One idea explains why the same people so often have migraine, fibromyalgia and irritable bowel syndrome together, why all three resist treatment, and why all three were dismissed for decades as psychosomatic. It is an elegant hypothesis with some real supporting data. It is also not an accepted medical diagnosis and cannot currently be tested for in a patient. Its own author openly lists major gaps in the evidence.
This article describes a hypothesis, not an established condition. “Clinical endocannabinoid deficiency” is not a diagnosis you can be given. It is not in any diagnostic manual, and it cannot be measured in a living patient. Anyone offering to test you for it, or to treat it, is well ahead of the evidence. Nothing here is medical advice.
1. The Idea
The concept was proposed by Ethan Russo and set out most fully in Russo EB, “Clinical Endocannabinoid Deficiency Reconsidered: Current Research Supports the Theory in Migraine, Fibromyalgia, Irritable Bowel, and Other Treatment-Resistant Syndromes,” Cannabis and Cannabinoid Research, 2016; 1(1): 154–165.
The proposition is that some disorders arise from deficient endocannabinoid tone. That could mean too little of the body’s own cannabinoids, anandamide (AEA) and 2-arachidonoylglycerol (2-AG), too little production of them, or reduced cannabinoid receptor density. Where that deficiency exists, Russo argues, you would expect a lowered pain threshold together with derangements of digestion, mood and sleep. (For the basics of the endocannabinoid system itself, see topic 1306.)
That prediction is what makes the hypothesis interesting, because it describes a real and recognisable clinical picture.
2. Why It Fits Three Conditions in Particular
Russo focuses on migraine, fibromyalgia and irritable bowel syndrome, because they share an unusual set of features:
- All three involve hypersensitivity: hyperalgesia and photophobia in migraine, widespread pain and central sensitisation in fibromyalgia (tender-point counts are no longer part of the diagnostic criteria), and visceral hypersensitivity in IBS.
- All three are diagnosed by subjective criteria rather than by a test.
- All three show high rates of comorbidity with each other.
- All three were, for decades, dismissed as psychosomatic.
A single underlying mechanism that produced all four of those characteristics would be a genuinely useful piece of medicine. That is the appeal, and it is legitimate.
3. The Evidence He Presents
| Finding | What it showed (as reported in Russo 2016) |
|---|---|
| Cerebrospinal fluid in migraine | Chronic migraineurs showed statistically significant reductions in CSF anandamide versus controls (p < 0.0001). Russo describes this as perhaps the strongest evidence for the theory |
| Platelet endocannabinoids in migraine | Profoundly reduced 2-AG and AEA in both episodic and chronic migraine patients (p < 0.0001) |
| Colonic biopsies in IBS | TRPV1-immunoreactive nerve fibres approximately 3.5-fold higher in IBS patients, consistent with visceral hypersensitivity |
| Genetics | CB1 receptor gene (CNR1) variants associated with migraine, haplotypic significance p = 0.008 |
This is not nothing. Measured deficits in a specific signalling system, in a specific patient population, with a genetic association: that is the shape of a real biological hypothesis rather than a slogan. (The migraine evidence is covered in more detail in article bai97.)
4. What Russo Himself Says Is Missing
The most persuasive thing about the 2016 paper is how openly it states its own gaps. Those statements should travel with the theory wherever it is quoted. As of 2016, Russo wrote that:
- There had been little direct objective evidence of the theory in patients “until quite recently”.
- Endocannabinoid measurements in fibromyalgia and IBS had not been systematically examined. The strongest data came from migraine, and the other two conditions were largely an extension by analogy.
- Anatomic and physiological scanning techniques were not yet capable of real-time direct assessment of endocannabinoid levels in living patients.
- Randomised controlled trials in these conditions were justified by the data but scarce.
- Contradictory findings were to be expected, particularly given the reciprocal relationships between AEA and 2-AG signalling.
Since 2016 (status as of October 2026). The most direct later test in fibromyalgia did not support the simple version of the hypothesis. A systematic review and meta-analysis (Kurlyandchik, Lauche, Tiralongo, Warne and Schloss, Pain Reports, 2022) found no significant difference in plasma anandamide or 2-AG between people with fibromyalgia and healthy controls. Only some related lipids (OEA and SEA) were higher in patients, and the authors called the evidence inconclusive. For IBS, the patient measurements remain small pilot studies. One of these (2013) reported different patterns by IBS subtype, such as higher 2-AG in diarrhoea-predominant IBS, rather than a uniform deficit. Cannabis RCTs in fibromyalgia remain few and short (see Section 6). None of them can select patients by endocannabinoid deficiency, because no validated test exists. As of October 2026, the case for the theory still rests mainly on the 2016 review and the migraine data it summarises.
Read together, these points make the position clear. This is a hypothesis that deserves further research. It is not a mechanism that explains the patient in front of you.
5. The Circular Argument to Watch For
One specific reasoning error attaches itself to this theory. It is worth naming because it appears constantly in online discussion.
“Cannabis helps my condition, therefore I must have an endocannabinoid deficiency, therefore cannabis is correcting a deficiency rather than merely masking symptoms.”
Every step is unsupported. A treatment working does not establish what was wrong: aspirin relieving a headache does not mean the headache was caused by aspirin deficiency. The third step also smuggles in a claim of real consequence. “Correcting a deficiency” implies a treatment that addresses a cause, while “relieving symptoms” implies something else entirely. Nothing in the evidence distinguishes between them.
The theory also cannot currently be falsified in an individual, because there is no test. A hypothesis that cannot be checked against the patient it is being applied to is not functioning as a diagnosis, whatever it is called.
6. What the Fibromyalgia Trials Actually Show
The best-known recent review, Strand et al., “Cannabis for the Treatment of Fibromyalgia: A Systematic Review,” Biomedicines, 2023, covered 9 studies with 564 patients, of which only 4 were RCTs. It rated the evidence for short-term pain reduction as low quality, and one of the four RCTs found no difference from placebo. The review authors concluded that cannabis was generally well tolerated in short-term studies; long-term safety is not established. These are data on short-term symptom relief, not on correcting a deficiency. The full chronic-pain evidence is covered in article bai167.
7. Why the Theory Still Earns Its Place
Rejecting a hypothesis for being unproven is as much an error as accepting it for being appealing. This idea has done three things well:
It gave a biological frame to conditions that were treated as character flaws. Patients with fibromyalgia, migraine and IBS spent decades being told the problem was psychological. A plausible mechanism changes how they are treated in a consulting room, and that is not a small thing.
It generates testable predictions. Measure endocannabinoid levels in these patients, look for CNR1 variants, run the trials. Those are real experiments. Some have now been done, and as Section 4 shows, the results have not all been kind to the theory.
It points somewhere specific. If deficient endocannabinoid tone is the problem, the therapeutic target may be the body’s own system, for example enzymes such as FAAH that break endocannabinoids down, rather than plant cannabinoids taken from outside the body. This is one research direction, but FAAH inhibitor trials have so far failed in pain. The inhibitor PF-04457845 did not relieve osteoarthritis knee pain in a placebo-controlled trial, and one 2016 trial (BIA 10-2474) caused serious harm, including one death.
8. What This Means for a Patient
If you have migraine, fibromyalgia or IBS and this theory has been offered to you as an explanation, the useful way to see it is: an interesting idea about why your condition might exist, not a diagnosis and not a treatment plan.
The practical questions stay the same. Has the condition been properly assessed? Which treatments with established evidence have been tried? If a cannabinoid is being considered, is it a supervised trial with a defined endpoint and a stopping rule? Is it judged on symptoms, which is what the evidence supports, rather than on a deficiency nobody can measure?
In Thailand, cannabis for medical use requires a prescription or medical certificate from an authorised practitioner under the Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis) B.E. 2568 (supply limited to about 30 days of use); the Cannabis Control Act remains a draft. For the current prescription rules, see article bai247 and topic 1358.
If you live with two or three of migraine, fibromyalgia and IBS together (common enough that the overlap is part of what prompted this theory), how were they treated? Separately by different specialists, or as one problem? That question has practical consequences, and the answer varies enormously by health system.
This is general educational information, not medical advice. Clinical endocannabinoid deficiency is a research hypothesis, not a recognised diagnosis.