pharmacodynamic and pharmacogenetic study). PMID 21803011, doi:10.1053/j.gastro.2011.07.036, NCT01786109, after publication, the 75-person single-dose results in the registration belong to this study, NCT01253408, but the results of that registration belong to Wong 2012), PMC3202649
- Klooker 2011. Double-blind, placebo-controlled, randomized crossover study (rectal barostat test). PMID 20718944, doi:10.1111/j.1365-2982.2010.01587.x, Full text could not be read (abstract)
- Esfandyari 2007. Double-blind, placebo-controlled, parallel RCT (colonic barostat-manometry). PMID 17395895, doi:10.1152/ajpgi.00565.2006, Full text could not be read (abstract)
- Esfandyari 2006. Double-blind, placebo-controlled, parallel RCT. PMID 16918762, doi:10.1111/j.1365-2982.2006.00834.x, Full text could not be read (abstract), the full text of Wong 2012 reports the dose in this study as 7.5 mg twice a day, while the abstract says 5 mg
- McCallum 1999. Double-blind, randomized, placebo-controlled crossover study. PMID 9892882, doi:10.1046/j.1365-2036.1999.00441.x, Full text could not be read (abstract)
- Bateman 1983. Placebo-controlled study (randomization not stated in the abstract). PMID 6307330, doi:10.1111/j.1365-2125.1983.tb01561.x, PMC1427915 (scanned page images only, text could not be read)
- Bilbao 2022. Systematic review and meta-analysis of RCTs, separated by drug type (up to October 2021; GRADE). PMID 35982439, doi:10.1186/s12916-022-02459-1, PROSPERO CRD42021229932, PMC9389720
- Giossi 2022. Systematic review and meta-analysis (RCTs and observational studies in chronic primary pain searched; PubMed, EMBASE, Cochrane, up to October 31, 2021; GRADE). PMID 36129666, doi:10.1007/s40122-022-00434-5, PMC9633894, also a row in the fibromyalgia topic. One of the authors reported receiving conference support from Jazz Pharmaceuticals (the maker of Epidiolex)
- Volz 2016. Systematic review (up to March 2015; CENTRAL, PubMed, cannabis-med.org, ClinicalTrials.gov; at least 4 weeks and at least 10 people per arm; Cochrane risk-of-bias tool). PMID 26809974, doi:10.1007/s00482-015-0087-0, In German, full text could not be read (English abstract)
- NASEM 2017 (US National Academies report). Expert committee report (assessment based on good- and fair-quality systematic reviews and subsequent primary studies). PMID 28182367, doi:10.17226/24625, National Academies Press, chapter 4 (nap.nationalacademies.org/read/24625/chapter/6, read on October 28, 2026), the NCBI Bookshelf copy could not be opened because of bot verification
- AGA 2025 (Staller). Clinical guideline (American Gastroenterological Association; GRADE). PMID 40976635, doi:10.1053/j.gastro.2025.08.004, Full text could not be read (the publisher page redirects with a script, text-mining rights reserved)
- GESA 2026 (Kellar). Professional society position statement (Gastroenterological Society of Australia; literature review and modified Delphi consensus; 20 statements, with level of evidence and strength of recommendation). PMID 42235947, doi:10.1111/jgh.70456, PMC13326989
- DGNM 2026 (Storr). Clinical guideline, S1 level (expert consensus; German Society of Neurogastroenterology and Motility), English edition. PMID 42095009, doi:10.1155/grp/2610765, PMC13140302, original German text: Storr 2025, PMID 40199346
- ESPGHAN/NASPGHAN 2025 (Groen). Clinical guideline (European and North American pediatric gastroenterology, hepatology and nutrition societies; GRADE; 86 RCTs reviewed). PMID 40444524, doi:10.1002/jpn3.70070, PMC12314588
- CAG 2019 (Andrews). Professional society position statement (Canadian Association of Gastroenterology; non-comprehensive literature review and expert opinion; no formal grading of evidence in the statements). PMID 31294362, doi:10.1093/jcag/gwy064, PMC6507278. According to the declaration of interests, 3 of the 7 authors have ties to companies that make cannabis or cannabinoid products (advisory board and management roles, research support and product donations, speaker fees)
- Barbash 2019. Prospective, uncontrolled before-and-after study (single center). PMID 31993268, doi:10.7759/cureus.6430, PMC6970440
- Santonicola 2025. Prospective observational study; no control group. PMID 40556630, doi:10.55563/clinexprheumatol/o5ck22, Full text could not be read (abstract)
- Parkman 2025. Multicenter patient registry, cross-sectional analysis (NIDDK Gastroparesis Clinical Research Consortium, GpR3). PMID 40286861, doi:10.1016/j.cgh.2025.01.033, NCT01696747, PMC12353374
- Kilani 2026. Retrospective cohort, 1:1 propensity score matched (TriNetX electronic health records). PMID 40197432, doi:10.14309/ajg.0000000000003479, Full text could not be read (abstract)
- McCarty 2022. Retrospective cross-sectional analysis (US National Inpatient Sample, 2008–2014). PMID 33780213, doi:10.1097/MCG.0000000000001526, PMC8435035
- Akturk 2025. Retrospective chart review, case-control comparison (single center). PMID 40741458, doi:10.2337/cd24-0096, PMC12304556 (read from the PMC page), short paper by the same group: Karakus 2025, PMID 39984123, no abstract, not read)
- Parkman 2020. Multicenter patient registry, cross-sectional analysis (NIDDK Gastroparesis Clinical Research Consortium, GpR2). PMID 31758430, doi:10.1007/s10620-019-05963-2, NCT01696747, PMC7242137
- Patel 2020. Retrospective cross-sectional analysis (US National Inpatient Sample, 2010–2014). PMID 31573379, doi:10.1080/10826084.2019.1664591, Full text could not be read (abstract)
- Choi 2022. Retrospective cohort (US Nationwide Readmissions Database, 2016). PMID 33471483, doi:10.1097/MCG.0000000000001498, Full text could not be read (abstract)
Points that could not be verified from the full text
- The full text of the functional dyspepsia study (Atieh 2022) could not be read (publisher access is paid): numbers of participants, side effects and median values of measurements are from the ClinicalTrials.gov registration (NCT03941288), and statistical results are from the abstract. The values given in the registration as week-4 gastric half-emptying time for the gastroparesis groups (medians 246 and 198 minutes) are the same as the values in the publication's baseline table; which is correct could not be verified. The gastroparesis publication gives the average dose as 9.43 mg/kg; whether this is daily or per dose is not clear. The abstract says 5 patients, and the main text says 4, could not reach the full dose. The registration's change history could not be opened (access blocked); so whether the total symptom score was ever entered in the registration could not be verified
- The full texts of the THC studies of the stomach and colon (McCallum 1999, Esfandyari 2006 and 2007, Klooker 2011, Bateman 1983) could not be read; details are from the abstracts. In McCallum 1999 and Klooker 2011, the abstract does not say how THC was given. The American Gastroenterological Association's 2025 guideline, the electronic records analysis of 41,374 people (Kilani 2026), the irritable bowel syndrome hospital record analyses (Patel 2020, Choi 2022) and the fibromyalgia study (Santonicola 2025) are also from the abstracts only; the certainty of evidence and the rationale behind the AGA guideline's CBD recommendation could not be read
- The single-dose dronabinol study (Wong 2011) is cited in the publication under number NCT01253408; however, the results in this registration belong to the 36-person 2-day study (Wong 2012), while the results of the 75-person single-dose study belong to registration NCT01786109. The purity and THC content of the CBD in the CBD gum study are not stated in the publication
- The NCBI copy of the US National Academies report could not be opened because it asked for bot verification; the report was read on the publisher's (National Academies Press) website. For cannabinoid hyperemesis syndrome, the American Gastroenterological Association's 2024 clinical practice update and the Rome Foundation's 2025 consensus on idiopathic gastroparesis could be seen only as abstracts; the abstracts contain no recommendation on cannabinoid treatment
Research history
Here we record what changed on this page, when and why, with dates. The newest change is at the top.
The last sentence of the page's short summary was rewritten in a calmer, more guiding tone: instead of "Talk to your doctor before you start.", it now says whom you can talk with, and about what, if you are considering CBD. The conclusion and the strength of evidence did not change.
Page published. This first review examined 28 studies.