Some women go into surgery with every textbook sign of endometriosis — crippling period pain, pain during sex, pelvic pain that never quite switches off — and come out without a diagnosis, because the surgeon and the pathologist cannot find the disease. A new CBD, endometriosis, anxiety and depression study from Los Angeles looks at exactly this group, and at what these women reach for when the pain and the low mood do not let up.
The paper was published in October 2026 in the European Journal of Obstetrics & Gynecology and Reproductive Biology: X by a team from Cedars-Sinai Medical Center led by gynecologists Raanan Meyer and Kelly Wright. It draws on 454 women who had surgery between October 2019 and March 2024 and finds that those with suspected but unconfirmed endometriosis were more likely to be living with depression than women whose diagnosis was confirmed — and more likely to say they used cannabis for it.
This page first covered a different piece of the puzzle: a small pilot study at Berlin’s Charité university hospital that gave 27 women with confirmed endometriosis an oral THC and CBD extract for their pain. That study is still described in full below. The Cedars-Sinai paper adds the mental-health side and a group of patients that most cannabis studies leave out.
Key Takeaways
The Cedars-Sinai study followed 454 women scheduled for gynecologic surgery: 308 had endometriosis confirmed under the microscope, 101 had no sign of it and no suspicion of it, and 45 had symptoms that strongly suggested endometriosis but no lesions were found — a group the authors call clinically suspected endometriosis with negative pathology. These 45 women described pain much like confirmed patients, and after the researchers accounted for other factors they had more than twice the odds of depression needing treatment (adjusted odds ratio 2.27, 95% CI 1.03–5.03) and nearly twice the odds of using cannabis for depression (aOR 1.92, 95% CI 1.01–3.65); anxiety rates did not differ significantly between groups. Women without endometriosis were far less likely than confirmed patients to use cannabis or CBD for pain. The study is observational, relies on self-reports, recorded no doses and had only 45 women in the key group, so it shows associations, not causes, and it says nothing about whether CBD or cannabis helps. The earlier Charité pilot, in which 27 women took about 10 mg of THC and 12 mg of CBD a day on top of their usual treatment, reported average pain falling from 6.81 to 4.93 out of 10 over three months, fewer days on painkillers and better quality of life, but nearly every participant had side effects and there was no placebo group. Taken together, the two studies describe real patient demand and a plausible research direction, not a proven treatment.
The CBD, Endometriosis, Anxiety and Depression Study: Who Was Included
Endometriosis is a condition in which tissue similar to the lining of the womb grows outside it, mostly in the pelvis. The standard way to confirm it is still surgery: a surgeon looks inside the abdomen through a small camera, removes suspicious tissue and a pathologist checks it under the microscope. If the tissue does not show endometrial-type cells, the official answer is “no endometriosis” — even if the woman’s symptoms say otherwise.
That gap is what the Cedars-Sinai team wanted to study. Their data come from a long-running research program called Biologic and Epidemiologic Markers of Endometriosis (BEME), run at a large specialist hospital in Los Angeles. Women aged 18 to 60 who were due for surgery for pelvic pain, suspected endometriosis, fibroids or other benign conditions were invited to take part, and each filled in a detailed questionnaire based on international endometriosis research standards. Questions about cannabis and CBD were added on top.
The 454 women were split into three groups. In 308 (67.8%), endometriosis was confirmed by pathology; 39.3% of them had the more advanced stages 3 or 4. Another 101 (22.2%) formed the comparison group: no one suspected endometriosis before surgery and none was seen during it. Many of them were having fibroids removed. The remaining 45 (9.9%) are the focus of the new paper: women whose symptoms and examination pointed strongly to endometriosis, who had tissue removed — biopsies of the pelvic lining were taken in every case — and whose pathology came back negative.
The definitions matter for reading the results. Anxiety and depression were counted if a woman said a doctor had diagnosed the condition and it had required medication or therapy at some point. Cannabis and CBD use meant using them on ten or more occasions in her lifetime, and women could tick what they used them for: pain, anxiety or depression.
Depression Stood Out in Women Without a Confirmed Diagnosis
On the surface, the three groups looked fairly similar. Anxiety needing treatment was reported by 52.9% of women with confirmed endometriosis, 60.0% of those with suspected but unconfirmed disease and 46.5% of the comparison group. For depression the figures were 42.2%, 60.0% and 40.6%. These raw differences were not statistically significant (p = .292 for anxiety and p = .063 for depression), meaning chance could not be ruled out.
The picture changed once the researchers adjusted for other factors, mainly race, which differed between the groups. Compared with confirmed cases, women with suspected but unconfirmed endometriosis had 2.27 times the odds of depression (95% CI 1.03–5.03, p = .043). The result held when pain characteristics were added to the model. For anxiety there was no meaningful difference (aOR 0.74, 95% CI 0.33–1.65). The comparison group did not differ from confirmed cases on either measure.
The pain profile of the unconfirmed group was much closer to endometriosis than to the comparison group. Pain during sex was reported by 72.7% of confirmed patients, 66.7% of the unconfirmed group and 41.6% of the comparison group. Pelvic pain outside the period affected 67.9%, 68.9% and 50.5%. Severe period cramps that needed medication and bed rest were reported by 72.1%, 57.8% and 38.6%. The women in the unconfirmed group were also about the same age as confirmed patients (34.8 and 35.6 years on average), while the comparison group was older (41.9).
One detail the authors flag as a possible clue: 37.8% of women in the unconfirmed group had a hysterectomy during their surgery, compared with 22.7% of confirmed patients. The authors suggest this might reflect a more catastrophic experience of symptoms that pushes women and doctors toward a definitive operation. They did not measure this, so it remains a hypothesis.
Cannabis and CBD: Used for Pain, and Sometimes for Mood
According to the paper’s main table, around half of the women in every group reported having used cannabis at least ten times (51.7%, 51.7% and 55.9%), and the groups did not differ significantly. Lifetime CBD use was reported by 46.4% of confirmed patients, 50.0% of the unconfirmed group and 31.7% of the comparison group, again without a statistically significant difference.
The reasons for use told a clearer story. Cannabis was used for pain by 28.9% of women with confirmed endometriosis, 22.2% of the unconfirmed group and 10.9% of the comparison group. For CBD, the figures were 26.3%, 20.0% and 10.9%. After adjustment, the comparison group was much less likely than confirmed patients to use cannabis for pain (aOR 0.32, 95% CI 0.16–0.64) and CBD for pain (aOR 0.36, 95% CI 0.18–0.72). In other words, pelvic pain itself — confirmed or not — goes hand in hand with reaching for cannabinoids.
The headline association with mood is narrower than the paper’s title suggests. Women in the unconfirmed group had 1.92 times the odds of using cannabis for depression compared with confirmed patients (95% CI 1.01–3.65, p = .047). But the raw numbers behind this are small: three women in the unconfirmed group and 19 among confirmed patients said they used cannabis for depression (6.7% and 6.2%). For CBD specifically, there was no difference in use for depression (aOR 0.99, 95% CI 0.28–3.55). So the signal concerns cannabis in general, not CBD-only products, and it rests on a handful of women.
If you want to understand what CBD is and how it differs from THC, our complete CBD guide covers the basics. What matters here is that neither this study nor the Charité pilot tested CBD on its own as a treatment.
Why Pain and Low Mood Travel Together
The authors’ explanation starts with something doctors have long observed: chronic pain and depression feed each other. Pain wears people down, and a low mood can make pain feel louder. Brain imaging studies have found changes in women with chronic pelvic pain, including a 2024 study in Human Reproduction that linked lower grey-matter volume in women with pelvic pain without endometriosis to depressive symptoms. The Cedars-Sinai team sees this as one plausible pathway behind their findings.
There is also a more mundane possibility: the disease may have been there but was missed. Endometriosis lesions can be microscopic. The same research group has previously found that in about 15% of patients whose first tissue slide was negative, examining deeper layers of the sample revealed endometriosis after all. Some experts have proposed the term “endometriosis stage 0” for women with strongly suspected disease that is not confirmed by histology. Diagnostic practice is also shifting, with growing calls to diagnose endometriosis from symptoms and imaging rather than waiting for surgery.
Then there is the human side. A woman who has gone through an operation in the hope of an answer and is told “nothing was found” may face doubt from others and from herself, with no clear treatment plan. The study cannot measure this, but it is a reasonable context for the higher rate of depression.
What This Study Cannot Tell Us
The limitations are substantial, and the authors list most of them themselves. The study is observational: it shows that depression and cannabis use for depression were more common in one group, not that one causes the other. The arrow could easily point the other way — depression may lead women to try cannabis, rather than cannabis leading to depression. Anxiety and depression were self-reported as past diagnoses, so recall errors are possible.
Only 45 women were in the key group, which limits how much weight any single figure can carry; the confidence intervals for both main findings start barely above 1. The researchers did not record doses, products or how cannabis was taken, so nothing can be said about dose and effect. The cohort was mostly white (69.2% of confirmed patients and 82.2% of the unconfirmed group), and the comparison group was not a group of healthy women: almost half had fibroids removed, and fibroids themselves are linked to anxiety and depression.
Careful readers will also notice that some rows in the published tables repeat identical figures for cannabis and CBD, and some percentages do not match the stated group sizes. This does not overturn the main results, but it is another reason to treat individual numbers with caution.
Finally, the authors stress that these associations should be read alongside the known risks of cannabis and CBD products. Long-term users and people with a personal or family history of psychiatric illness are the most vulnerable to adverse effects such as cognitive decline, psychosis and severe cyclic vomiting. The study was funded by the US National Institutes of Health, and the authors declared no conflicts of interest.
Inside the Cannabinoid Extract Endometriosis Pain Study: Who Took Part
For many women with endometriosis, the month splits into two kinds of days: days with pain, and days spent waiting for it. The women who joined a small study at Berlin’s Charité university hospital had already worked through the standard toolbox — hormone therapy, surgery, ibuprofen and diclofenac, and in some cases opioids — and were still reaching for painkillers on most days of the month. So the researchers added one more thing to their existing treatment: a measured dose of a standardized cannabis extract, taken by mouth twice a day, for three months.
The results were published on 22 September 2026 in PLOS One by a team from the Endometriosis Research Centre Charité, with gynecologist Sylvia Mechsner as senior author. They point to lower pain, fewer days on painkillers and a better quality of life. They also come with caveats that matter just as much: only 27 women completed the study, there was no placebo group, and every participant knew exactly what she was taking.
The pain is not only about inflammation at the site of the lesions. Over time, the nervous system itself can become oversensitive — a process called central sensitization. A useful comparison is a smoke alarm whose sensitivity has been turned up so high that it goes off when someone makes toast. The signal is real, but the system amplifies it. This matters for the study, because it is one reason why standard painkillers often fall short. The authors note that endometriosis affects roughly 10% of women of reproductive age worldwide.
Between 1 November 2023 and 31 May 2025, the Charité team enrolled 30 women. To qualify, participants needed a confirmed endometriosis diagnosis, chronic pelvic pain that was not tied to the menstrual cycle for at least three months, an average pain score of at least 4 out of 10, and a separate clinical indication for cannabis-based medicine documented by a Charité physician. Women with heart disease, known addiction disorders, regular cannabis use in the previous six months, or who were pregnant, breastfeeding or planning a pregnancy were excluded. Nobody was allowed to start or change hormonal therapy in the three months before or during the study, so that any change could not simply be put down to new hormones.
One woman dropped out during the build-up phase because of side effects, and two stopped returning questionnaires. That left 27 participants with an average age of 32.9 years. Most (81.5%) were already on hormone therapy. Just over half (51.9%) had the mildest surgical stage of the disease, rASRM stage I — a reminder that in endometriosis, how the lesions look during surgery says little about how much pain a woman feels. On average, participants had undergone 1.9 operations.
The baseline picture was one of pain that ordinary treatment was not controlling. In the month before the study, average pain intensity was 6.8 out of 10. Almost everyone had used anti-inflammatory painkillers, many also used paracetamol and metamizole, and seven women were actively taking opioids such as tramadol, tilidine or oxycodone when the study began. Twelve had never used opioids. Cannabis experience was limited: 37.0% had never used it, 44.4% had tried it only in the distant past, and none used it regularly.
The Dose: What the Participants Actually Took
All participants received an oral cannabinoid-based medicinal product classified as chemotype 2, meaning it contains THC and CBD in roughly equal amounts. The extract contained 12 mg of THC and 14 mg of CBD per millilitre. Rather than starting at full strength, the team raised the dose slowly over four weeks to limit side effects, and called each participant by phone after two and four weeks to check how she was coping.
After this build-up, the target was about 10 mg of THC and 12 mg of CBD per day, split into two doses: 0.3 ml in the morning (3.6 mg THC and 4.2 mg CBD) and 0.5 ml in the evening (6.0 mg THC and 7.0 mg CBD). Over the three months, the median doses stayed at exactly those levels. The authors read this as a sign that the effect did not fade and that participants did not need more and more of the extract to get the same relief — although three months is too short to say anything about the longer term.
The extract was supplied by the cannabis company Khiron and dispensed through Bezirks Apotheken, which the authors acknowledge in the paper. The study received no specific funding, and the authors declared no competing interests. These figures describe a supervised research protocol, not a guide for anyone to follow on their own.
What Happened to Pain and Painkiller Use
Pain intensity fell early and stayed lower. Compared with a baseline score of 6.81, the average was 5.15 after the first month, 4.59 after the second and 4.93 after the third. Each of these differences was statistically significant, meaning they were unlikely to be a fluke of chance within this group.
The number of days with pain also went down. According to the results section, the median fell from 30 days per month at baseline to 20 after two months and 17 after three. Readers checking the paper should note that its baseline table reports the painkiller and pain-day measures differently — a median of 30 days on painkillers and 13 days in pain despite them — so the exact starting value for each measure is not fully clear from the published text.
The days on which women needed painkillers dropped significantly, with a measurable change already after the first month. Because some participants were on anti-inflammatory drugs and seven were on opioids, the authors describe the extract as potentially “opioid-sparing” and “NSAID-sparing”. That is an interpretation rather than a separate finding: the paper does not report how many of the seven women taking opioids reduced or stopped them. It is still a relevant question. A Cochrane review of anti-inflammatory painkillers for endometriosis found little solid evidence that they work well for this pain, and long-term opioid use carries well-known risks. Similar opioid-reduction signals have appeared in an Australian study of chronic pain patients prescribed medical cannabis, though in a very different patient group.
One smaller observation deserves attention. Women on hormone therapy and those not on it reported the same pain intensity, but hormone users reported more days of pain per month. The difference did not reach statistical significance and the groups were unequal in size, so the authors treat it as a hint rather than a result.
Beyond Pain: Fatigue, Anxiety and Quality of Life
Endometriosis is increasingly seen as a whole-body condition, and the study measured more than pain. After three months, fatigue scores improved, as did anxiety measured with a standard seven-question screening tool. Scores on the Central Sensitization Inventory — a questionnaire estimating how much the nervous system is amplifying pain — also dropped, and so did the Pain Disability Index, which asks how much pain interferes with work, social life and daily activities.
Interestingly, improvements in central sensitization tracked only moderately with improvements in day-to-day pain scores. In plain terms, the two did not simply move as one, which the authors take as a sign that the extract may be acting on how the nervous system processes pain, not just on how much pain is felt at a given moment. This is a hypothesis, not something this design can prove.
The study’s main goal was defined in advance: a 15-point improvement in the pain section of the Endometriosis Health Profile-30 (EHP-30), a questionnaire where scores run from 0 (best health) to 100 (worst). The pain score fell from 59.34 to 38.38 — a larger drop than the target. Every other section improved too. Feelings of control and powerlessness went from 72.76 to 43.52, emotional well-being from 52.01 to 34.26, self-image from 58.33 to 37.04, social support from 53.94 to 40.51, work-related impairment from 51.67 to 34.81, and sexual relationships from 54.26 to 40.19.
The anxiety result is worth reading next to the Cedars-Sinai data. In Berlin, anxiety scores fell during treatment with a THC and CBD extract; in Los Angeles, women with suspected but unconfirmed endometriosis were more likely to use cannabis for depression. Neither study can show that cannabinoids improve mood, but both point to the same unmet need.
Side Effects: Mostly Mild, But Almost Universal
The extract was described as well tolerated, but that should not be read as side-effect-free. Nearly all participants reported at least one side effect, whatever their dose. Fatigue, concentration and memory problems were the most persistent, affecting around 60% of women throughout the three months.
Other effects faded with time. Anxiety or panic fell from 26% to 4%, and paranoia from 4% to 0%. Nausea was reported by 29.6% at month one, 25.9% at month two and 14.8% at month three. Changes in appetite were common — 44.4%, 55.6% and 40.7% across the three months — though rated as mild. A faster heartbeat was reported by 18.5% in the first two months and 11.1% in the third. One participant reported diarrhea; nobody reported vomiting.
In this study, “mild” means a score of 1 on a 0–3 scale: noticeable, but not distressing or limiting. Body weight was not systematically recorded, which the authors flag as a gap. When side-effect scores were added up rather than averaged, higher doses were moderately linked to a heavier overall burden.
How Much Weight Can These Studies Carry?
The Charité authors are direct about the limits of their pilot, and they are substantial. Twenty-seven women is a very small group. Everyone knew they were receiving a cannabis extract, and there was no placebo arm, so expectation could explain part of the improvement. That matters especially in chronic pain, where placebo responses can be large — and even more so among patients who have exhausted other options and may pin their hopes on a new treatment. Participants were also not randomly drawn from all women with endometriosis, and all outcomes were self-reported through questionnaires.
Three months is short. Longer-term effects of regular THC on memory, hormonal regulation and the body’s own cannabinoid system are not well understood, particularly in women of reproductive age, and long-term THC exposure is known to make cannabinoid receptors less responsive in some people. The authors call for randomized, placebo-controlled trials with more participants and longer follow-up.
The Cedars-Sinai paper does not fill that gap, because it was never designed to test a treatment. What it adds is a warning against looking only at women with a confirmed diagnosis: the patients most likely to be living with depression, and to be using cannabis to cope with it, may be the ones the healthcare system has not yet given a name to their pain.
The studies land in a field where patient interest runs well ahead of clinical proof. A 2024 survey of German-speaking patients by the Charité research group found cannabis was rated the most effective self-management method, with a mean score of 7.6 out of 10, and around 90% of users said they had cut back on conventional painkillers. An international survey covered earlier by The Cannex found similar enthusiasm alongside legal and social barriers. A 2026 scoping review of cannabis use in endometriosis mapped a growing but mostly observational literature. A 2022 systematic review of medical cannabis for gynecologic pain and a 2024 evidence review on endometriosis-associated pain both describe the clinical evidence as limited. The kind of randomized trial the Charité team calls for is what drug developers are now attempting, as with the prescription-grade CBD candidate for endometriosis being developed in the UK.
Where This Fits Legally
The Charité study took place in Germany, where cannabis for medical purposes has been regulated under the Medical Cannabis Act (MedCanG) since 1 April 2024. It is prescribed by a doctor and dispensed only through pharmacies, which is how participants received their extract.
The Cedars-Sinai data come from California, which has had a medical cannabis program since 1996 and legal adult use since 2016. The authors suggest that the recent legalization and the stigma that still surrounds cannabis may mean their figures underestimate real use.
At the federal level in the United States, a federal order placed FDA-approved drug products containing marijuana and marijuana covered by a state medical marijuana license in Schedule III in April 2026. A broader question — whether all remaining marijuana should move from Schedule I to Schedule III — went through a DEA administrative hearing in June and July 2026 and is still awaiting a final decision. No cannabis-based medicine is approved by the FDA for endometriosis.
CBD products bought in shops sit under a separate set of rules, and these are about to change. Section 781 of Public Law 119-37, passed in November 2025, rewrites the federal definition of hemp and limits finished hemp-derived products to 0.4 mg of total THC per container. Many full-spectrum CBD oils contain more than that. A stopgap funding measure approved by Congress in September 2026 delayed the rule for naturally derived hemp products until 11 December 2026, while products with synthetic cannabinoids fall under it from 12 November 2026. Unless Congress acts again, much of what is currently sold as CBD in the US may disappear from legal shelves.
FAQ
There is no solid clinical proof yet. In the Cedars-Sinai cohort, 26.3% of women with confirmed endometriosis said they used CBD for pain, and patient surveys report that many find it helpful, but no randomized trial has shown that CBD on its own relieves endometriosis pain. The Charité pilot used a combined THC and CBD extract, had no placebo group and cannot answer this question for CBD alone.
Yes, the link is well documented. In the Cedars-Sinai study, 52.9% of women with confirmed endometriosis reported anxiety and 42.2% depression that had needed treatment at some point, and among women with suspected but unconfirmed endometriosis both figures were 60.0%. After adjustment, the unconfirmed group had more than twice the odds of depression compared with confirmed cases.
No study has shown that cannabis makes endometriosis lesions grow. But cannabis carries its own risks: the Cedars-Sinai authors point to cognitive decline, psychosis and severe cyclic vomiting in long-term users, especially in people with a personal or family history of psychiatric illness. In the Charité pilot, tiredness, concentration and memory problems affected around 60% of participants. Whether cannabis use worsens or eases depression cannot be told from the new study.
No CBD or cannabis product is FDA-approved for endometriosis. Hemp-derived CBD can be sold, but a new federal definition of hemp limits finished products to 0.4 mg of total THC per container; for naturally derived hemp products it takes effect on 11 December 2026, which may make many full-spectrum CBD oils illegal. Medical marijuana is available through state programs, and since April 2026 state-licensed medical marijuana is in federal Schedule III; qualifying conditions differ from state to state.
Disclaimer
This article is for informational and educational purposes only and does not constitute medical, legal or pharmaceutical advice. It describes the findings of an observational study and a single pilot study and does not recommend the use of cannabis or cannabis-based products for endometriosis, anxiety, depression or any other condition. Anyone considering treatment options should speak with a qualified healthcare professional. Laws governing cannabis and cannabis-based medicines vary by country and region and change over time; readers are responsible for checking the rules that apply where they live.
