Ashwagandha: what it is and what the research says

Ashwagandha (Withania somnifera) is a small evergreen shrub from the nightshade family, the same family as tomatoes and potatoes. It grows wild across India, the Middle East and parts of Africa, and most of the supply for supplements is farmed in India. The root is the part used in most human research, and it is the part most commercial extracts are made from.

The name is Sanskrit. Ashwa means horse and gandha means smell, because the fresh root has a strong, horsey scent. The Latin species name, somnifera, means sleep bringing. In English it also goes by Indian winter cherry, after the small orange red berry that sits inside a papery husk.

You will sometimes see it sold as Indian ginseng. That name is a marketing habit rather than botany. Ashwagandha is not related to the ginseng family at all, and the two plants contain different compounds. Our Panax Ginseng guide covers the real thing.

Ashwagandha is usually filed under adaptogens, a loose label with no legal or medical definition for plants said to raise resistance to stress in a general way. Ginseng, rhodiola and mushrooms such as Reishi get the same label.

Ashwagandha has been part of Ayurveda, the traditional medicine system of India, for centuries, and it also appears in Unani medicine. Classical Ayurvedic texts class it as a rasayana. That is a tonic taken over long periods for strength and vitality, not a remedy for one complaint.

The traditional preparation is simple. The root is dried, ground to a fine powder and stirred into warm milk, sometimes with honey or ghee, and drunk in the evening. That evening habit fits the somnifera name.

Today you will find root extract capsules and tablets (the form used in most clinical trials), loose root powder, liquid tinctures, teas, drinks and gummies. In the UK it is sold in Boots, Holland & Barrett, most large supermarkets and by hundreds of online brands, and most buyers are after something for everyday stress or sleep. Traditional use describes how people have taken a plant for generations. It is not proof that it works.

The compounds researchers talk about most are withanolides, a group of natural steroidal lactones found in the root and leaf. The plant also contains alkaloids and other compounds. In a lab study published by Dutta and colleagues at Oregon Health & Science University in 2025, the effects seen in fruit flies and mouse nerve cells came from an alkaloid rich part of the root extract, while the withanolide rich part showed none. Read on PubMed So withanolide content is a useful quality marker rather than the whole story.

When a label says an extract is standardised, it means each batch is tested to hit a set level of withanolides, for example 5%. Labs use different methods to measure withanolides, though, so two products that both say 5% are not always identical.

Root and leaf are chemically different. Leaves carry more of a compound called withaferin A. Most of the research on stress and sleep used root extract, and a few trials used root and leaf blends.

Many trials used branded extracts, and you will see their names on UK labels:

• KSM-66, a root only extract, usually tested at 300mg twice a day
• Shoden, a root and leaf extract with a high withanolide content, tested at 120mg to 240mg a day
• Sensoril, Witholytin and others, each with its own dose and standardisation

Because these extracts differ, a dose from one trial does not transfer neatly to another product. An extract ratio such as 10:1 tells you roughly how much dried root went into each gram of extract, in this case about ten grams. Kalma, our ashwagandha gummy, uses a 10:1 root extract standardised to 5% withanolides.

Stress is where most of the human research sits. The trials usually recruit adults who describe themselves as stressed, give them an extract or a placebo for six to twelve weeks, then compare scores on standard stress and anxiety questionnaires. Some also measure cortisol, a hormone that rises as part of the body's stress response and normally peaks in the morning.

• Chandrasekhar and colleagues, 2012. 64 adults in India with a history of chronic stress took 300mg of root extract twice a day or a placebo for 60 days. The ashwagandha group scored lower on every stress scale used, and their serum cortisol fell by 27.9%. Read on PubMed
• Salve and colleagues, 2019. 60 stressed adults took 250mg a day, 600mg a day or a placebo for eight weeks. Both ashwagandha groups had lower perceived stress scores and lower cortisol than placebo. The changes were larger in the 600mg group. Read on PubMed
• Lopresti and colleagues, 2019. 60 stressed adults took 240mg of Shoden or a placebo for 60 days. Anxiety scores and morning cortisol fell more in the ashwagandha group. A broader stress questionnaire did not quite reach statistical significance. Read on PubMed
• Smith and colleagues, 2023. 120 overweight adults aged 40 to 75 took 200mg of Witholytin twice a day for 12 weeks. Stress scores fell in both groups, with no real difference between ashwagandha and placebo. Fatigue scores were lower with ashwagandha. Read on PubMed

Pooled together, the trials lean positive but cautious. A 2022 meta-analysis by Akhgarjand and colleagues combined 12 trials with 1,002 participants and found lower anxiety and stress scores with ashwagandha than placebo, with the stress effect clearest at 300mg to 600mg a day. The authors rated the certainty of the evidence as low, because the trials varied so much. Read on PubMed An earlier 2021 review cited by the US National Institutes of Health found seven trials with 491 adults, all in India.

In 2022 an international taskforce of psychiatrists (WFSBP and CANMAT) gave ashwagandha root extract a provisional recommendation for generalised anxiety disorder, a diagnosed condition that needs a doctor. They said more data was needed before anything stronger. Read on PubMed

Sleep is the second big area of research. There are fewer trials than for stress, but several used wrist worn sleep monitors as well as questionnaires, which gives firmer data than self report alone.

• Langade and colleagues, 2019. 60 adults with insomnia took 300mg of root extract twice a day or a placebo for ten weeks. By the end, the ashwagandha group took about 29 minutes to fall asleep against about 34 minutes on placebo. Their sleep efficiency, the share of time in bed actually spent asleep, rose from about 76% to 83%, compared with 75% to 80% on placebo. Anxiety scores fell too. Read on PubMed
• Deshpande and colleagues, 2020. 150 healthy adults who woke feeling unrefreshed took 120mg of Shoden or a placebo for six weeks. Self rated sleep quality rose by 72% in the ashwagandha group and 29% in the placebo group. Read on PubMed
• Langade and colleagues, 2021. 80 adults, half with insomnia, took 600mg a day or a placebo for eight weeks. People with insomnia saw the clearest changes. Those without insomnia also rated their sleep higher, but their anxiety scores and alertness on waking did not change. Read on PubMed

The best single summary is a 2021 systematic review and meta-analysis by Cheah and colleagues in PLOS One. It pooled five trials with 400 adults and found a small but significant effect on overall sleep. The effect was larger in people with diagnosed insomnia, at 600mg a day or more, and in trials of eight weeks or longer. People also scored higher on alertness when they got up, and there was no effect on quality of life. Read on PubMed

Ashwagandha is not a sedative in the way a sleeping pill is, and none of these trials suggest it replaces the basics: a regular bedtime, daylight in the morning, less caffeine after lunch and a screen free last half hour.

Ashwagandha marketing often goes well beyond stress and sleep. The research here is thinner, usually one or two small trials per topic.

• Memory. In a 2017 pilot trial by Choudhary and colleagues, 50 adults with mild cognitive impairment took 300mg twice a day or a placebo for eight weeks. The ashwagandha group scored higher on several memory and attention tests. Read on PubMed For a mushroom with a similar research profile, see our Lion's Mane guide.
• Strength training. In a 2015 trial by Wankhede and colleagues, 57 young men new to weight training took 300mg twice a day or a placebo for eight weeks. The ashwagandha group added 46kg to their bench press on average against 26kg on placebo. Read on PubMed
• Male hormones and fertility. A 2019 crossover trial in 57 overweight men aged 40 to 70 found testosterone rose 14.7% more with ashwagandha than placebo over eight weeks, though fatigue and vigour scores did not differ. Read on PubMed A 2013 pilot in 46 men with low sperm counts reported higher counts after 90 days on 675mg a day. Read on PubMed
• Perimenopause. In a 2021 trial by Gopal and colleagues, 100 women with perimenopausal symptoms took 300mg twice a day or a placebo for eight weeks. Scores on a menopause rating scale fell further with ashwagandha. Read on PubMed

Harvard Health lists blood sugar, inflammation, joint health, constipation and sexual health as areas where the evidence is extremely limited. Immunity claims fall in the same bucket. If a product leads with any of these, treat it with suspicion.

If you skipped straight here from the top, this is the section to read.

What it does show. Across dozens of placebo controlled trials, adults taking ashwagandha root extract for six to twelve weeks have tended to report lower stress and anxiety scores than people on placebo. Several trials also measured lower cortisol. Sleep trials show a small effect overall, larger in people with insomnia. Short term use was well tolerated in these studies.

What it does not show.

• Most trials are small, often 50 to 150 people, and short. Evidence on use over many months or years is missing.
• Most were run in India, many with extracts supplied by the companies that make them, and a lot of the published work comes from a handful of research groups.
• Outcomes are mostly questionnaires. People who expect to feel calmer often report feeling calmer, which is why placebo groups report gains too.
• Extracts, doses and withanolide levels vary so much that the reviews rate their own certainty as low.
• Some well run trials, such as the 2023 Witholytin study, found no difference on stress.
• No trial we know of has tested a gummy. The research used capsules and tablets.

The US National Institutes of Health fact sheet makes a further point. Much of the research was carried out within a traditional medicine setting, so how ashwagandha behaves as a stand alone supplement bought off a shelf is less clear. You can see every study we reference, ingredient by ingredient, on our research page.

There is no official recommended dose of ashwagandha in the UK, and no safe upper level has been set. What we can tell you is what the trials used.

• Most stress and sleep trials used 240mg to 600mg of root extract a day, most often 300mg twice a day. Harvard Health puts the full studied range at 125mg to 600mg.
• Extracts with a high withanolide content were tested at lower amounts, such as 120mg to 240mg of Shoden.
• Trials using raw root powder used several grams a day, because powder is far less concentrated than extract.
• The 2022 psychiatry taskforce mentioned above referred to 300mg to 600mg of root extract standardised to 5% withanolides.

Higher doses have been studied much less, and one of the thyroid case reports involved 1,950mg a day. Follow the label and do not take two ashwagandha products at once, which is easy to do by accident when it turns up in sleep blends and greens powders.

No trial has compared morning with evening dosing head to head, so timing comes down to how you get on with it.

• Many sleep trials gave the dose in the evening or split it across the day.
• If it makes you feel drowsy, move it to the evening and take care driving until you know how it affects you.
• If it upsets your stomach, take it with food. Breakfast is an easy anchor for most people.
• The main thing is consistency. Take it at the same time every day so you do not miss doses.

Expect to wait. The trials ran for six to twelve weeks and the sleep meta-analysis found larger effects at eight weeks or more, so give it two full months and keep a weekly note of sleep, stress and any side effects.

Ashwagandha appears well tolerated for up to about three months. A few small studies ran to twelve months, but long term safety is not well mapped, and nobody has tested whether taking breaks makes a difference.

Most people in trials had no side effects or mild ones. The common ones are:

• Stomach upset, loose stools or nausea
• Drowsiness
• Headache, less often
• For a minority, feeling more wired or anxious rather than calmer

Liver. Rare cases of liver injury have been linked to ashwagandha supplements. A 2020 case series by Björnsson and colleagues described five people in Iceland and the US who developed jaundice two to twelve weeks after starting. None developed liver failure, and liver tests returned to normal within one to five months in the four who were followed up. Read on PubMed A 2021 report from the Freeman Hospital in Newcastle described a 39 year old woman with jaundice and nausea after taking an over the counter ashwagandha product. Read on PubMed These reactions look unpredictable rather than dose related. Stop taking it and see your GP promptly if you notice yellowing of the skin or eyes, dark urine, itching all over, or unusual tiredness, nausea or stomach pain.

Thyroid. In a 2018 trial of 50 adults with an underactive thyroid that did not yet need treatment, 600mg a day for eight weeks lowered TSH and raised thyroid hormones compared with placebo. Read on PubMed There are also case reports of an overactive thyroid in people taking ashwagandha. So it may interact with thyroid medication.

Speak to your GP or pharmacist before taking ashwagandha if you:

• Are pregnant, trying to conceive or breastfeeding (we advise not taking it at all)
• Are under 18
• Have a thyroid condition or take levothyroxine or other thyroid medication
• Have liver or kidney disease
• Have an autoimmune condition such as rheumatoid arthritis, lupus or multiple sclerosis, or take immunosuppressants
• Take medicines for diabetes or blood pressure, as ashwagandha may add to their effect
• Take sedatives, sleeping tablets, anti anxiety medicines, antidepressants or anti seizure medicines
• Have surgery coming up. Tell the surgical team, as it may add to the effect of sedatives and anaesthetics

Ashwagandha is legal in the UK and is sold as a food supplement. It is also under official review, and the position may change.

The Food Standards Agency. From 8 July to 1 September 2024 the FSA ran a call for evidence on ashwagandha in food supplements. It said its risk assessments had linked ashwagandha with effects on thyroid hormone levels, low blood sugar and possible liver toxicity, and that there were no safe levels or set limits for its use in the UK. It asked the Committee on Toxicity (COT), its independent scientific advisers, to review whether a safe level of intake can be set. The FSA published a summary of the 19 responses on 25 June 2026. The COT has worked through several drafts of its statement and the item was still on its agenda on 8 September 2026. What it will conclude, and whether the UK will set a limit, is an open question. On 26 September 2026 there is no UK ban and no formal consumer warning. We follow the FSA's guidance and will update this page when the statement is out.

The NHS. The NHS does not recommend ashwagandha and has no page on it. Its general guidance on herbal medicines says not to take them if you are pregnant or breastfeeding, have kidney or liver problems, or take certain medicines, naming statins, antidepressants, metformin, ramipril, warfarin and the combined pill.

Elsewhere in Europe. Denmark banned ashwagandha in 2023 after a risk assessment raised concerns about pregnancy and hormones. In 2024 the French food safety agency, ANSES, advised against it for groups including pregnant and breastfeeding women and people with hormone related conditions. The American Herbal Pharmacopoeia, whose monograph is often cited for the pregnancy concern, has said there is no evidence that the root causes miscarriage. Regulators are still cautious, and so are we.

Claims. There are no authorised health claims for ashwagandha in Great Britain. Claims for plant ingredients have been on hold for years, so no brand can legally make a stress or sleep claim for it. Only vitamins and minerals such as vitamin B6 carry authorised claims, which is why this page sticks to what the studies found.

If you decide to try ashwagandha, a few checks on the label make a big difference:

• Root or leaf. Most human research used root extract.
• Withanolide content. A stated percentage, such as 5%, means the extract is standardised.
• Amount per serving. Compare it with the 240mg to 600mg range used in most trials.
• Everything else in the pot. Blends can hide small amounts, and sugar content varies in gummies.
• Testing. Harvard Health suggests looking for independent testing. Athletes in the UK often look for Informed Sport certification.

Capsules make it easier to reach trial style amounts and usually contain no sugar. Gummies are easier to take every day but tend to carry less per serving, some sugar and, often, gelatine.

Kalma is our ashwagandha gummy. Every two gummies contain a 10:1 ashwagandha root extract standardised to 5% withanolides, plus 2.8mg of vitamin B6, which is 200% of the NRV. Vitamin B6 contributes to normal psychological function, to the normal functioning of the nervous system and to the reduction of tiredness and fatigue. Those are authorised health claims for B6, not for ashwagandha. Kalma is apple flavoured, set with fruit pectin, vegan, and a pack of 60 lasts a month at £24.99. The studies above did not test Kalma, and it is not for use during pregnancy or breastfeeding or for under 18s.

If you are also curious about mushrooms, our mushroom gummies are covered ingredient by ingredient in guides such as Cordyceps and Chaga.