Loss of Appetite, Weight Loss and CBD Oil: Research, Dosage and Warnings

Q: Does CBD oil help with loss of appetite and weight loss?

We don't know yet. The only controlled study of pure CBD in patients with loss of appetite or weight loss was in anorexia nervosa; it included 32 people and lasted 3 weeks. There was a small difference in weight, but no clear difference from placebo in the main symptoms measured. CBD oil has not been tested for appetite in HIV, in older age or in other illnesses. On top of that, CBD lowered appetite in some people in the epilepsy studies. That is why the evidence for CBD is very weak.

Q: Does cannabis or THC boost appetite?

THC can increase appetite, but its benefit in preventing weight loss appears small. In AIDS-related loss of appetite, dronabinol (synthetic THC) increased appetite more than placebo over 6 weeks (38% vs 8%), but the difference in weight was not significant; megestrol, another appetite-boosting medicine, worked better for weight. In patients over 65 with reduced appetite, a spray containing THC and CBD did not change the amount eaten over one day. THC can cause dizziness, sleepiness and clouded thinking; the strength of evidence is weak. The situation is different for loss of appetite and weight loss related to cancer: in two large studies of cancer patients with this problem, THC-containing products were superior neither to placebo (243 patients) nor to the appetite stimulant megestrol (469 patients); megestrol was more effective for appetite. Reviews combining these studies also rated the certainty of the result as low or very low. Appetite and weight loss in cancer is covered on a separate page, where the evidence for THC is rated “very weak”; the “weak” rating on this page rests on increases in appetite or the amount eaten in a few small controlled studies in HIV/AIDS (the effect on weight was small and inconsistent).

Q: What is dronabinol, and is it the same as CBD oil? Is it available in Turkey?

Dronabinol is a laboratory-made form of THC, the compound in cannabis; it is a prescription medicine that can cause a high and is not the same thing as CBD oil. In the US it is approved in adults for AIDS-related loss of appetite with weight loss and for chemotherapy nausea that has not responded to other medicines. Dronabinol does not appear on the list of approved (licensed) medicines dated October 16, 2026, or on the list of active substances that can be imported from abroad dated September 28, 2026, issued by TİTCK (Turkish Medicines and Medical Devices Agency).

Q: I'm living with HIV; can cannabis or CBD affect my HIV medicines or my viral load?

No harmful effect was seen in short-term studies: in 67 people on effective HIV treatment, 21 days of smoked cannabis or dronabinol did not worsen viral load or immune cells; in 80 people, 12 weeks of a CBD-rich oil did not change viral load either. However, these were short, small studies; there are no long-term data. Some HIV medicines such as ritonavir can raise the level of dronabinol in the blood; CBD can also affect how some medicines are broken down. That is why it is important to talk about any product with your HIV doctor and your pharmacist.

Q: My older relative has no appetite; would a CBD or cannabis product help?

There are very few studies on loss of appetite in older adults. In a study of 17 patients over 65 with reduced appetite, a mouth spray containing THC and CBD did not change the amount eaten; in an older study of 15 people with Alzheimer's disease who refused to eat, weight rose more during the dronabinol period than during the placebo period. CBD oil has never been tested for this purpose, and there are only two small controlled studies with THC-containing products, so the evidence is very limited. THC-containing products can increase dizziness, sleepiness and the risk of falls in older adults. The best course is to talk with your relative's doctor about the cause of the loss of appetite and about nutritional support.

Q: Can CBD or cannabis be used in anorexia nervosa?

There are only a few small, short studies on this, and they are not enough to support a treatment recommendation. In a 3-week pilot study of pure CBD, CBD was well tolerated, but there was no clear difference from placebo in anxiety and eating disorder symptoms; in a study of dronabinol, symptoms did not change either; and a 2026 analysis pooling mental health studies also found no clear effect in anorexia nervosa. High-dose THC, meanwhile, caused intense restlessness in some people. Anorexia nervosa is an illness treated by a specialist team; if you are considering a product, talk with your treatment team first. If you are going through a hard time, sharing it with your doctor or someone you trust is a good place to start.

Q: In Turkey, can CBD oil be obtained with a prescription for loss of appetite?

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 or 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 about using it, talk with your doctor first.

What does the research say?

THC-containing medicines in HIV/AIDS: Most of the well-designed studies on this topic (studies in which participants were assigned to groups by chance and compared with an inactive drug, that is, a placebo) were done in the 1990s, before effective HIV treatment became widespread. In the largest one, 139 people with AIDS-related loss of appetite and weight loss took dronabinol (synthetic THC) or placebo for 6 weeks: appetite rose 38% from baseline with dronabinol and 8% with placebo, and nausea eased slightly. Weight, however, stayed stable with dronabinol and fell by an average of 0.4 kg with placebo; this difference was not significant. The US approval is based on this study. In another study of 12 patients (5 completed), dronabinol slightly increased body fat percentage, while weight showed only a trend. In a comparison of 52 patients, those taking dronabinol lost an average of 2 kg over 12 weeks, while those taking megestrol, an appetite-boosting hormone medicine, gained 6.5 kg, and adding dronabinol did not boost megestrol's effect.

In three small laboratory studies of people with HIV who regularly smoked cannabis (30, 10 and 7 people), dronabinol and smoked cannabis increased the calories eaten; in the first study the increase was seen only in those with muscle loss. The dronabinol doses used in these studies were several times the starting dose recommended on the label, and in the 16-day study with the highest dose, the rise in calories eaten disappeared after the first 8 days. In a 21-day in-hospital study of 67 people on effective HIV treatment, smoked cannabis and dronabinol did not harm viral load or immune cells; in this study, weight also rose slightly more than with placebo.

CBD in people living with HIV: In a 12-week study of 80 people whose virus had been suppressed for years, a CBD-rich oil (containing very little THC) was no different from placebo on the main scores of a quality-of-life scale; only the physical function score was slightly better. Appetite and weight were not measured in this study; viral load and immune cells did not change, and there was no clinically meaningful difference in liver and kidney values. In a 10-person open pilot study, one person taking high-dose pure CBD (800 mg a day) developed severe liver inflammation (hepatitis); other serious illnesses were also found in this person at the time.

What do the reviews say? A Cochrane review (2013) looked at seven small, short studies in HIV/AIDS and concluded that there was not enough evidence on the benefits and safety of cannabis or THC medicines. A meta-analysis that pooled studies in palliative care (2018) found better results than placebo for weight and appetite in HIV, based on two small studies, but rated the quality of the evidence as very low, and the authors concluded there was no high-quality evidence showing a benefit. The US National Academies' 2017 report also found the evidence for appetite and weight in HIV/AIDS to be limited. The report notes that with effective HIV treatment, AIDS-related wasting has nearly disappeared, which is why it found no good-quality new studies published after that.

Loss of appetite in older age: In a 2025 study of 17 patients over 65 with reduced appetite, a mouth spray containing THC and CBD, given as two doses on the same day, did not change the amount eaten compared with placebo; appetite-related hormones and the hunger the person felt did not change either. Of the 73 people enrolled, only 17 were able to take part until the end. In a small 1997 study of 15 people with Alzheimer's disease who refused to eat, weight rose more during the dronabinol period than during the placebo period. In two studies that reviewed the records of patients given dronabinol in nursing homes and a dementia unit (28 and 40 people), slightly more than half the patients gained weight in one (the average change was not significant), and an improvement in the amount of food eaten at meals was reported in the other; these studies had no comparison group. In a review that pooled 46 controlled studies in people over 50, THC-containing products increased side effects, especially dry mouth, dizziness and sleepiness; pure CBD did not.

Other illnesses besides cancer: In a 4-week study of 27 people with systemic sclerosis (scleroderma) who had loss of appetite or malnutrition, an oil containing THC and CBD was not clearly different from placebo for appetite, weight or calories taken in; one person taking the oil had to be hospitalized for severely low sodium. An older study that reviewed the records of 25 patients who used oral cannabinoid medicines for loss of appetite caused by hepatitis C treatment reported that weight loss slowed; however, the study was retrospective and small. In a 2024 study that reviewed the records of people with ALS (344 patients), complaints of poor appetite decreased in the short term among those using medical marijuana, but no link with maintaining weight was seen; in this study, too, the groups were not formed by chance and the content of the products used was not reported.

Does CBD increase or decrease appetite? In the controlled studies of the approved CBD medicine in Lennox–Gastaut and Dravet syndromes, appetite decreased in 16–22% of people taking CBD (5% with placebo), and 16% lost at least 5% of their body weight (8% with placebo); the effect grew with the dose. In the study in tuberous sclerosis, which used a higher dose, weight loss of at least 5% was 31% versus 8%. Two reviews looking at CBD's effect on appetite and weight also showed that reduced appetite was more common with CBD; most of these studies were done in epilepsy with high doses. On the other hand, in a single-dose study of 15 healthy adults, people ate slightly more at lunch after pure CBD. So CBD's effect on appetite may vary by person, dose and situation; there is no study showing that CBD increases appetite in patients with loss of appetite.

Anorexia nervosa: Anorexia nervosa is a serious mental illness marked by a fear of gaining weight and distress about body image; its treatment is planned by a specialist team, and as of October 2026 there is no approved medicine for this illness. There are three small controlled studies with cannabinoids. In a study of 25 women who had been ill for a long time, dronabinol added to standard treatment produced a small increase in body weight over 4 weeks compared with placebo, but eating disorder symptoms did not change. In an 11-person study from 1983, THC showed no effect different from that of a sedative medicine, and three people were withdrawn from the study because of intense restlessness and feeling unwell. In a 3-week pilot study of pure CBD in 32 women aged 18–40, CBD was well tolerated; there was a small difference in weight, but no clear difference from placebo in anxiety and eating disorder symptoms, and both groups improved similarly. A large 2026 analysis that pooled controlled studies in mental health conditions also found no clear effect of cannabinoids in anorexia nervosa. These results are not enough to support a treatment recommendation.

Strength of evidence: ●○○○ Very weak evidence. Finding: Not studied enough. What does this mean?

Ongoing research on this topic

Trials testing CBD for this topic on ClinicalTrials.gov: 1 under way; results are awaited from 1 that ended in the last two years. A trial being under way does not show that CBD works.

NCT05416697: Phase 3, completed, results awaited, 30 participants
NCT04878627: Early phase 1, active, not recruiting, target 40 participants

All trials on this topic and their history

Possible side effects and drug interactions

With THC-containing products, the most common side effects involved the nervous system: an excessively high mood (euphoria), dizziness, sleepiness, clouded thinking and confusion. According to the US prescribing information, about one in three people taking dronabinol in the AIDS studies had such a side effect, mostly in the first two weeks. Older adults may be more sensitive to these effects and to a drop in blood pressure on standing up; the risk of falls may increase. In reviews pooling studies in people over 50, THC-containing products increased side effects, while pure CBD did not. In anorexia nervosa, three people were withdrawn from a study with high-dose THC because of intense restlessness.

CBD was generally well tolerated, but it can lower appetite in some people and can raise liver enzymes at high doses: in a small study of people living with HIV, one person taking 800 mg of CBD a day developed severe liver inflammation. Both CBD and THC can affect how the body breaks down some medicines; this matters for people taking HIV medicines. Severely low sodium was reported in one person taking an oil containing THC and CBD in systemic sclerosis.

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

How was CBD used in studies?

In the HIV/AIDS studies, dronabinol was mostly used at 2.5 mg twice a day, before lunch and dinner; for those with side effects it was lowered to 2.5 mg once a day (study dose; not a personal recommendation). This is also the starting dose in the US prescribing information; for older adults or if side effects occur, starting with 2.5 mg once a day in the evening is recommended. In the laboratory studies, people who regularly smoked cannabis were given several times this amount (up to 40 mg a day) (study dose; not a personal recommendation). Dronabinol is a prescription medicine and is not the same thing as CBD oil.

In the study of older adults, the mouth spray containing THC and CBD was given twice in one day as 3 sprays each time (8.1 mg THC and 7.5 mg CBD each time); in systemic sclerosis, the oil containing THC and CBD was started at 1 drop twice a day and raised to 2 drops after 1 week (study doses; not a personal recommendation). In anorexia nervosa, pure CBD was started at 1.25 mg per kilogram of body weight twice a day and raised once a week, reaching 6.25 mg twice a day in the third week; in people living with HIV, the CBD-rich oil was used at 1 mg per kilogram twice a day (study doses; not a personal recommendation). There is no dose of CBD shown to help patients with loss of appetite.

Study details (for health professionals)

Most randomized studies of cannabinoids in loss of appetite and weight loss not related to cancer were done in HIV/AIDS and with THC-containing products. In the largest (Beal 1995; 139 randomized, 88 evaluable, 6 weeks), dronabinol 2.5 mg twice a day increased appetite more than placebo (38% vs 8%; p = 0.015), while the difference in weight was not significant (p = 0.14); the US approval is based on this study. In Struwe 1993 (12 randomized, 5 completed), only body fat percentage increased; in Timpone 1997 (52 patients, open-label), megestrol produced much more weight gain than dronabinol and adding dronabinol did not increase it; in Abrams 2003 (67 patients, 21 days), smoked cannabis and dronabinol did not adversely affect viral load or CD4 count, and weight, a secondary outcome, rose more than with placebo (as reported by reviews). Three laboratory studies of people with HIV who regularly smoked cannabis (Haney 2005 and 2007, Bedi 2010; 7–30 people) showed that cannabis and dronabinol increased calorie intake, and that at the highest dose this effect disappeared after 8 days. The Cochrane review (Lutge 2013; 7 RCTs) found the evidence on efficacy and safety insufficient; the Mücke 2018 meta-analysis found superiority over placebo in HIV for weight (2 RCTs: Abrams 2003 and Beal 1995; SMD 0.57) and appetite (Beal 1995 only; SMD 0.57) but rated the quality of evidence for the HIV outcomes as very low; NASEM 2017 found the evidence limited for HIV/AIDS and insufficient for anorexia nervosa; Bilbao 2022 found a benefit of moderate certainty for appetite with dronabinol (SMD −0.51; high heterogeneity between studies) and no difference in comparisons with active drugs. In the only placebo-controlled study of CBD in people living with HIV (Barré 2024; 80 people, 12 weeks, CBD-rich full-spectrum oil), the main quality-of-life scores did not change and appetite was not measured; in the safety publication, viral load did not change. In older age: in patients over 65 with reduced appetite, a spray containing THC and CBD did not change calorie intake over one day (Nielsen 2025; 17 of 73 completed; +10 kcal, 95% CI −55 to 75); in 15 people with Alzheimer's disease who refused to eat, weight rose more with dronabinol (Volicer 1997); two uncontrolled chart reviews (28 and 40 people) reported improvement. In systemic sclerosis, an oil containing THC and CBD was no different from placebo for appetite, weight and calories over 4 weeks (Pisprasert 2025; 27 patients); in the records of 344 patients with ALS who did or did not use medical marijuana, use was linked with fewer complaints of poor appetite in the short term but not with maintaining weight (Austin 2024). In anorexia nervosa, dronabinol produced a small weight gain over 4 weeks compared with placebo without changing symptoms (Andries 2014; 25 women); THC was no different from diazepam and 3 patients had severe dysphoria (Gross 1983; 11 women); pure CBD was well tolerated in a 3-week pilot study, with a small difference in body mass index (p = 0.046) and no difference in the prespecified behavioral measures (Sahota 2026; 32 women); a 2026 meta-analysis pooling 54 RCTs in mental health conditions found no significant effect on outcomes related to anorexia nervosa (Wilson 2026). CBD's effect on appetite: in the epilepsy studies of the approved CBD medicine, reduced appetite was 16–22% vs 5% (US prescribing information); two reviews showed that reduced appetite was more common with CBD (Spanagel 2021: OR 2.46; Pinto 2022); in 15 healthy volunteers, 193 kcal more was eaten after a single dose of pure CBD (Johnson 2026). October 2026 search (PubMed, Cochrane, ClinicalTrials.gov, Europe PMC): on this topic we found 15 randomized controlled trials with cannabinoids (12 placebo-controlled, 3 with an active drug or comparing two products; 11 with THC-containing products, 2 with pure CBD [one in healthy volunteers], 1 with a CBD-rich oil, 1 comparing pure CBD with THC:CBD), 1 non-randomized controlled laboratory study, 2 retrospective cohorts with a comparison group, 3 observational studies without a comparison group and 10 systematic reviews, meta-analyses or expert reports (1 Cochrane); all are in the table below. Of these, 12 concern HIV/AIDS, 5 older age, 5 anorexia nervosa, 3 other illnesses not related to cancer, 3 CBD's general effect on appetite and 3 more than one area. The results of a dronabinol study in HIV treatment-related nausea, which enrolled 103 people and ended in 2005, were never published; no studies from Turkey were found.

Evidence assessment: Pure CBD: very weak (in patients with loss of appetite, the only pure CBD study is a 3-week pilot study of 32 people with anorexia nervosa; in HIV, a CBD-rich oil was tested for quality of life, not appetite; in the epilepsy studies, CBD lowered appetite)

CBD alone

There is only one controlled study of pure CBD in patients with loss of appetite or weight loss: in a 3-week pilot study of 32 people with anorexia nervosa, approved (licensed) pure CBD was well tolerated and there was a small difference in body mass index compared with placebo (p = 0.046), but no difference in the main outcomes measured, anxiety and eating disorder symptoms. In 80 people living with HIV, a CBD-rich full-spectrum oil (THC below 0.3%) did not change the main quality-of-life scores over 12 weeks; appetite and weight were not measured; viral load and immune values were not affected. Pure CBD has not been tested for loss of appetite in older age, HIV-related wasting or other illnesses not related to cancer. In the epilepsy studies of the approved CBD medicine, CBD lowered appetite and increased weight loss (dose-dependent); in 15 healthy volunteers, on the other hand, people ate slightly more after a single dose of pure CBD.

Products containing THC

In HIV/AIDS, dronabinol (synthetic THC) increased appetite more than placebo in a 6-week study of 139 people (38% vs 8%; p = 0.015), but the difference in weight was not significant (stable weight vs −0.4 kg; p = 0.14); in a comparison of 52 people, megestrol led to much more weight gain (+6.5 kg vs −2.0 kg). Smoked cannabis and high-dose dronabinol increased calories eaten in the laboratory among people with HIV who regularly smoked cannabis; in a 21-day study, weight rose slightly more than with placebo and viral load did not change. Most of these studies were small, short and done before the era of effective HIV treatment. The Cochrane review found the evidence insufficient and the US National Academies found it limited; a meta-analysis based on two small studies found better results than placebo for appetite and weight in HIV but rated the quality of evidence as very low. In 17 patients over 65 with reduced appetite, a spray containing THC and CBD did not change the amount eaten over one day; in 15 people with Alzheimer's disease who refused to eat, weight rose more with dronabinol. In a study of 27 patients with systemic sclerosis, an oil containing THC and CBD was no different from placebo. In anorexia nervosa, dronabinol produced a small weight gain in a study of 25 people but did not change symptoms; an older 11-person study with THC found no difference and three people had severe restlessness; a 2026 meta-analysis pooling mental health studies also found no significant effect in anorexia nervosa.

Evidence: THC-containing products: weak (in HIV/AIDS, 4 small-to-medium RCTs and 3 small laboratory studies: appetite rose, the effect on weight was small and inconsistent, weaker than megestrol; most done before effective HIV treatment; no difference in one small RCT each in older adults and in systemic sclerosis; in anorexia nervosa, one of two small RCTs found a small weight gain and the other found no difference; insufficient evidence according to Cochrane, limited according to the US National Academies)

Approved drug regimen

Pure CBD is not approved for loss of appetite or weight loss. Dronabinol, a synthetic THC (capsule and oral solution), is approved in the US in adults for AIDS-related loss of appetite with weight loss (the drug's first US approval was in 1985; this use was added in 1992). The starting dose in the prescribing information is 2.5 mg twice a day (1 hour before lunch and dinner); for older adults or if side effects occur, starting with 2.5 mg once a day in the evening or at bedtime is recommended; the maximum is 10 mg twice a day. The approval is based on the 6-week study of 139 people (Beal 1995). Source: US prescribing information (DailyMed: dronabinol capsule, versions February 14, 2023 and May 13, 2026; dronabinol oral solution, version April 21, 2026) §1, §2.1, §14.1. The prescribing information reviewed lists no use for loss of appetite in older age, anorexia nervosa or other illnesses not related to cancer. Turkey: dronabinol, nabilone and any other THC or CBD medicine are not on the Licensed Human Medicinal Products List of TİTCK (Turkish Medicines and Medical Devices Agency) (October 16, 2026); dronabinol and nabilone do not appear on the Foreign Active Substance List (September 28, 2026), where only cannabidiol oral solution and a mouth spray containing THC and CBD are listed among medicines that require a red prescription (Turkey's controlled-drug prescription) and can be imported with TİTCK's written approval.

Dose range in studies

Dronabinol in HIV/AIDS 2.5 mg twice a day (2.5 mg/day with side effects; label maximum 10 mg twice a day); in HIV laboratory studies 10–30 mg single dose and 5–10 mg four times a day; in Alzheimer's disease 5 mg/day; in anorexia nervosa dronabinol 2.5 mg twice a day, THC 7.5–30 mg/day; smoked cannabis 1.8–3.95% THC; nabiximols (older adults) 2 × 3 sprays in one day (each 8.1 mg THC + 7.5 mg CBD); THC:CBD oil (systemic sclerosis) 2.92 mg THC + 3.24 mg CBD/day; pure CBD in anorexia nervosa 2.5–12.5 mg/kg/day (3 weeks); CBD-rich oil in HIV 2 mg/kg/day; pure CBD in the open HIV study 200–800 mg/day

Starting dose and titration

Dronabinol: per the prescribing information, start at 2.5 mg twice a day, or 2.5 mg in the evening for older adults or with side effects; increased slowly if needed. In anorexia nervosa, CBD was increased weekly: 1.25 → 3.75 → 6.25 mg per kilogram twice a day; in systemic sclerosis, THC:CBD oil from 1 drop to 2 drops twice a day after 1 week; in the open HIV study, participants increased the dose themselves as tolerated; with the CBD-rich oil (HIV), the dose was not increased. These are research protocols, not personal recommendations

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.

Loss of Appetite and Weight Loss and CBD: studies reviewed
Study	Participants	Product	Dose and titration	Duration	Result	Source
Beal 1995 Multicenter, double-blind, placebo-controlled, parallel-group RCT (USA)	139 randomized; efficacy could be evaluated in 88 people Patients with AIDS-related loss of appetite and weight loss of at least 2.3 kg (the era before effective HIV treatment became widespread)	Dronabinol (synthetic THC, capsule); placebo	2.5 mg twice a day (1 hour before lunch and dinner); 2.5 mg once a day for those with side effects (13 of 72 people, according to the US prescribing information) None; dose lowered with side effects	6 weeks	Appetite (100 mm visual scale) rose 38% from baseline with dronabinol and 8% with placebo (p = 0.015); mood improvement 10% vs −2% (p = 0.06), reduction in nausea 20% vs 7% (p = 0.05). Weight stayed stable with dronabinol and fell by an average of 0.4 kg with placebo (difference not significant, p = 0.14); those gaining at least 2 kg were 22% vs 10.5% (p = 0.11). According to the Cochrane review, RR 2.09 for gaining at least 2 kg (95% CI 0.72–6.06); randomization was done by center, and methodological quality could not be assessed. Dronabinol's US approval for loss of appetite in AIDS is based on this study (prescribing information §14.1). In the 12-month open-label extension (94 patients, no placebo), the increase in appetite continued and weight stayed stable for at least 7 months Side effects: Mostly mild to moderate: euphoria (an excessively high mood), dizziness, abnormal thinking; stopping treatment 8.3% vs 4.5%	PMID 7730690, doi:10.1016/0885-3924(94)00117-4, PMID 9223837, the authors' ties to the manufacturer are not in the sources read
Barré 2024 Double-blind, placebo-controlled, parallel-group RCT (France; the trial's primary aim was genes for intracellular cleanup [autophagy])	80 randomized; 79 analyzed (39 CBD, 40 placebo) Adults living with HIV whose virus had been suppressed for a long time (median 12 years; median age 54.6 in the table, 56.4 in the text; 30% women); not selected for loss of appetite or weight loss	Full-spectrum CBD oil (50 mg CBD per mL; Little Green Pharma, CBD 50 LGP Classic; THC below 0.3%, amount not stated); placebo medium-chain oil	1 mg per kilogram of body weight twice a day (2 mg/kg/day) None (the dose was not increased slowly; the authors note this as a limitation)	12 weeks + 4 weeks follow-up	No difference in the primary outcomes (physical and mental summary scores of the SF-36 quality-of-life scale); among the subscales, only physical function was higher with CBD at week 12 (+7.72; 95% CI 0.55–14.89; p = 0.035). Appetite and weight were not measured. In the safety publication of the same trial (Couton 2026), viral load, intracellular HIV DNA and the CD4/CD8 ratio did not change; no clinically meaningful difference in creatinine and liver enzymes; total bilirubin decreased with CBD (p = 0.046); in men, heart rate was about 8 beats/min lower (exploratory analysis) Side effects: Number of side effects similar in the two groups (56 vs 53; treatment-related 7 vs 8); most common treatment-related: throat discomfort, tiredness, worse sleep quality, nausea	PMID 39296339, doi:10.1093/ofid/ofae492, NCT05306249, PMID 41934259, one author is an employee of the manufacturing company
Abrams 2003 Randomized, placebo-controlled, three-arm study (USA; 21 days as hospital inpatients)	67 randomized; 62 in the primary analysis (20 cannabis, 22 dronabinol, 20 placebo) Adults on protease inhibitor-based HIV treatment (viral load <50 copies/mL in 58%; median CD4 340); not selected for loss of appetite	Smoked cannabis cigarette (3.95% THC), dronabinol capsule 2.5 mg or placebo capsule	Three times a day, before meals None	21 days	The primary outcome was safety: at day 21, the change in viral load compared with placebo was −0.07 log10 with cannabis (95% CI −0.30 to 0.13) and −0.04 with dronabinol (−0.20 to 0.14); CD4 and CD8 counts were not adversely affected; a small drop in protease inhibitor levels with no clinical importance (Kosel 2002). Weight was a secondary outcome: according to the Cochrane review as reported in the US National Academies report, median weight gain was 3.0 kg with cannabis (p = 0.021), 3.2 kg with dronabinol (p = 0.004) and 1.1 kg with placebo Side effects: Not reported (in the abstract); full text could not be read	PMID 12965981, doi:10.7326/0003-4819-139-4-200308190-00008, Secondary publications of the same trial: Kosel 2002 (protease inhibitor levels, PMID 11872997, Bredt 2002 (immune cells, PMID 12412840, NIMH, NCRR)
Timpone 1997 (DATRI 004) Randomized, open-label, multicenter, four-arm study (phase I/II; no placebo, compared with megestrol)	52 randomized; 47 attended at least one follow-up visit; 39 completed 12 weeks Patients with HIV-related wasting syndrome (mean CD4 59; the era before effective HIV treatment became widespread)	Dronabinol tablet and the appetite-boosting hormone medicine megestrol acetate, alone or together	Dronabinol 2.5 mg twice a day; megestrol 750 mg/day; megestrol 750 mg/day + dronabinol; megestrol 250 mg/day + dronabinol None	12 weeks	Mean weight change at 12 weeks: dronabinol alone −2.0 kg, megestrol 750 mg +6.5 kg, megestrol 750 mg + dronabinol +6.0 kg, megestrol 250 mg + dronabinol −0.3 kg (p = 0.0001 between groups); adding dronabinol did not increase megestrol's effect on weight. No significant difference between groups in side effects, stopping treatment, new AIDS-defining illness or CD4 change. Weight and hunger were linked to the blood level of megestrol, not of dronabinol Side effects: Serious side effects attributed to dronabinol involved the nervous system: confusion, anxiety, emotional lability, euphoria, hallucinations; with megestrol, shortness of breath, changes in liver enzymes and raised blood sugar	PMID 9071430, doi:10.1089/aid.1997.13.305, NCT00000737, sponsor US National Institute of Allergy and Infectious Diseases, collaborators Roxane Laboratories and Bristol-Myers Squibb)
Sahota 2026 Double-blind, placebo-controlled, parallel-group pilot RCT (USA; main aims were safety, tolerability and blood drug levels)	32 analyzed (16 CBD, 16 placebo); in addition, 1 person from each group was withdrawn at visit 2 because of a rash and not included in the analysis (the total number randomized is not clear in the text) Women aged 18–40 diagnosed with restrictive-type anorexia nervosa or atypical anorexia nervosa (mean age 22.7; some were in an intensive outpatient program)	Approved (licensed) pure CBD medicine (Epidiolex; 100 mg CBD per mL); the manufacturer supplied the drug and placebo	Per kilogram of body weight, twice a day: week 1 1.25 mg, week 2 3.75 mg, week 3 6.25 mg, with food (up to 12.5 mg/kg/day; total daily 515–861 mg in the last week) Weekly increase (schedule above)	21 days	Safety and tolerability were good. No significant difference between groups on the prespecified behavioral measures (anxiety, depression, eating disorder symptoms, desire to eat, meal consumption); both groups improved in the first week (the authors note an expectancy effect). The group × time interaction for body mass index was small but significant in favor of CBD (p = 0.046); this measure was not the study's main aim. Effect sizes for body shape concern and feeling a loss of control over eating were large in favor of CBD but not significant. The authors write that a larger study is needed because of the small sample and short duration Side effects: Limited and not serious: mild abdominal complaint in 1 person in the CBD group; a resolving rash of unclear cause in 1 person in each group (withdrawn from the study as a precaution)	PMID 41906184, doi:10.1002/eat.70034, NCT04878627, not recruiting', target 40 people, planned completion June 2026, according to the publication, data collection ended on December 31, 2024). Funded by the University of California San Diego Center for Medicinal Cannabis Research, the authors reported no conflicts of interest
Nielsen 2025 Triple-blind, placebo-controlled, crossover RCT (Denmark; investigator-initiated)	73 people enrolled; 17 completed both study days (17 analyzed) Patients over 65 with reduced appetite admitted from the emergency department for a medical reason (study days after discharge; median age 78, 76.5% women)	Nabiximols (Sativex; a mouth spray containing THC and CBD, 2.7 mg THC + 2.5 mg CBD per spray); placebo spray	2 doses on the study day, each 3 sprays (8.1 mg THC + 7.5 mg CBD per dose): the first 25 min before breakfast, the second 25 min before an ad libitum lunch None (fixed single-day dose)	Two separate study days (14 days apart)	Primary outcome (calories taken in at meals): no significant difference between spray and placebo (+10 kcal; 95% CI −55 to 75). In a secondary analysis of the same trial (Bornæs 2026), appetite-related hormones (GLP-1, ghrelin) and self-reported appetite were also no different from placebo (composite appetite score −5.2 mm; 95% CI −24.4 to 4.0). Blood THC levels varied widely from person to person (Storgaard 2026). Because only 17 of 73 people completed, the result cannot be generalized to the wider older population Side effects: 16 possibly related side effects (5 on the placebo day, 1 before the drug); all mild and temporary; no related serious side effects	PMID 40069983, doi:10.1016/j.clnu.2025.02.024, NCT05503147, PMID 37314893, PMID 42514343, PMID 40974011, the authors reported no conflicts of interest
Pisprasert 2025 Placebo-controlled, parallel-group RCT (Thailand; triple-blind according to the registry)	27 randomized (14 oil, 13 placebo); 26 analyzed Patients with systemic sclerosis (scleroderma) and loss of appetite or malnutrition, aged 18–70 (mean 55.9; most with diffuse skin involvement)	Cannabis oil made by the university, with roughly equal THC and CBD (0.73 mg THC + 0.81 mg CBD per drop), under the tongue; placebo drops	Week 1: 1 drop twice a day, then 2 drops twice a day (2.92 mg THC + 3.24 mg CBD a day) Dose doubled after 1 week if tolerated	4 weeks (+2 weeks follow-up)	At week 4, no significant difference between groups in appetite measures (hunger, enjoyment of food, ability to eat a lot), weight (+0.15 vs −0.3 kg; p = 0.36), daily calorie and protein intake, inflammation markers or quality of life. Only in within-group comparison did the hunger score rise in the oil group (p < 0.001) and not with placebo; calorie intake also rose from baseline in the oil group (p = 0.046). At baseline, calorie intake was higher in the oil group (the authors note the imbalance as a limitation). The paper gives the product concentration in two different ways (2.7 mg THC + 2.5 mg CBD per mL and 27 mg THC + 30 mg CBD per mL); the per-drop doses fit the latter Side effects: Side effects in 10 patients in the oil group and 7 with placebo (difference not significant); in 1 patient, thirst, drinking a lot of water, severely low sodium (hyponatremia) and sleepiness after the first drop, leading to hospitalization and withdrawal from the study; sleepiness in 2 patients, dizziness in 2 patients	PMID 41137182, doi:10.1186/s42238-025-00342-3, NCT05416697, PMID 40430155, difference in sleep quality not significant, p = 0.09). Khon Kaen University funded the study and made the product
Andries 2014 Double-blind, placebo-controlled, crossover RCT (Denmark; added to standard treatment)	25 randomized; 24 completed Adult women with anorexia nervosa for at least 5 years (under treatment at a specialist eating disorder center)	Dronabinol (synthetic THC); placebo	2.5 mg twice a day None	4 weeks of dronabinol + 4 weeks of placebo (with a 4-week washout in between)	Primary outcome (body weight): a small but significant increase during the dronabinol period compared with placebo (0.73 kg; p < 0.01). No change on the scale measuring eating disorder symptoms (EDI-2). In secondary publications of the same trial, the intensity of physical activity rose by 20% during the dronabinol period (daily energy expenditure higher than with placebo) and the stress hormone cortisol in urine decreased; the growth factor IGF-I did not change Side effects: No serious psychiatric side effects were seen (abstract)	PMID 24105610, doi:10.1002/eat.22173, NCT00760695, PMID 24890912, Andries 2015, PMID 26248813
Haney 2005 Double-blind, double-dummy, placebo-controlled, within-subject laboratory study (drug order randomized; USA)	30 (15 with marked loss of muscle mass, 15 without) Adults living with HIV who regularly smoke cannabis	Dronabinol capsule (0, 10, 20, 30 mg) and smoked cannabis (0%, 1.8%, 2.8%, 3.9% THC)	Single dose per session (doses above) None	Eight separate 7-hour sessions (single dose per session)	Cannabis and dronabinol clearly increased calories eaten compared with placebo in the group with muscle mass loss, but not in those with normal muscle mass. Cannabis (all doses) and the low dronabinol doses (10 and 20 mg) were well tolerated; effects on cognitive performance were small. According to the authors, these doses were at least 4–8 times the recommended starting dose Side effects: 30 mg dronabinol was poorly tolerated by some participants	PMID 15778874, doi:10.1007/s00213-005-2242-2, NCT00079560, primary outcome daily calorie intake, sponsor US National Institute on Drug Abuse)
Volicer 1997 Placebo-controlled crossover study (2 × 6 weeks; a controlled clinical trial in PubMed; randomized and double-blind according to reviews)	15 (11 completed; 1 patient who died was included in the analysis) Patients with probable Alzheimer's disease who refused to eat (hospital dementia unit, USA)	Dronabinol (synthetic THC)	2.5 mg in the morning and at noon (5 mg a day; per the study table of Cochrane 2021; one review says 2.5 mg a day) None (fixed dose)	12 weeks	Body weight rose more during the dronabinol period than during the placebo period (number not in the abstract); the severity of disturbed behavior also decreased, and in those who took dronabinol first, this effect continued into the placebo period. The only controlled study of food refusal in older adults; full text could not be read Side effects: Euphoria (an excessively high mood), sleepiness and tiredness more common during the dronabinol period, treatment not stopped; the study was stopped in 1 patient because of a major seizure and in 2 patients because of severe infection; 1 patient died of a heart attack	PMID 9309469, The only study included in the Krishnan 2009 Cochrane review (data could not be analyzed), also in the Bosnjak Kuharic 2021 Cochrane review. According to Cochrane 2021, supported by the US Department of Veterans Affairs and the dronabinol manufacturers (Roxane Laboratories, UNIMED)
Struwe 1993 Double-blind, placebo-controlled, crossover RCT (USA)	12 randomized; 5 completed Patients with symptomatic HIV infection and weight loss of at least 2.25 kg	Dronabinol; placebo	5 mg twice a day, before meals None	2 × 5 weeks (with a 2-week washout in between)	During the dronabinol period, body fat percentage rose by 1% (p = 0.04) and symptom distress decreased (p = 0.04); weight (+0.5 kg; p = 0.13), appetite score (p = 0.14) and prealbumin (p = 0.11) showed only a trend. 7 people dropped out: 2 could not tolerate the drug, 2 had disease progression, 2 were nonadherent and 1 switched to an experimental HIV treatment Side effects: 2 people stopped because they could not tolerate the drug; side effect details not in the abstract	PMID 8395916, doi:10.1177/106002809302700701, Supported by the US National Institutes of Health (NCRR)
Gross 1983 Double-blind, crossover randomized study (USA; compared with an active drug, no placebo)	11 women Women with anorexia nervosa (mean age 23.6)	THC (delta-9-tetrahydrocannabinol) capsule; comparison: the sedative drug diazepam	THC from 2.5 mg three times a day up to a maximum of 10 mg (7.5–30 mg/day); diazepam from 1 mg three times a day up to a maximum of 5 mg Doses were increased gradually	Period lengths are not in the sources read (no abstract in the PubMed record)	The primary outcome (weight) was similar in the THC and diazepam periods (the paper has no abstract; as reported by reviews: Lim 2017, Rosager 2021). In a meta-analysis (Bilbao 2022), the effect size was SMD −0.06 (95% CI −0.65 to 0.53). In one review, risk of bias was rated high with the Cochrane tool (Lim 2017) Side effects: During the THC period, 3 patients were withdrawn from the study because of severe dysphoria (intense restlessness and feeling unwell)	PMID 6308069, No abstract in the PubMed record, details from reviews (Lim 2017, Rosager 2021, Bilbao 2022)
Mboumba Bouassa 2022 (CTN PT028) Randomized, open-label pilot study (Canada; two products compared, no placebo)	10 randomized (5 THC:CBD, 5 CBD); 8 completed (the target was 26 people; closed early when the capsule supply ran out and participants could not be found) Adults on effective HIV treatment with a suppressed virus (median age 57.5; 80% men); not selected for loss of appetite	Oral capsule (>98% purified cannabinoid in oil): THC:CBD 1:1 (2.5 mg + 2.5 mg per capsule) or CBD alone (200 mg per capsule)	THC:CBD up to 15 mg + 15 mg/day; CBD up to 800 mg/day (lowered in the protocol to a maximum of 400 mg/day after two liver events) Participants increased the dose themselves as tolerated	12 weeks	Primary outcome (no significant toxicity on the WHO toxicity scale): 7 of 10 people. CD4 count, CD4/CD8 ratio and viral load did not change; no significant change on quality-of-life scales. Appetite was not measured; in the THC:CBD group, increased appetite was recorded as a side effect in 1 person Side effects: Most common was sleepiness (50%). 2 people in the CBD group were withdrawn for safety reasons at week 6: in one, pre-existing anemia worsened (with a temporary rise in liver enzymes); in the other, life-threatening acute hepatitis developed on 800 mg/day CBD (pancreatic cancer was diagnosed at the time; diabetes, fatty liver, heavy alcohol use and a possible interaction with HIV medicines were also counted among the factors)	PMID 36551926, doi:10.3390/biomedicines10123168, PMID 30659041, PMID 37508476, PMID 39063999, PMID 40879543, one author reports receiving research support from this company
Haney 2007 Double-blind, placebo-controlled, within-subject laboratory study (two 16-day inpatient stays; USA)	10 Adults living with HIV who regularly smoke cannabis	Dronabinol (5 and 10 mg) and smoked cannabis (2.0% and 3.9% THC); placebo	Each dose four times a day for 4 days (only one active on any given day) None	2 × 16 days as inpatients (each active condition 4 days)	Compared with placebo, cannabis and dronabinol increased daily calorie intake and body weight in a dose-dependent way; the effects of the two products were similar; only the 3.9% THC cannabis improved sleep scores. All active conditions except low-dose dronabinol produced a marked high (participants rated this positively); cognitive performance was not impaired Side effects: Feeling high except with low-dose dronabinol; few signs of discomfort	PMID 17589370, doi:10.1097/QAI.0b013e31811ed205, Supported by the US National Institute on Drug Abuse
Johnson 2026 Double-blind, placebo-controlled, crossover RCT (United Kingdom)	15 (4 women) Healthy adults (volunteers without loss of appetite)	CBD isolate (pure CBD); placebo	298 mg single dose None	Single dose (two separate days)	At an ad libitum lunch served 3 hours after CBD, 193 kcal more was eaten (95% CI 80–306; p = 0.003); no difference in self-reported appetite measures such as hunger; the hunger hormone ghrelin was lower with CBD; no difference in post-meal sugar and fat metabolism. Not tested in patients Side effects: Not reported (in the abstract)	PMID 41825697, doi:10.1016/j.appet.2026.108531, NCT05618756, no registration number in the abstract, institution, dose and design match the registry). In the conflict-of-interest statement, 4 of the 10 authors report ties to food, beverage and supplement companies, one (L. James) states receiving financial support from Bridge Farm Nurseries Ltd
Bedi 2010 Double-blind, placebo-controlled, within-subject laboratory study (two 16-day inpatient stays; order counterbalanced, randomization not stated; USA)	7 men Men taking HIV medicines who smoked cannabis on average 4 days a week (not selected for weight loss)	Dronabinol (from the manufacturer Unimed); placebo	5 mg four times a day for the first 2 days, then 10 mg four times a day for 14 days (40 mg/day; 8 times the recommended starting dose) Dose doubled after 2 days	2 × 16 days as inpatients	Dronabinol increased daily calorie intake in the first 8 days (mean 3,580 vs 3,228 kcal) but not on days 9–16; weight rose 1.0 kg in the first 8 days but this was not significant. The increase in desire to eat and the improvement in mood lasted 16 days. The authors think tolerance to the appetite effect developed at the high dose; according to the US prescribing information, however, tolerance to the appetite effect does not develop at recommended doses, and the effect lasted up to 5 months in patients with AIDS Side effects: Well tolerated; few negative subjective or cognitive effects; dry mouth and thirst more common	PMID 20824270, doi:10.1007/s00213-010-1995-4, Supported by the US National Institute on Drug Abuse, the authors reported no conflicts of interest
Lutge 2013 (Cochrane) Cochrane systematic review (no meta-analysis performed; search up to July 2012)	7 RCTs (8 publications) Adults living with HIV/AIDS	Dronabinol and smoked cannabis		Studies 21–84 days	Evidence on effects on illness and death is limited; studies were short, small and focused on short-term outcomes; randomization and allocation concealment were adequate in only four publications (three studies); THC's quickly noticeable effects made blinding difficult. In the single large study (Beal 1995; 88 evaluable), the chance of gaining at least 2 kg was twice as high with dronabinol (RR 2.09; 95% CI 0.72–6.06; not significant), mean +0.1 kg vs −0.4 kg. Authors: there is no evidence for the efficacy and safety of cannabis and cannabinoids in this area; there are no long-term data in people on effective HIV treatment	PMID 23633327, doi:10.1002/14651858.CD005175.pub3, Update of an earlier version (CD005175.pub2). Full text could not be read, abstract was read
NASEM 2017 (US National Academies report) Expert committee report (assessment based on good- and fair-quality systematic reviews and later primary studies)	For HIV, 2 systematic reviews (Whiting 2015: 4 RCTs, 255 patients; Lutge 2013: 7 RCTs); for anorexia nervosa, no suitable review, 1 RCT (Andries 2014) HIV/AIDS-related wasting, loss of appetite and cachexia in cancer, and anorexia nervosa	Cannabis and oral cannabinoids			Conclusion 4-4(a): there is limited evidence that cannabis and oral cannabinoids are effective in increasing appetite and decreasing weight loss in HIV/AIDS. Conclusion 4-4(b): there is insufficient evidence to say whether they are effective for loss of appetite and cachexia in cancer and for anorexia nervosa. All four HIV studies in Whiting 2015 had a high risk of bias. According to the committee, AIDS-related wasting nearly disappeared once effective HIV treatment became widespread in the mid-1990s; this is why the committee found no good-quality new studies published after that	PMID 28182367, doi:10.17226/24625, National Academies Press, Chapter 4 (nationalacademies.org/read/24625/chapter/6, read on October 28, 2026). Also a row in the Irritable Bowel and Stomach Problems and Glaucoma topics (results for those topics there)
Bilbao 2022 Systematic review and meta-analysis split by pharmacology (GRADE; search up to October 2021)	152 RCTs, 12,123 participants (10 RCTs, 599 people in the dronabinol–placebo comparison for appetite) All medical indications (appetite: studies of nausea, vomiting or loss of appetite; also anorexia nervosa)	Dronabinol, nabilone, CBD and nabiximols (smoked cannabis excluded)			Appetite: dronabinol increased appetite scores compared with placebo (SMD −0.51; 95% CI −0.87 to −0.15; moderate certainty; high heterogeneity between studies, I² 81%); there was no difference compared with active drugs (3 RCTs, 795 people; SMD −0.02; low certainty); no significant effect with nabilone, CBD (single study; SMD 0.10; p = 0.80) or nabiximols. In anorexia nervosa, dronabinol vs placebo SMD −0.47 (−0.89 to −0.05; single study), THC vs diazepam SMD −0.06 (−0.65 to 0.53; single study). Which disease groups the appetite studies came from is not given separately in the main text (supplementary tables were not read)	PMID 35982439, doi:10.1186/s12916-022-02459-1, Also a row in the Tourette Syndrome, Irritable Bowel and Stomach Problems, Parkinson's Disease and Glaucoma topics
Mücke 2018 Systematic review and meta-analysis (palliative medicine; search up to March 15, 2017)	9 RCTs, 1,561 participants (3 studies in HIV, 258 people) Palliative care: patients with cancer, HIV/AIDS and Alzheimer's disease	Cannabinoids (dronabinol, smoked cannabis, THC extracts)		HIV studies median 6 weeks (3–12)	In HIV, compared with placebo, cannabinoids were superior for weight gain (2 RCTs: Abrams 2003 and Beal 1995, 192 people; SMD 0.57; 95% CI 0.22–0.92; p = 0.001) and appetite (Beal 1995 only; SMD 0.57; 0.11–1.03; p = 0.02), and no different for nausea and vomiting; Timpone 1997 was not in the placebo analysis but in the megestrol comparison; psychiatric side effects were more common (risk difference 0.05; 0.00–0.11). Megestrol produced more weight gain than dronabinol. Quality of evidence for the HIV outcomes was very low (low or very low across the review as a whole); according to the authors, there is no convincing, unbiased, high-quality evidence that cannabinoids are of value for loss of appetite and cachexia in cancer or HIV Side effects: Psychiatric side effects increased in HIV; no difference in stopping because of side effects	PMID 29400010, doi:10.1002/jcsm.12273, PMID 26809975
Rosager 2021 (systematic review) Systematic review (PROSPERO-registered; PRISMA)	4 publications: 2 original RCTs (Gross 1983, Andries 2014) and 2 secondary publications of the same RCT Anorexia nervosa	THC and dronabinol			Few studies and a low level of evidence. The only well-designed study (Andries 2014) found a small weight gain with dronabinol compared with placebo over 4 weeks; the other study, which used high-dose THC (Gross 1983), found no effect, and according to the authors the side effects may have masked the effect. More research is needed on the effect on eating disorder symptoms	PMID 32240516, doi:10.1007/s40519-020-00891-x
Wilson 2026 Systematic review and meta-analysis of RCTs (mental disorders and substance use disorders; 1980–May 13, 2025; PROSPERO CRD42023392718; GRADE)	54 RCTs; 2,477 participants (all mental disorders; the number of studies and people in the anorexia nervosa analysis is not in the abstract) People with a mental disorder or substance use disorder (including anorexia nervosa)	Cannabinoids (THC, CBD and products containing both)			No significant difference was found between cannabinoids and placebo on outcomes related to anorexia nervosa (which studies and measures were pooled is not in the abstract; Sahota 2026 is not included because the search ended in May 2025). Significant benefit was seen only in cannabis use disorder, insomnia, tics and Tourette syndrome, and autism; certainty of evidence was low for most outcomes, and 44% of studies had a high risk of bias. The correction published in the journal (June 16, 2026) corrected the axis of a figure and the study counts in the table; the results did not change Side effects: Any side effect more common with cannabinoids (OR 1.75; 95% CI 1.25–2.46); serious side effects and dropping out did not increase	PMID 41856154, doi:10.1016/S2215-0366(26)00015-5, Correction: Lancet Psychiatry 2026, 13(8):e11, PMID 42302809, 2026, the axis of a figure and the study counts in the table were corrected, results unchanged)
Velayudhan 2021 (safety) Systematic review and meta-analysis of RCTs (mean age ≥50, all indications; 1990–October 2020; GRADE)	46 RCTs, 6,216 people (3,469 received cannabinoids; mean age 58.6) Adults over 50 (dementia, Parkinson's, MS, cancer, etc.)	THC, THC:CBD, CBD			Safety review: THC alone (all-cause side effects IRR 1.42; 1.12–1.78) and THC:CBD (1.58; 1.26–1.98) increased side effects; pure CBD did not (1.02; 0.90–1.16). Serious side effects and deaths did not increase; with THC:CBD, stopping because of side effects increased (RR 1.40; 1.08–1.80), and stopping was linked to the THC dose Side effects: Dry mouth, dizziness and sleepiness more common with THC-containing products	PMID 33780450, doi:10.1371/journal.pmed.1003524, PMID 39602500, 58 RCTs, 6, 611 people, THC dose-dependently increased dizziness, balance and coordination difficulties, thinking and perception problems and sleepiness)
Spanagel and Bilbao 2021 (systematic review and meta-analysis) Comparative systematic review and meta-analysis (PROSPERO-registered; GRADE)	15 RCTs for appetite, 1,974 people People with physical or mental illness (unwanted effects of approved cannabinoids on sleep and appetite)	Dronabinol, nabilone and CBD			Approved THC medicines (dronabinol, nabilone) did not adversely affect appetite; reduced appetite was more common with CBD than with placebo (OR 2.46; 95% CI 1.74–4.01; moderate certainty). According to the authors, the side effect of reduced appetite with approved cannabinoids appears to come from CBD	PMID 34181977, doi:10.1016/j.neuropharm.2021.108680, Full text could not be read (paywalled), abstract and as reported in the Bilbao 2022 discussion
Pinto and Martel 2022 (systematic review) Systematic review (no meta-analysis)	11 RCTs Placebo-controlled RCTs of CBD reporting appetite or weight (mostly as a side effect); mostly epilepsy studies	CBD (mostly pharmaceutical-grade pure CBD; in one study a CBD-dominant cannabis extract, in one inhaled CBD-rich cannabis)	Between 5 mg/kg and 1,500 mg a day	Search up to July 31, 2022	The authors conclude that CBD reduces appetite and that this is linked to weight loss; in seven studies reduced appetite was more common in those taking CBD (as a dose-related side effect in epilepsy), and in one study appetite increased. Only three studies measured weight, and none of the three found a weight difference compared with placebo. There are no studies in patients with loss of appetite; risk of bias was at the 'some concerns' level in most	PMID 36180814, doi:10.1007/s40261-022-01205-y
Charron 2019 (systematic review) Systematic review (narrative synthesis)	Not reported (in the abstract) People living with HIV	Cannabis and cannabinoids (THC, CBD)			Findings on cannabis's effect on the immune system in HIV are conflicting; cannabis smoke is thought to harm the lungs, and respiratory symptoms are more common in people who smoke cannabis. Safety data specific to people living with HIV regarding the lungs, cognitive function, cancer risk and cardiovascular disease are limited. The authors recommend that people living with HIV not use it until more safety data are available	PMID 31332397, doi:10.24875/AIDSRev.19000060
Austin 2024 Retrospective cohort (comparison group: those with the same symptoms who did not receive a medical marijuana prescription; not randomized)	344 patients with ALS Patients followed at a single ALS clinic with target symptoms including loss of appetite (Texas, USA)	Medical marijuana (product type and THC/CBD content not in the abstract)	Not reported (in the abstract)	2-year chart review	Medical marijuana use was linked in the short term with fewer complaints of poor appetite, pain and anxiety; there was no link with maintaining body mass index. ALS progressed faster in users, but these patients' symptoms were already more severe before the marijuana prescription (association; cause and effect not shown). The authors recommended a randomized study Side effects: Not reported (in the abstract)	PMID 39307005, doi:10.1016/j.jns.2024.123243
Costiniuk 2008 Retrospective cohort (comparison group: those not taking oral cannabinoid medicines; not randomized)	191 patients (25 took oral cannabinoid medicines) Patients receiving chronic hepatitis C treatment with interferon and ribavirin; in those taking the medicines, the main reasons were treatment-related loss of appetite (72%) and nausea (32%) (Canada)	Oral prescription medicines containing cannabinoids (which medicines were used is not in the abstract)	Not reported (in the abstract)	Throughout hepatitis C treatment (those starting between August 2003 and January 2007)	64% of those taking the medicines reported subjective improvement in symptoms; before starting there was a median weight loss of 4.5 kg, and one month after starting the medicine weight loss slowed (an additional median 0.5 kg). Interferon dose reductions were similar (8% vs 5%); the proportion who completed hepatitis C treatment and achieved a lasting cure of the virus was higher among those taking the medicines. Retrospective and small; prior cannabis smoking was more common among those taking the medicines (24% vs 10%) Side effects: Not reported (in the abstract)	PMID 18414712, doi:10.1155/2008/725702
DeJesus 2007 Retrospective chart review (no comparison group)	117 (those with weight loss before starting) Patients with HIV/AIDS started on dronabinol (USA)	Dronabinol	Not reported (in the abstract)	3–12 months of dronabinol (between 1993 and 2003)	63% of patients maintained or gained weight; in those using it for 1 year, mean weight gain was 3.7 ± 10.6 pounds (about 1.7 kg). Those reporting loss of appetite fell from 71% to 26% at month 1 (p < 0.001); nausea decreased from 38% (p = 0.031 at month 6). Because there was no comparison group, how much of the improvement was due to the drug cannot be known Side effects: Not reported (in the abstract)	PMID 17538000, doi:10.1177/1545109707300157, Funding not in the abstract (PubMed: non-US government support)
Woodward 2014 Retrospective systematic chart review (inpatient unit); no control group	40 Inpatients with severe dementia given dronabinol for behavioral or appetite problems (geriatric neuropsychiatry unit, USA)	Dronabinol (added to existing treatment)	Not reported (in the abstract) Not reported (in the abstract)	Up to 7 days after starting dronabinol	Improvement in the proportion of meals eaten, sleep duration, all domains of the Pittsburgh Agitation Scale and clinical global impression (numbers not in the abstract); no comparison group Side effects: 26 side effects, none led to stopping treatment	PMID 23597932, doi:10.1016/j.jagp.2012.11.022
Wilson 2007 Retrospective observational study (no comparison group)	28 (22 women) Long-term care residents with loss of appetite and marked weight loss not responding to nutritional support (five facilities, USA; mean age 79.5, range 46–98)	Dronabinol	Not reported (in the abstract)	12 weeks of dronabinol	15 people (53.5%) gained weight (10 gained more than 5 pounds, 6 more than 10 pounds); 11 people lost weight. Mean change +3 ± 8 pounds (about 1.4 kg; not significant, p = 0.2). Those who lost weight were younger; 64% of those who lost weight and 26% of those who gained weight died during follow-up. No comparison group Side effects: Well tolerated (abstract; no details)	PMID 17435963

Reported side effects

Dronabinol, AIDS studies: nervous system side effects 33% (mostly in the first 2 weeks); feeling good or high at low doses 8%; dizziness, euphoria, paranoid reaction, sleepiness and abnormal thinking 3–10% (US prescribing information §6.1)
Beal 1995: euphoria, dizziness, abnormal thinking; stopping treatment 8.3% vs 4.5%
Timpone 1997: serious side effects attributed to dronabinol were confusion, anxiety, emotional lability, euphoria and hallucinations
Gross 1983: severe dysphoria in 3 patients during the THC period; they were withdrawn from the study
Nielsen 2025 (spray containing THC and CBD in older adults): all possibly related side effects were mild and temporary; no related serious side effects
Pisprasert 2025 (oil containing THC and CBD): severe hyponatremia and sleepiness in 1 patient (hospitalization); sleepiness and dizziness in 2 patients each
Mboumba Bouassa 2022: sleepiness 50%; life-threatening acute hepatitis in 1 person on 800 mg/day CBD
Barré 2024 (CBD-rich oil): side effects similar to placebo; the most common treatment-related one was throat discomfort
Sahota 2026 (pure CBD): mild abdominal complaint in 1 person
Velayudhan 2021: in people over 50, THC (IRR 1.42) and THC:CBD (IRR 1.58) increased side effects, while pure CBD did not

Monitoring done in studies

When starting dronabinol and when increasing the dose, blood pressure, pulse and fainting are monitored; fall precautions are taken in older adults (dronabinol prescribing information §5.2, §8.5)
Liver enzymes should be monitored in people using high-dose CBD; hidden liver disease should be looked for (Mboumba Bouassa 2022)
In people living with HIV: studies found no harmful change in viral load or CD4 count in the short term (Abrams 2003; Barré 2024 and Couton 2026; Mboumba Bouassa 2022); there are no long-term data
Ongoing trials (October 2026): studies are under way in people living with HIV looking at the effect of THC and CBD on cognitive function (NCT04883255; 138 people, recruiting), gut microbiota and brain inflammation (NCT05514899; 90 people, recruiting) and blood levels of HIV medicines (NCT04800159; 40 people); no ongoing CBD study with appetite or weight as the main outcome was found (although the anorexia nervosa study's registration appears ongoing, its results have been published)

Who should be careful

The positive results on this page came from prescription medicines containing THC; they cannot be cited as evidence for CBD oils sold in Turkey
People whose appetite is already low: CBD can lower appetite; in the epilepsy studies of the approved CBD medicine, reduced appetite was 16–22% vs 5%, and weight loss of at least 5% was 16% vs 8% (US prescribing information, May 29, 2026); two reviews point the same way (Spanagel 2021, Pinto 2022)
Older adults: THC-containing products can cause dizziness, sleepiness, confusion and a drop in blood pressure on standing up, and can increase the risk of falls; the prescribing information recommends fall precautions in older adults with dementia and starting at a low dose (dronabinol prescribing information §8.5; Velayudhan 2021)
History of mental illness: dronabinol can worsen mania, depression and schizophrenia (prescribing information §5.1); in anorexia nervosa, intense restlessness and feeling unwell (dysphoria) were seen with THC in three patients (Gross 1983)
History of seizures: seizures have been reported with dronabinol (prescribing information §5.3); in the study of people with Alzheimer's disease who refused to eat, one patient had a major seizure (Volicer 1997)
Liver disease or high-dose CBD: in people living with HIV, one person developed life-threatening acute hepatitis on 800 mg of CBD a day; the researchers recommend monitoring liver enzymes (Mboumba Bouassa 2022)
Pregnancy and breastfeeding: dronabinol can harm the baby and is not recommended during pregnancy; mothers living with HIV are advised not to breastfeed (dronabinol prescribing information §8.1, §8.2)
Smoked cannabis: the smoke is thought to harm the lungs; safety data in people living with HIV are limited (Charron 2019)
Anyone taking long-term medicines should consult a doctor before starting; the studies on this page did not systematically assess drug interactions

Drug interactions

The studies on this page did not systematically assess drug interactions; details are in the site's Drug interactions section
Dronabinol is broken down in the body by the enzymes CYP2C9 and CYP3A4: ritonavir, ketoconazole, itraconazole, clarithromycin, erythromycin and grapefruit juice (CYP3A4), as well as amiodarone and fluconazole (CYP2C9), can raise dronabinol levels and its side effects (US prescribing information §7.3)
Sleep-inducing medicines, alcohol, benzodiazepines and opioids: sleepiness and dizziness increase when combined with dronabinol (prescribing information §5.1, §7.1)
Highly protein-bound drugs with a narrow therapeutic range (warfarin, cyclosporine, amphotericin B): side effects should be monitored when starting dronabinol; this interaction has not been confirmed in the body (prescribing information §7.4)
HIV protease inhibitors: smoked cannabis lowered the peak blood level of indinavir by 14% over 14 days, with no significant change for nelfinavir; the researchers judged this to have no clinical importance in the short term (Kosel 2002, Abrams 2003 trial)
CBD and HIV medicines: no systematic data; in a case of hepatitis seen with high-dose CBD, the researchers also counted a possible interaction with HIV medicines among the factors (Mboumba Bouassa 2022)
In people whose CYP2C9 enzyme works poorly, blood levels of dronabinol may rise (prescribing information §8.6)

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

Talk with your doctor, and your dietitian if you have one, about the cause of unintended weight loss and loss of appetite; this helps plan the right support for you.
If you take HIV medicines, talk about cannabis, THC or CBD products with your HIV doctor and your pharmacist; some medicines such as ritonavir can raise the level of dronabinol in the blood.
If you are considering it for an older loved one, talk with their doctor about the risk of falls, sleepiness and confusion; THC-containing products can increase these effects in older adults.
If you or a loved one is living with anorexia nervosa or another eating disorder, talk with your treatment team before trying any product; these products do not replace treatment.
Check the product's label and certificate of analysis (COA), and find out whether it contains THC; THC-containing products can affect driving.

Choose your age group, medicines and situation to prepare a short summary you can take to your doctor. Nothing you enter is sent anywhere.

Open the personal warnings tool Convert drops ↔ mg

Bottom line

It is not known whether CBD oil helps with loss of appetite and weight loss not related to cancer; there are very few studies on this, and CBD can lower appetite in some people. Prescription medicines containing THC increased appetite in HIV/AIDS, but their effect on weight was small and the evidence is weak; a 2025 study of older adults with reduced appetite found no difference. Talking with your doctor about the cause of your loss of appetite and the nutritional support that is right for you is the safest path.

Frequently asked questions

Does CBD oil help with loss of appetite and weight loss?

We don't know yet. The only controlled study of pure CBD in patients with loss of appetite or weight loss was in anorexia nervosa; it included 32 people and lasted 3 weeks. There was a small difference in weight, but no clear difference from placebo in the main symptoms measured. CBD oil has not been tested for appetite in HIV, in older age or in other illnesses. On top of that, CBD lowered appetite in some people in the epilepsy studies. That is why the evidence for CBD is very weak.

Does cannabis or THC boost appetite?

THC can increase appetite, but its benefit in preventing weight loss appears small. In AIDS-related loss of appetite, dronabinol (synthetic THC) increased appetite more than placebo over 6 weeks (38% vs 8%), but the difference in weight was not significant; megestrol, another appetite-boosting medicine, worked better for weight. In patients over 65 with reduced appetite, a spray containing THC and CBD did not change the amount eaten over one day. THC can cause dizziness, sleepiness and clouded thinking; the strength of evidence is weak. The situation is different for loss of appetite and weight loss related to cancer: in two large studies of cancer patients with this problem, THC-containing products were superior neither to placebo (243 patients) nor to the appetite stimulant megestrol (469 patients); megestrol was more effective for appetite. Reviews combining these studies also rated the certainty of the result as low or very low. Appetite and weight loss in cancer is covered on a separate page, where the evidence for THC is rated “very weak”; the “weak” rating on this page rests on increases in appetite or the amount eaten in a few small controlled studies in HIV/AIDS (the effect on weight was small and inconsistent).

What is dronabinol, and is it the same as CBD oil? Is it available in Turkey?

Dronabinol is a laboratory-made form of THC, the compound in cannabis; it is a prescription medicine that can cause a high and is not the same thing as CBD oil. In the US it is approved in adults for AIDS-related loss of appetite with weight loss and for chemotherapy nausea that has not responded to other medicines. Dronabinol does not appear on the list of approved (licensed) medicines dated October 16, 2026, or on the list of active substances that can be imported from abroad dated September 28, 2026, issued by TİTCK (Turkish Medicines and Medical Devices Agency).

I'm living with HIV; can cannabis or CBD affect my HIV medicines or my viral load?

No harmful effect was seen in short-term studies: in 67 people on effective HIV treatment, 21 days of smoked cannabis or dronabinol did not worsen viral load or immune cells; in 80 people, 12 weeks of a CBD-rich oil did not change viral load either. However, these were short, small studies; there are no long-term data. Some HIV medicines such as ritonavir can raise the level of dronabinol in the blood; CBD can also affect how some medicines are broken down. That is why it is important to talk about any product with your HIV doctor and your pharmacist.

My older relative has no appetite; would a CBD or cannabis product help?

There are very few studies on loss of appetite in older adults. In a study of 17 patients over 65 with reduced appetite, a mouth spray containing THC and CBD did not change the amount eaten; in an older study of 15 people with Alzheimer's disease who refused to eat, weight rose more during the dronabinol period than during the placebo period. CBD oil has never been tested for this purpose, and there are only two small controlled studies with THC-containing products, so the evidence is very limited. THC-containing products can increase dizziness, sleepiness and the risk of falls in older adults. The best course is to talk with your relative's doctor about the cause of the loss of appetite and about nutritional support.

Can CBD or cannabis be used in anorexia nervosa?

There are only a few small, short studies on this, and they are not enough to support a treatment recommendation. In a 3-week pilot study of pure CBD, CBD was well tolerated, but there was no clear difference from placebo in anxiety and eating disorder symptoms; in a study of dronabinol, symptoms did not change either; and a 2026 analysis pooling mental health studies also found no clear effect in anorexia nervosa. High-dose THC, meanwhile, caused intense restlessness in some people. Anorexia nervosa is an illness treated by a specialist team; if you are considering a product, talk with your treatment team first. If you are going through a hard time, sharing it with your doctor or someone you trust is a good place to start.

In Turkey, can CBD oil be obtained with a prescription for loss of appetite?

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 or 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 about using it, talk with your doctor first.

References

Show 31 references

Beal 1995. Multicenter, double-blind, placebo-controlled, parallel-group RCT (USA). PMID 7730690, doi:10.1016/0885-3924(94)00117-4, PMID 9223837, the authors' ties to the manufacturer are not in the sources read
Barré 2024. Double-blind, placebo-controlled, parallel-group RCT (France; the trial's primary aim was genes for intracellular cleanup [autophagy]). PMID 39296339, doi:10.1093/ofid/ofae492, NCT05306249, PMID 41934259, one author is an employee of the manufacturing company
Abrams 2003. Randomized, placebo-controlled, three-arm study (USA; 21 days as hospital inpatients). PMID 12965981, doi:10.7326/0003-4819-139-4-200308190-00008, Secondary publications of the same trial: Kosel 2002 (protease inhibitor levels, PMID 11872997, Bredt 2002 (immune cells, PMID 12412840, NIMH, NCRR)
Timpone 1997 (DATRI 004). Randomized, open-label, multicenter, four-arm study (phase I/II; no placebo, compared with megestrol). PMID 9071430, doi:10.1089/aid.1997.13.305, NCT00000737, sponsor US National Institute of Allergy and Infectious Diseases, collaborators Roxane Laboratories and Bristol-Myers Squibb)
Sahota 2026. Double-blind, placebo-controlled, parallel-group pilot RCT (USA; main aims were safety, tolerability and blood drug levels). PMID 41906184, doi:10.1002/eat.70034, NCT04878627, not recruiting', target 40 people, planned completion June 2026, according to the publication, data collection ended on December 31, 2024). Funded by the University of California San Diego Center for Medicinal Cannabis Research, the authors reported no conflicts of interest
Nielsen 2025. Triple-blind, placebo-controlled, crossover RCT (Denmark; investigator-initiated). PMID 40069983, doi:10.1016/j.clnu.2025.02.024, NCT05503147, PMID 37314893, PMID 42514343, PMID 40974011, the authors reported no conflicts of interest
Pisprasert 2025. Placebo-controlled, parallel-group RCT (Thailand; triple-blind according to the registry). PMID 41137182, doi:10.1186/s42238-025-00342-3, NCT05416697, PMID 40430155, difference in sleep quality not significant, p = 0.09). Khon Kaen University funded the study and made the product
Andries 2014. Double-blind, placebo-controlled, crossover RCT (Denmark; added to standard treatment). PMID 24105610, doi:10.1002/eat.22173, NCT00760695, PMID 24890912, Andries 2015, PMID 26248813
Haney 2005. Double-blind, double-dummy, placebo-controlled, within-subject laboratory study (drug order randomized; USA). PMID 15778874, doi:10.1007/s00213-005-2242-2, NCT00079560, primary outcome daily calorie intake, sponsor US National Institute on Drug Abuse)
Volicer 1997. Placebo-controlled crossover study (2 × 6 weeks; a controlled clinical trial in PubMed; randomized and double-blind according to reviews). PMID 9309469, The only study included in the Krishnan 2009 Cochrane review (data could not be analyzed), also in the Bosnjak Kuharic 2021 Cochrane review. According to Cochrane 2021, supported by the US Department of Veterans Affairs and the dronabinol manufacturers (Roxane Laboratories, UNIMED)
Struwe 1993. Double-blind, placebo-controlled, crossover RCT (USA). PMID 8395916, doi:10.1177/106002809302700701, Supported by the US National Institutes of Health (NCRR)
Gross 1983. Double-blind, crossover randomized study (USA; compared with an active drug, no placebo). PMID 6308069, No abstract in the PubMed record, details from reviews (Lim 2017, Rosager 2021, Bilbao 2022)
Mboumba Bouassa 2022 (CTN PT028). Randomized, open-label pilot study (Canada; two products compared, no placebo). PMID 36551926, doi:10.3390/biomedicines10123168, PMID 30659041, PMID 37508476, PMID 39063999, PMID 40879543, one author reports receiving research support from this company
Haney 2007. Double-blind, placebo-controlled, within-subject laboratory study (two 16-day inpatient stays; USA). PMID 17589370, doi:10.1097/QAI.0b013e31811ed205, Supported by the US National Institute on Drug Abuse
Johnson 2026. Double-blind, placebo-controlled, crossover RCT (United Kingdom). PMID 41825697, doi:10.1016/j.appet.2026.108531, NCT05618756, no registration number in the abstract, institution, dose and design match the registry). In the conflict-of-interest statement, 4 of the 10 authors report ties to food, beverage and supplement companies, one (L. James) states receiving financial support from Bridge Farm Nurseries Ltd
Bedi 2010. Double-blind, placebo-controlled, within-subject laboratory study (two 16-day inpatient stays; order counterbalanced, randomization not stated; USA). PMID 20824270, doi:10.1007/s00213-010-1995-4, Supported by the US National Institute on Drug Abuse, the authors reported no conflicts of interest
Lutge 2013 (Cochrane). Cochrane systematic review (no meta-analysis performed; search up to July 2012). PMID 23633327, doi:10.1002/14651858.CD005175.pub3, Update of an earlier version (CD005175.pub2). Full text could not be read, abstract was read
NASEM 2017 (US National Academies report). Expert committee report (assessment based on good- and fair-quality systematic reviews and later primary studies). PMID 28182367, doi:10.17226/24625, National Academies Press, Chapter 4 (nationalacademies.org/read/24625/chapter/6, read on October 28, 2026). Also a row in the Irritable Bowel and Stomach Problems and Glaucoma topics (results for those topics there)
Bilbao 2022. Systematic review and meta-analysis split by pharmacology (GRADE; search up to October 2021). PMID 35982439, doi:10.1186/s12916-022-02459-1, Also a row in the Tourette Syndrome, Irritable Bowel and Stomach Problems, Parkinson's Disease and Glaucoma topics
Mücke 2018. Systematic review and meta-analysis (palliative medicine; search up to March 15, 2017). PMID 29400010, doi:10.1002/jcsm.12273, PMID 26809975
Rosager 2021 (systematic review). Systematic review (PROSPERO-registered; PRISMA). PMID 32240516, doi:10.1007/s40519-020-00891-x
Wilson 2026. Systematic review and meta-analysis of RCTs (mental disorders and substance use disorders; 1980–May 13, 2025; PROSPERO CRD42023392718; GRADE). PMID 41856154, doi:10.1016/S2215-0366(26)00015-5, Correction: Lancet Psychiatry 2026, 13(8):e11, PMID 42302809, 2026, the axis of a figure and the study counts in the table were corrected, results unchanged)
Velayudhan 2021 (safety). Systematic review and meta-analysis of RCTs (mean age ≥50, all indications; 1990–October 2020; GRADE). PMID 33780450, doi:10.1371/journal.pmed.1003524, PMID 39602500, 58 RCTs, 6, 611 people, THC dose-dependently increased dizziness, balance and coordination difficulties, thinking and perception problems and sleepiness)
Spanagel and Bilbao 2021 (systematic review and meta-analysis). Comparative systematic review and meta-analysis (PROSPERO-registered; GRADE). PMID 34181977, doi:10.1016/j.neuropharm.2021.108680, Full text could not be read (paywalled), abstract and as reported in the Bilbao 2022 discussion
Pinto and Martel 2022 (systematic review). Systematic review (no meta-analysis). PMID 36180814, doi:10.1007/s40261-022-01205-y
Charron 2019 (systematic review). Systematic review (narrative synthesis). PMID 31332397, doi:10.24875/AIDSRev.19000060
Austin 2024. Retrospective cohort (comparison group: those with the same symptoms who did not receive a medical marijuana prescription; not randomized). PMID 39307005, doi:10.1016/j.jns.2024.123243
Costiniuk 2008. Retrospective cohort (comparison group: those not taking oral cannabinoid medicines; not randomized). PMID 18414712, doi:10.1155/2008/725702
DeJesus 2007. Retrospective chart review (no comparison group). PMID 17538000, doi:10.1177/1545109707300157, Funding not in the abstract (PubMed: non-US government support)
Woodward 2014. Retrospective systematic chart review (inpatient unit); no control group. PMID 23597932, doi:10.1016/j.jagp.2012.11.022
Wilson 2007. Retrospective observational study (no comparison group). PMID 17435963

Points that could not be verified from the full text

Studies whose full text could not be read: Beal 1995 and 1997, Struwe 1993, Timpone 1997, Abrams 2003 (the publisher's page required bot verification) and its secondary publications, Haney 2005 and 2007, Andries 2014 and its secondary publications, Nielsen 2025 (the full text of its secondary analysis was read), Volicer 1997, Gross 1983 (no abstract on PubMed either; details from reviews), Johnson 2026, Couton 2026, DeJesus 2007, Wilson 2007, Woodward 2014, Costiniuk 2008, Lutge 2013 (Cochrane), Charron 2019, Rosager 2021, Spanagel 2021, Wilson 2026 and Austin 2024. The numbers in these rows were taken from abstracts, registry records, the US prescribing information and as reported by reviews. The weight result of Abrams 2003 comes only from reviews (the Cochrane review as reported in the US National Academies report, and the Mücke 2018 full text).
A study with unpublished results: a 103-person phase 2 study comparing dronabinol with placebo in nausea and vomiting caused by HIV treatment (NCT00642499; sponsored by the drug's license holder at the time, completed in April 2005); the registry lists no results and no publication.
In the systemic sclerosis study (Pisprasert 2025), the concentration of the cannabis oil is written in two different ways in the paper (2.7 mg THC + 2.5 mg CBD per mL and 27 mg THC + 30 mg CBD per mL); the table uses the paper's per-drop doses. In the registry the product is called CBD oil, but it also contains THC.
The registration of the anorexia nervosa CBD study (NCT04878627) still appears ongoing and targets 40 people; the publication gives an analysis of 32 people and states that data collection ended in December 2024. It is not known whether results for the remaining people will be published separately.
Turkey: the statement that dronabinol and nabilone are not licensed in Turkey is based on their absence from the TİTCK (Turkish Medicines and Medical Devices Agency) lists (Licensed Human Medicinal Products List October 16, 2026; Foreign Active Substance List September 28, 2026); whether they could be imported by some other route was not investigated in this search.

Research history

Here we record what changed on this page, when and why, with dates. The newest change is at the top.

October 30, 2026
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.
October 30, 2026
The frequently asked questions now explain why the strength of evidence for THC-containing products is “Weak” on this page and “Very weak” on the Appetite Loss in Cancer page: in loss of appetite related to HIV/AIDS, appetite or the amount eaten increased in a few small controlled studies; in two large studies with cancer patients, however, THC-containing products did no better than either placebo or the appetite stimulant megestrol. No new study was added; the conclusion and the strength of evidence did not change.
October 30, 2026
The correction to the Wilson 2026 meta-analysis published in the journal was read: only the axis of one figure and the study counts in a table were corrected; the results on this page did not change.
Related studies: Wilson 2026, published April 2026 (PMID 41856154, doi:10.1016/S2215-0366(26)00015-5, Correction: Lancet Psychiatry 2026, 13(8):e11, PMID 42302809, 2026, the axis of a figure and the study counts in the table were corrected, results unchanged))

Older changes (1)

October 28, 2026
Page published. This first review examined 31 studies.

Evidence changes across all topics