Ashwagandha: what it is, what the evidence says and why several European countries have restricted it

If you have walked into a health food shop, a pharmacy or any supplement shop in the last few years, you will have seen ashwagandha. Capsules, powders, gummies: almost always with the same promise on the packaging or in the video that led you to it, calm, less stress, better rest. It is one of the best-selling plants of the moment and also one of the ones giving European regulators the most to talk about, for reasons that have nothing to do with marketing. We are not going to tell you whether to take it or not: we are going to tell you what it is, what the trials have measured, how big those effects really are and why in some EU countries it can no longer be sold in food supplements.

What ashwagandha actually is

Ashwagandha is the common name for Withania somnifera (L.) Dunal, a plant traditionally used in Ayurvedic medicine, of which the root is mainly used. That is how it is described by LiverTox, the database of the National Institute of Diabetes and Digestive and Kidney Diseases in the United States (NIDDK/NIH), one of the most rigorous references available on plant products and their relationship with the liver.

Its biological activity is attributed mainly to withanolides, a group of steroidal lactones which have been described as having anti-inflammatory, neuroprotective and cytotoxic effects, according to the safety assessment published in Frontiers in Nutrition in 2025. That detail matters more than it might seem: withanolides are molecules with a steroidal structure and genuine pharmacological activity. The fact that something comes from a plant does not make it harmless; it makes it, at the very least, something that deserves the same scrutiny as any other active substance.

You will see the word adaptogen everywhere in connection with this plant. It is worth knowing that it is not a recognised pharmacological or regulatory category: it is a term from the herbal medicine literature, not a classification validated by any medicines agency. In the same way, the idea that ashwagandha regulates the hypothalamic-pituitary-adrenal axis is a proposed mechanism, not one demonstrated or validated by the EMA, EFSA or the AEMPS.

What the trials have measured: anxiety, stress and cortisol

This is where there is data, and it is worth looking at it calmly, because the headline and the small print say different things.

A meta-analysis published in Explore in 2024 (PMID 39348746) brought together placebo-controlled trials and found three signals. First: ashwagandha reduces anxiety measured on the Hamilton Anxiety Scale, with a mean difference of −2.19 points (95% CI: −3.83 to −0.55). Second: it reduces serum cortisol, with a mean difference of −2.58 (95% CI: −4.99 to −0.16), a confidence interval whose upper limit almost touches zero, in other words almost compatible with no effect. Third: it reduces perceived stress measured on the PSS scale, with a mean difference of −4.72 points (95% CI: −8.45 to −0.99).

The doses used in the trials included in that meta-analysis ranged from 125 to 600 mg a day. That range matters because of what it implies in reverse: outside it, there is simply no trial data. Four of the included studies reported mild to moderate adverse effects.

A second meta-analysis, published in BJPsych Open with searches up to 8 September 2024, points in the same direction and adds the most honest nuance in the whole literature: it found a statistically significant reduction in cortisol, PSS and HAM-A, but no statistically significant improvement in participants' quality of life. Translated: the scales moved, people's lives did not, at least not measurably. That disconnect between the marker and the outcome that matters is one of the things that should most temper the enthusiasm.

And on cortisol it is worth being especially clear, because it is marketing's favourite argument. Cortisol is a surrogate marker: lowering it slightly on a blood test does not automatically mean being less stressed, or sleeping better, or losing weight. The chain of lower cortisol equals weight loss has not been demonstrated, and anyone selling it to you as though it had is selling you a leap of logic, not a result.

What cannot be claimed (even if the packaging hints at it)

This is probably the most useful part of the article, because it is the part that appears in no advert.

• Sleep and insomnia: there is no verified figure we can give you on sleep quality. If it is promised to you, it is not backed by the data we have been able to check.
• Testosterone, male infertility, strength, muscle mass or sports performance: zero verified figures. It is one of the most heavily promoted uses and one of the worst supported by checkable evidence.
• Cognition, memory or concentration: again, there is no verified data we can cite.
• Thyroid: not only is there no demonstrated benefit, but the risk assessments carried out by the British Food Standards Agency identified signals of thyroid toxicity. Saying it is good for the thyroid is exactly the opposite of what the safety data suggest.
• Authorised health claims: there are none for ashwagandha in the EU. Botanical claims have been on hold on the EFSA list for years, so phrases such as reduces stress or balances cortisol on labelling or in advertising are not authorised.

The safety signal: the liver, and why it has become a regulatory matter

LiverTox holds a nuanced position that is worth reproducing exactly as it stands: ashwagandha is considered generally safe and in clinical trials no rises in serum enzymes or serious hepatotoxicity were reported, but cases of clinically apparent liver injury have recently been described. Both halves of that sentence are true at the same time, and that is precisely the problem: trials are short, with limited sample sizes and selected participants; idiosyncratic liver reactions are rare and appear once the product reaches millions of people.

There are published, peer-reviewed cases of acute liver injury induced by ashwagandha. In one of them causality was formally assessed using RUCAM, the standardised scale used precisely to attribute liver damage to a specific product. In another, the person developed jaundice after roughly a year of use. These are case reports, not population incidence: they do not tell us how many people are affected, but they do show that the phenomenon exists and is documented.

The agencies have been taking note. The risk assessments by the United Kingdom's Food Standards Agency found associations with thyroid toxicity, hypoglycaemia and potential liver toxicity, which is why the FSA referred the matter to the Committee on Toxicity (COT) for a more in-depth assessment.

The European map: banned in one country, advised against in another, under assessment in several

This is the point that most surprises anyone who buys ashwagandha with the peace of mind of thinking they are buying a herbal tea.

• Denmark banned the use of ashwagandha in foods and food supplements in 2023, following a risk assessment by the Technical University of Denmark (DTU) published in 2020. The key Danish conclusion was that no safe lower intake limit could be established.
• The Netherlands: the RIVM advises against the use of food supplements containing Withania somnifera, with a particularly forceful argument, namely that the potential effects are serious and can occur when the product is used according to the labelling on the packaging. In other words, they are not talking about misuse, they are talking about normal use. The Spanish General Council of Official Colleges of Pharmacists has echoed this warning.
• Sweden: in September 2022 the Swedish Food Agency stated that the Danish assessment is also applicable in Sweden.
• France: in 2021 ANSES launched an assessment of the risks of using ashwagandha in food supplements.

The full picture, set out in the regulatory review published in Phytotherapy Research in 2025, is that of a plant whose European assessment remains open and which has already led to restrictive decisions in at least one member state.

And in Spain, where does it stand?

In Spain, ashwagandha is sold as a food supplement, not as a medicine. That distinction is not bureaucratic, it explains almost everything. Food supplements are foods: they do not go through the authorisation process required of a medicine, they do not have to demonstrate efficacy against placebo to reach the market and, legally, they cannot be credited with properties for preventing, treating or curing disease. A supplement does not treat anxiety, and if a package or an advert says it does, it is breaking the rules.

AESAN manages food supplement alerts through the SCIRI system and the national official control plan, and suspected adverse reactions are reported via AESAN or the health authority of your autonomous community. On other points we prefer not to comment, because we have not been able to verify them: we do not claim that ashwagandha is expressly permitted or expressly prohibited under Spanish supplement regulations, nor that an authorised medicine containing Withania somnifera does or does not exist, nor that there is a European monograph. The honest wording is the one you have already read: in Spain it is sold as a food supplement, several EU countries have banned it or advised against it, and the European assessment remains open.

A note on brands: the trade names of standardised extracts you will see on packaging are not validated by any regulator. They are product brands, not seals of health approval.

How this fits in if what you are dealing with is stress, sleep or weight

Many people come to ashwagandha not out of botanical interest, but because they sleep badly, live with sustained stress or want to lose weight and have read that cortisol is to blame. It is worth separating the problems out, because each one has its own path and its own tools.

If sleep is the main problem and there is snoring, breathing pauses or marked daytime sleepiness, the relevant diagnosis may lie elsewhere: here we explain what obstructive sleep apnoea is and how it is managed, a treatable and very underdiagnosed cause of poor rest. And if what worries you is your weight along with your blood pressure, waist circumference or blood sugar, it is worth first understanding what metabolic syndrome is and why it appears and how prediabetes is managed, because there you do have diagnostic criteria and strategies with solid evidence behind them.

And if the underlying question is what really works for weight loss beyond supplements, we have explained in detail what GLP-1 drugs are and how they work in obesity: these are prescription medicines, with their indications, their adverse effects and their need for medical supervision, and precisely for that reason they are in a different league from a food supplement.

In summary, and what to do if the underlying problem is your weight

Ashwagandha has a small, measurable effect on anxiety and stress scales, an effect on cortisol that borders on statistical irrelevance, no demonstrated improvement in quality of life and a liver and thyroid safety signal that has already led one European country to ban it and others to review it. It is neither a poison nor the solution: it is a product with more promise than proof.

If what brought you here is your weight and the suspicion that something metabolic is going wrong, there is a medical route available in Spain today: medical weight loss treatment, with a prior assessment, a prescription where appropriate from a registered doctor and ongoing follow-up, which is what we do at AiON. You can start by checking whether you are eligible in a few minutes.