Ashwagandha
Ayurveda tradition
A musky Ayurvedic root that's been used for centuries as a rejuvenating tonic - and has since become the most heavily trialled adaptogen on the shelf. Here's the honest version: what tradition says, what the human studies actually found (and their real limitations), the liver and thyroid cautions nobody puts on the front of the label, what Health Canada lets us claim, and how to buy it well. No hype, no BS.
In one minute
An Ayurvedic root, used for centuries as a rasayana (rejuvenating tonic) - and now one of the most heavily trialled herbs on the shelf.
The strongest human evidence is for perceived stress, sleep quality and lowered cortisol - most of it from small trials of 6-12 weeks.
Nearly all the good trials used a branded, standardised extract (KSM-66, Sensoril, Shoden), not generic root powder. That matters when you buy.
Real safety signals exist: rare but documented liver injury, and measurable effects on thyroid hormones.
Not for pregnancy, and a genuine 'ask your doctor first' if you have liver disease, a thyroid condition, an autoimmune condition, or take sedatives.
Health Canada permits traditional Ayurvedic uses plus an adaptogen claim - nothing more.
At a glance
History & natural history
Ashwagandha has been used in India for a very long time - but the story of how it became a global bestseller is much more recent, and worth separating from the folklore.
The plant itself
Withania somnifera is a small, hardy evergreen shrub with dull green leaves, greenish-yellow flowers and bright red-orange berries in a papery husk - which is why one of its English names is winter cherry. It grows in dry, poor soil across India, the Middle East, parts of Africa and southern Europe. Botanically it sits in the Solanaceae, the nightshade family, alongside tomatoes, potatoes, peppers, tobacco and deadly nightshade. The part used medicinally is almost always the root: thick, pale, and carrying a distinctly musky smell.
The name
Ashwa-gandha is Sanskrit for 'smell of horse'. The usual explanation is a double one - the root genuinely does smell faintly equine, and traditional texts said it conferred the strength and vigour of a stallion. The Latin species name adds a second clue: somnifera means 'sleep-bringing', which tells you what European botanists understood it to do.
Its place in Ayurveda
In Ayurveda, ashwagandha is classified as a rasayana - a rejuvenating tonic. Rasayana medicines are not aimed at a single complaint; they are given to rebuild strength and vitality, especially in old age or during recovery from illness. It's also described as a nervine that balances aggravated Vata, the elemental principle associated with movement and, when disturbed, with restlessness and anxiety. It appears in the Ayurvedic Pharmacopoeia of India and in classical materia medica going back centuries. You'll sometimes see it called 'Indian ginseng' - a Western marketing coinage. It is not related to ginseng in any way.
How it's grown and made
Most commercial ashwagandha is cultivated rather than wild-harvested, largely in Madhya Pradesh and Rajasthan in India. The roots are lifted, washed, dried and either milled into powder or extracted with water or ethanol. That extraction step is where modern products diverge sharply from the traditional preparation: a concentrated standardised extract may deliver in a 300 mg capsule what would take several grams of raw powder, and root-only extracts differ chemically from those made with root and leaf together.
The modern turn
Ashwagandha's Western moment came from research, not tradition. Beginning in the 2000s and accelerating through the 2010s, a run of randomised placebo-controlled trials - most conducted in India, many using newly developed branded extracts - reported reductions in perceived stress, lower cortisol and better sleep. That evidence, combined with the arrival of the word 'adaptogen' in wellness marketing, turned a classical Ayurvedic root into one of the best-selling supplements in North America. The same period brought the first published reports of ashwagandha-associated liver injury, in 2017, and a growing regulatory conversation that continues today. Both halves of that story belong on the same page.
The science, graded honestly
We grade the evidence for every use so you can see the difference between "proven," "promising," and "traditional."
## The short version
Ashwagandha doesn't stimulate you and it doesn't sedate you in the way a sleeping pill does. The best current explanation is that it takes the edge off the body's stress response - the chain of signals that runs from the brain to the adrenal glands and ends in cortisol, your main stress hormone.
## The cortisol story
Across several trials, people taking ashwagandha ended up with lower blood cortisol than people taking a dummy capsule. In the most-quoted example, a 60-day trial in 64 chronically stressed adults, average serum cortisol fell about 28 percent in the ashwagandha group versus about 8 percent on placebo. Researchers describe this as a dampening of an over-active stress axis rather than a shutdown of it - which is the whole idea behind the word 'adaptogen'.
## The calming side
Laboratory work suggests some of ashwagandha's compounds interact with GABA signalling - the brain's main 'slow down' system, and the same broad target as many calming medicines. This is a plausible mechanism, not a settled one, and it's studied mostly in cells and animals. It is, though, the most likely reason ashwagandha can make some people feel drowsy, and the reason Health Canada requires a caution about driving and operating machinery.
## The compounds
The signature molecules are the withanolides - a family of steroid-like plant compounds, of which withaferin A and withanone are the best known. Different extracts pull different amounts and ratios of these out of the plant, and root-only extracts differ chemically from root-and-leaf extracts. That is not a technicality: it is why two ashwagandha products can behave differently, and why the research is hard to summarise in one sentence.
## Why it's slow
Nothing here is an acute effect. Trials that found benefit generally ran 6 to 12 weeks, and the sleep meta-analysis found effects were clearer at 8 weeks or more. Consistency matters more than dose size.
Stress & cortisol
Reasonably supportedThis is the reason most people buy ashwagandha, and it's where the human evidence is genuinely strongest - so it's worth walking through what the studies actually did.
The famous one is Chandrasekhar and colleagues, published in the Indian Journal of Psychological Medicine in 2012. The researchers recruited 64 adults with a history of chronic stress and split them into two groups. One group took 300 mg of a concentrated full-spectrum root extract twice a day; the other took identical dummy capsules. Neither the participants nor the researchers handing out the capsules knew who got which - that's what 'double-blind' means, and it's how you stop hope from doing the work. After 60 days, the ashwagandha group scored significantly lower on every stress questionnaire used. More interestingly, they also measured a physical marker: serum cortisol, the body's main stress hormone. It fell about 28 percent in the ashwagandha group compared with about 8 percent on placebo. A subjective feeling and an objective blood measurement moving together is a meaningfully stronger result than either alone.
It is not one study standing alone, which is unusual for a botanical. A 2021 systematic review pulled together seven randomised trials covering 491 adults, using doses from 240 to 1,250 mg a day of extract over 6 to 8 weeks. Across them, ashwagandha significantly reduced stress and anxiety on validated rating scales, reduced sleeplessness and fatigue, and lowered serum cortisol versus placebo - with benefit appearing clearer at 500 to 600 mg a day than at lower doses. Trials published since have mostly pointed the same way, including a 90-day study of 130 stressed adults in India using a sustained-release extract, and a 30-day trial of 60 adults in Florida using a root-and-leaf extract, where the lower 225 mg dose also produced lower salivary cortisol.
The honest limitations. Every participant in that seven-trial review was recruited in India; so was everyone in most of what followed. The trials are small and short - two months is typical, and none tell you anything about a year of use. Many were conducted with a branded extract supplied by its manufacturer, and industry-funded supplement trials tend on average to report rosier results. And the effects, while statistically real, are modest: this is 'the week felt more manageable', not 'my stress disappeared'. One 2023 Australian trial in 120 tired, overweight adults found no reduction in perceived stress at all - it improved fatigue instead. Consistent direction, unimpressive size, imperfect evidence.
Chandrasekhar, Kapoor & Anishetty 2012, Indian J Psychol Med ↗Sleep quality
Reasonably supportedThe botanical name is a hint: somnifera is Latin for 'sleep-bringing'. The modern evidence is smaller than the stress evidence but points the same way, with one important nuance - it helps people who sleep badly far more than people who already sleep well.
The clearest single trial is Deshpande and colleagues, published in Sleep Medicine in 2020. They enrolled 150 adults aged 18 to 65 in India who reported poor, unrefreshing sleep, and randomised them to a root-and-leaf extract (Shoden, 120 mg a day standardised to withanolide glycosides) or placebo for six weeks. What makes this study better than most is that it didn't rely only on people's opinions: participants wore wrist actigraphy monitors, which measure movement and estimate sleep objectively. Self-reported sleep quality improved in both groups - placebo included, which tells you something about expectation - but improved considerably more on ashwagandha (72 percent versus 29 percent). The actigraphy backed it up, showing better sleep efficiency, longer total sleep, shorter time to fall asleep and less waking during the night.
A second trial by Langade and colleagues took 80 adults, half with diagnosed insomnia and half without, and gave them 600 mg a day of a root extract (KSM-66) or placebo for eight weeks. The people with insomnia improved on sleep quality, time to fall asleep, morning alertness and anxiety. The people without insomnia reported slightly better sleep but no change in anxiety or morning alertness. If your sleep is already fine, there may not be much room to improve it.
Pulling it together, Cheah and colleagues published a systematic review and meta-analysis in PLOS ONE in 2021, combining five randomised trials and 400 participants. Their conclusion was carefully worded: a small but statistically significant effect on overall sleep, more prominent in people diagnosed with insomnia, at doses of 600 mg a day or more, and over treatment periods of at least eight weeks.
The limitations again matter. All five trials in the meta-analysis were conducted in India. There was substantial variation between studies. 'Small but significant' means measurable, not dramatic - nobody should expect a sleeping-pill effect. And the reviewers themselves noted that safety data for long-term use simply isn't there.
Cheah et al. 2021, PLOS ONE - systematic review and meta-analysis ↗Anxiety
Emerging but limitedStress and anxiety get lumped together in supplement marketing, but they aren't the same thing, and the evidence for them isn't the same strength either. Everyday stress is a feeling. Clinical anxiety is a diagnosis. Ashwagandha's data on the first is better than its data on the second.
What exists is encouraging but thin. Several small randomised trials have measured anxiety alongside stress and found improvements on validated rating scales - the 2021 systematic review of seven trials in 491 Indian adults reported reduced anxiety as well as reduced stress. A trial of 54 adults with mild-to-moderate stress and anxiety in India, using a 500 mg standardised root extract with piperine, found significantly lower anxiety scores at day 60 than placebo. A small Iranian trial in people with diagnosed generalised anxiety disorder reported benefit too.
The most useful summary comes from an international expert taskforce convened by the World Federation of Societies of Biological Psychiatry and the Canadian Network for Mood and Anxiety Treatments. In their 2022 clinician guidelines on nutraceuticals in psychiatry, they gave ashwagandha a provisional recommendation for generalised anxiety disorder at 300 to 600 mg a day of root extract standardised to 5 percent withanolides - and explicitly stated they could not offer a stronger recommendation without more data. That is a genuinely fair reading: enough signal for a cautious professional to consider it, nowhere near enough to call it established.
NCCIH, the NIH's complementary health centre, is blunter still. Their assessment is that some ashwagandha preparations may be effective for insomnia and stress, but that the evidence on anxiety specifically is unclear.
The practical takeaway: if you have an anxiety disorder, this is a conversation to have with your doctor, not a self-treatment decision - particularly because ashwagandha may interact with sedatives and other psychiatric medication. Nothing on this page is a substitute for proper care.
Sarris et al. 2022, WFSBP/CANMAT clinician guidelines ↗Strength, muscle & testosterone
Emerging but limitedThis is where the internet gets loudest and the evidence gets thinner - so it's worth being precise about what was measured.
The study everyone cites is Wankhede and colleagues, 2015. They recruited 57 healthy men aged 18 to 50 with little previous resistance-training experience, and randomised them to 300 mg of root extract twice daily or a starch placebo. Both groups then followed the same eight-week resistance-training programme. At the end, the ashwagandha group had gained more bench-press and leg-extension strength, more arm and chest circumference, and showed a larger rise in testosterone and a larger fall in exercise-induced muscle damage markers than placebo.
Read that carefully, because the framing matters. These were untrained men starting a structured lifting programme - a group who will gain strength regardless. The finding is that ashwagandha appeared to improve the response to training, not that it built muscle on its own. And the testosterone change, while statistically significant, started from normal levels in healthy young men; it is not comparable to what a medical testosterone intervention does. A 2021 systematic review of herbs and testosterone concluded that ashwagandha was among the few with any supportive human data, while noting the studies were small and heterogeneous.
NCCIH's own summary is the sober one: there is some limited evidence that taking ashwagandha for two to four months may increase testosterone levels and sperm quality, and there isn't enough evidence to say whether it helps athletic performance. Those two sentences sit side by side in the same fact sheet, which tells you how narrow the finding really is.
One consequence you won't see on a pre-workout label: because ashwagandha may raise testosterone, expert bodies including NCCIH advise that men with hormone-sensitive prostate cancer should avoid it. A hormonal effect large enough to be worth advertising is large enough to be worth a warning.
Wankhede et al. 2015, J Int Soc Sports Nutr ↗Thyroid hormones
PreliminaryThis one is on the page as a safety issue rather than a benefit, because that's the honest way to present it - even though plenty of websites sell it as a thyroid remedy.
The key study is Sharma, Basu and Singh, published in the Journal of Alternative and Complementary Medicine in 2018. Fifty people with subclinical hypothyroidism - meaning raised TSH with thyroid hormones still in the low-normal range - were randomised to 300 mg of ashwagandha root extract twice daily or placebo for eight weeks. Blood tests at the end showed significantly lower TSH and significantly higher T3 and T4 in the ashwagandha group compared with placebo. In thyroid terms, that's the herb pushing hormone levels upward.
Supplement marketing reads that as 'supports thyroid function'. A more careful reading is that ashwagandha has real, measurable hormonal activity - and anything that can raise thyroid hormones in someone with an underactive thyroid can also push someone else too far. That isn't hypothetical. An earlier study found small increases in T4 in men taking a root-and-leaf extract for eight weeks, in a trial that wasn't even about the thyroid. And there are at least three published case reports of thyrotoxicosis - an overactive thyroid state - in women taking ashwagandha, including a 32-year-old who developed weight loss, shakiness and palpitations six weeks after starting and shortly after increasing her dose. Her symptoms and blood tests returned to normal after she stopped.
Why the grade is 'Preliminary': one small 50-person trial, one incidental finding and a handful of case reports is not a body of evidence. It is enough to establish that an interaction is plausible and worth taking seriously - not enough to characterise it properly, predict who it happens to, or recommend ashwagandha for any thyroid purpose.
The practical rule is simple. If you have hypothyroidism, hyperthyroidism, Hashimoto's, or you take levothyroxine or any other thyroid medication, do not start ashwagandha without talking to the clinician who manages it. NCCIH lists thyroid disorders among the conditions for which ashwagandha is not recommended, and lists thyroid hormone medication among the drugs it may interact with.
Sharma, Basu & Singh 2018, J Altern Complement Med ↗Traditional use (Ayurveda)
Traditional useLong before anyone measured a cortisol level, ashwagandha had a settled place in Ayurveda - and in Canada, that traditional record is actually the legal basis for most of what a product may say.
In the Ayurvedic system it is classified as a rasayana: a rejuvenating tonic, given not to treat a specific complaint but to rebuild strength and vitality, particularly in old age or during recovery from illness. It is also described as a nervine that balances aggravated Vata - the elemental principle associated in Ayurveda with movement, dryness and, when disturbed, restlessness and anxiety. The name itself carries the intent. Ashwa-gandha means 'smell of horse', a reference both to the root's distinctly musky odour and to the strength of a stallion it was said to impart.
Health Canada's Natural and Non-prescription Health Products Directorate publishes a compendial monograph for Withania somnifera root, most recently dated February 2025. It permits exactly these traditional statements, sourced to Ayurvedic pharmacopoeias and classical texts: traditionally used in Ayurveda as a rasayana (rejuvenative tonic); to relieve general debility, especially during convalescence or old age; to help relieve restlessness and/or nervousness (which helps to promote sleep); to balance aggravated Vata (nervine); and for memory enhancement. Separately, it permits one non-traditional statement drawn from Western herbal practice: used in Herbal Medicine as an adaptogen to help increase energy and resistance to stress over time.
The dose ceilings are set by the claim. For the traditional Ayurvedic uses, 2 to 6 grams of dried root per day for dry, powdered and non-standardised preparations. For the adaptogen claim, 2.5 to 6.5 grams of dried root or whole plant per day. Concentrated standardised extracts - the kind used in nearly all the clinical trials - are dosed far lower and expressed as their dried-root equivalent, which is why a 600 mg capsule can sit legitimately inside a 6 gram limit.
What traditional use is, and isn't: it is a genuine, documented record of how a plant has been used by a coherent medical system over a long time. It is not clinical proof of effect. Health Canada is explicit that these are traditional-use statements, and that's exactly how we present them.
Health Canada NNHPD monograph - Ashwagandha, Withania somnifera (February 2025) ↗Who it's for - and who should skip it
A reasonable fit if
You're under sustained, grinding stress and want something gentle alongside the boring fundamentals (sleep, movement, daylight, less caffeine).
You fall asleep fine but wake unrefreshed, or you lie awake with a busy head. The sleep data is modest but real.
You want a non-stimulant that you can take at night without it wrecking your sleep.
You're willing to give it 6-8 weeks and judge honestly.
Probably not your herb if
You're pregnant or trying to conceive. Several regulators advise against it in pregnancy, and Health Canada requires a talk-to-your-practitioner warning. Skip it.
You're breastfeeding.
You have any liver condition, or you drink heavily. Liver injury cases have clustered in people with pre-existing liver disease.
You have a thyroid condition or take thyroid medication. Ashwagandha can nudge thyroid hormones upward - see the practitioner note below.
You have an autoimmune condition, or take immune-suppressing medication.
You take sedatives, sleep medication, or anti-seizure medication.
You have hormone-sensitive prostate cancer - because ashwagandha may raise testosterone, several expert bodies advise avoiding it.
You have surgery coming up. Stop at least two weeks before.
A note on nightshades
Ashwagandha belongs to the Solanaceae - the same botanical family as tomatoes, potatoes, peppers and deadly nightshade. That doesn't make it dangerous, but if you know you react badly to nightshades, this is worth knowing before you start.
How to read the ashwagandha evidence
Ashwagandha has more randomised human trials behind it than almost any other adaptogen. That's genuinely unusual and worth respecting. But the quality of that pile of studies deserves an honest look, because the marketing rarely gives you one.
Three things to keep in mind
Most trials are small and short. Typical study: 50 to 150 people, 6 to 12 weeks. Useful for spotting a signal; not enough to settle a question or to tell you what a year of daily use does.
Almost all of them were run in India. The NIH's own review of the stress and anxiety literature noted that all 491 participants across seven trials were recruited in India, and that every trial in the main sleep meta-analysis was Indian too. That's not a slur on Indian research - it's the country where the herb is best known - but it does mean the findings haven't been widely replicated elsewhere
Many were funded or supplied by extract manufacturers. Industry-sponsored supplement trials tend, on average, to report more favourable results than independent ones. Several of the headline ashwagandha studies were run using a branded extract supplied by its maker.
What that adds up to
Not 'ignore the research' - the direction of the findings is consistent across many trials, which counts for something. But it does mean the honest verdict is 'reasonably supported, modest effect' rather than 'proven'. An international expert taskforce (WFSBP and CANMAT, 2022) landed in exactly that place: they gave ashwagandha a provisional recommendation for generalised anxiety disorder and said plainly that they couldn't say anything stronger without more data.
How it compares
Ashwagandha vs Reishi
Both are reached for when life feels loud. Reishi is a mushroom with a long TCM history as a 'shen' or spirit tonic, and its modern evidence sits mostly in immune and antioxidant territory - the calming reputation is largely traditional. Ashwagandha is the opposite balance: less mystique, more human trial data aimed squarely at stress, cortisol and sleep. If you want the better-studied stress herb, that's ashwagandha. If you want something food-like and gentle with a simpler safety profile, that's reishi.
Ashwagandha vs other calming botanicals
Rhodiola: better suited to stress that shows up as fatigue and mental flatness. Slightly stimulating for some people - the opposite temperament to ashwagandha.
Lemon balm and chamomile: mild, food-grade, good for daytime edginess. Much weaker evidence, but also almost no safety baggage.
Valerian: aimed narrowly at falling asleep, not at daytime stress. Evidence is mixed and the smell is memorable.
Magnesium (glycinate): not a herb, but frequently the more sensible first step - cheap, and with none of ashwagandha's cautions.
L-theanine: fast-acting daytime calm without drowsiness. Complements ashwagandha rather than competing with it.
Where ashwagandha genuinely wins
It's the one in this group with a stack of randomised, placebo-controlled human trials, a measurable biological marker moving in the expected direction (cortisol), and a recognised traditional-use monograph in Canada. Where it loses is safety simplicity - it is the one on this list with real contraindications.
Practitioner note: the two cautions that actually matter
Most botanicals on this site need a light-touch safety paragraph. Ashwagandha needs a real one. Two issues deserve your attention before you start.
1. The liver
Ashwagandha has been linked to rare but genuine cases of liver injury. The NIH's LiverTox database rates it 'B' - a likely cause of clinically apparent liver injury - and its December 2024 entry describes a pattern: jaundice and itching appearing roughly 2 to 12 weeks after starting, usually cholestatic or mixed in type, generally resolving within 1 to 4 months of stopping. Cases were first reported in 2017 and have increased since. Most were mild to moderate. But not all: a 2023 case series of eight patients from three Indian hospitals included three deaths, all in people who already had chronic liver disease, and there is a published case of a 41-year-old woman who needed an emergency liver transplant.
The honest framing: millions of people take ashwagandha and the number of reported cases is small. It is not a common event. But it is a real one, the mechanism isn't understood, and it doesn't appear to be neatly dose-predictable. LiverTox states plainly that ashwagandha should be avoided in people with cirrhosis or advanced chronic liver disease. If you develop yellowing of the eyes or skin, dark urine, persistent itching, unusual fatigue or nausea while taking it - stop and see a doctor.
Several regulators have acted on this and related concerns. Denmark banned ashwagandha in food supplements in 2023, following a Technical University of Denmark risk assessment; France's food safety agency ANSES issued an advisory in 2024 recommending against use in pregnancy, breastfeeding and endocrine disorders. Canada has not restricted it, but you deserve to know that other regulators have looked at the same evidence and drawn a stricter line.
2. The thyroid
Ashwagandha can move thyroid hormones. In a double-blind trial of 50 people with subclinical hypothyroidism, 600 mg/day of root extract for 8 weeks lowered TSH and raised T3 and T4 significantly compared with placebo. Enthusiasts read that as a benefit. Clinically it is better read as evidence of real hormonal activity - which cuts both ways.
There are also published case reports of thyrotoxicosis (an overactive thyroid state) in people taking ashwagandha, including a 32-year-old woman whose symptoms and abnormal thyroid tests resolved after she stopped. If you take levothyroxine or any thyroid medication, ashwagandha can plausibly shift your numbers and your dose. Don't start it without telling the person who manages your thyroid.
Also worth flagging
Autoimmune conditions: NCCIH advises against ashwagandha for people with autoimmune disorders, and it may interact with immunosuppressant medication.
Sedatives and sleep medication: additive drowsiness is plausible. Health Canada requires a caution about driving and machinery for a reason.
Blood sugar and blood pressure medication: possible additive effects; monitor.
Surgery: stop at least two weeks beforehand.
Long-term use: the honest answer is that nobody knows. Good safety data extends to roughly three months. Cycling - a couple of months on, a few weeks off - is a sensible precaution rather than a proven protocol.
— LiverTox, NIDDK/NIH - Ashwagandha (updated December 2024)
What we're actually allowed to say
Myth or fact?
Tap each one to see the verdict.
'Adaptogen' describes a hoped-for pattern of action, not a safety rating. Ashwagandha has documented cases of clinically apparent liver injury - the NIH's LiverTox database rates it 'B', a likely cause, with injury typically appearing 2 to 12 weeks after starting and usually resolving 1 to 4 months after stopping. Most cases were mild to moderate, but fatal cases and one emergency liver transplant have been reported. It also shifts thyroid hormones, may raise testosterone, and interacts with sedatives, immunosuppressants, and diabetes and blood pressure medication. Denmark banned it in food supplements in 2023 and France's ANSES advised against it for pregnancy, breastfeeding and endocrine disorders in 2024. This is a real, active botanical - which is precisely why it might work, and precisely why it needs respect. LiverTox, NIDDK/NIH - Ashwagandha ↗
There is a kernel of truth wrapped in a lot of exaggeration. In an eight-week trial, 57 previously untrained men doing a resistance-training programme showed a larger rise in testosterone on ashwagandha than on placebo, alongside better strength gains. But these were healthy young men with normal levels to begin with, the change was modest, and the effect was on how well they responded to training rather than on hormones in isolation. NCCIH's summary is that there is 'some limited evidence' that two to four months of use may increase testosterone and sperm quality - and, in the same fact sheet, that there isn't enough evidence to say it helps athletic performance. It is not comparable to a medical testosterone intervention, and the same hormonal effect is the reason experts advise men with hormone-sensitive prostate cancer to avoid it. NCCIH (NIH) - Ashwagandha: Usefulness and Safety ↗
Nobody actually knows whether that's safe, and the organisations that review the evidence say so plainly. NIH's Office of Dietary Supplements states that ashwagandha appears well tolerated for up to about three months of use and that its long-term safety is not known; NCCIH says the same. Almost every trial ran 6 to 12 weeks. There is no good data on what a year or five years of daily use does - to the liver, the thyroid, or anything else. That's not a reason to panic, but it is a reason to treat continuous indefinite use as an untested choice rather than a default. A sensible approach is to use it for a defined period, notice whether it's actually doing anything for you, and take breaks. NIH Office of Dietary Supplements - Ashwagandha ↗
The trials that built ashwagandha's reputation almost all used specific branded, standardised extracts - KSM-66 (root only), Sensoril and Shoden (root and leaf), and a handful of others - each with a different withanolide content and a different extraction method. The NIH review notes that root and leaf differ chemically, that doses across trials ranged from 240 mg to 1,250 mg of extract, and that this variety is precisely why it's difficult to translate the research into a single recommendation. A cheap generic root powder is a different product from a 5 percent-withanolide standardised root extract, and results from one do not transfer to the other. Look for a named, standardised extract with a stated withanolide percentage and third-party testing. NIH Office of Dietary Supplements - Ashwagandha ↗
Ashwagandha isn't a fast-acting calming agent, and it wasn't studied as one. Health Canada's own permitted wording is telling: an adaptogen to help increase energy and resistance to stress 'over time'. The trials that found benefit ran 6 to 12 weeks, and the sleep meta-analysis found effects were clearer once treatment reached at least eight weeks. Some people do notice mild drowsiness soon after a dose - which is why Health Canada requires a caution about driving and operating machinery - but that's a side effect, not the mechanism. If you want something that works within the hour, this isn't it. Judge ashwagandha after two months of consistent daily use, not after one capsule on a bad day. Health Canada NNHPD Ashwagandha monograph ↗
How to use it
Common forms
How much
Most successful trials used 300-600 mg per day of a standardised root extract, taken as one dose or split into two. The 2021 NIH review noted benefits looked clearer at 500-600 mg/day than at lower doses, and an international expert taskforce (WFSBP/CANMAT, 2022) put its provisional figure at 300-600 mg/day of extract standardised to 5% withanolides. Plain dried root powder is a different animal - far less concentrated. Health Canada's monograph allows 2-6 g of dried root per day for the traditional Ayurvedic uses, and 2.5-6.5 g/day of dried root or whole plant for the adaptogen claim. Extract milligrams and powder grams are not interchangeable; follow the label on the product you actually have. Timing is flexible. Evening suits most people, since mild drowsiness is the most common side effect. Taking it with food reduces stomach upset. Effects build over 6-12 weeks - judge it at two months, not two days. Good safety data only extends to about three months, so a defined period with breaks is more sensible than indefinite daily use.
Capsules with water, once daily.
Four things separate a research-grade ashwagandha from a filler:
A named, standardised extract. KSM-66 (root only, >5% withanolides) and Sensoril or Shoden (root and leaf, higher withanolide glycosides at much smaller doses) are the extracts behind most of the well-known trials. A generic 'ashwagandha extract' with no standardisation isn't the product the research tested.
A stated withanolide percentage. This is the actual active measure. Most clinical root extracts sit around 5% withanolides; root-and-leaf extracts are expressed differently, as withanolide glycosides. If a label gives you no number at all, you're buying on faith.
Root vs root-and-leaf. Health Canada's monograph covers root and whole plant. Root-only is the more traditional and better-trialled option for stress; root-and-leaf extracts appear in several sleep trials. They are not the same thing chemically.
Third-party testing. Ask for testing on heavy metals and identity. Ayurvedic supply chains have a documented history of heavy-metal contamination, and given ashwagandha's liver signal, product purity matters more here than for most botanicals.
One last thing: a bigger milligram number is not automatically better. A 1,000 mg capsule of unstandardised powder delivers less active material than a 300 mg standardised extract.
Frequently asked questions
Straight answers to the things people actually ask.
Give it about eight weeks. The trials that found benefit for stress, sleep and anxiety generally ran six to twelve weeks, and the 2021 PLOS ONE sleep meta-analysis found the effects were more prominent once treatment reached at least eight weeks. Health Canada's own permitted wording says an adaptogen that helps increase resistance to stress 'over time' - which is the regulator quietly telling you not to expect an instant effect. Some people notice mild drowsiness within an hour or two of a dose, but that's a side effect rather than the point. Take it daily, judge it after two months, and don't bother judging it after two days.
Most successful trials used 300 to 600 mg a day of a standardised root extract, either as one dose or split into two. The 2021 NIH review noted that benefits appeared greater at 500 to 600 mg a day than at lower doses, and the WFSBP/CANMAT expert taskforce's provisional figure for generalised anxiety disorder was 300 to 600 mg a day of extract standardised to 5 percent withanolides. If you're using plain dried root powder rather than an extract, Health Canada's monograph allows 2 to 6 grams a day for traditional Ayurvedic uses and 2.5 to 6.5 grams a day for the adaptogen claim - much bigger numbers, because powder is far less concentrated than extract. Always follow the label on the specific product you have; the numbers are not interchangeable. Timing is flexible. Evening suits most people, since drowsiness is the most common mild side effect, and taking it with food can reduce stomach upset.
For most people it isn't - but the concern is real and you should know the facts rather than a reassuring summary. The NIH's LiverTox database rates ashwagandha 'B', meaning a likely cause of clinically apparent liver injury. Cases were first reported in 2017 and have increased since. The typical picture is jaundice and itching appearing two to twelve weeks after starting, usually resolving within one to four months of stopping. Most cases have been mild to moderate. But a 2023 case series of eight patients from three Indian hospitals included three deaths - all in people who already had chronic liver disease - and a separate published case describes a 41-year-old woman who required an emergency liver transplant. Set against millions of users, reported cases number in the dozens, so this is rare. The practical rules: don't take ashwagandha if you have cirrhosis or advanced chronic liver disease, be cautious if you drink heavily or take other liver-taxing medication, and stop immediately and see a doctor if you notice yellowing of the eyes or skin, dark urine, persistent itching or unusual fatigue.
Not without talking to the clinician who manages it. Ashwagandha demonstrably moves thyroid hormones: in a 2018 double-blind trial of 50 people with subclinical hypothyroidism, 300 mg of root extract twice daily for eight weeks significantly lowered TSH and raised T3 and T4 compared with placebo. There are also published case reports of thyrotoxicosis - an overactive thyroid state - in people taking it, including a 32-year-old woman whose symptoms and abnormal blood tests resolved after stopping. NCCIH lists thyroid disorders among the conditions for which ashwagandha is not recommended, and thyroid hormone medication among the drugs it may interact with. If you take levothyroxine, ashwagandha could shift both your numbers and your correct dose. This isn't a reason for alarm; it's a reason for a conversation and, if you do proceed, monitoring.
No - avoid it. NCCIH states plainly that ashwagandha should be avoided during pregnancy and should not be used while breastfeeding. Health Canada's monograph requires every ashwagandha product to carry a warning to consult a health care practitioner before use if you are pregnant or breastfeeding. France's food safety agency ANSES recommended against use in pregnant and breastfeeding women in 2024, and part of Denmark's rationale for banning it in supplements involved concerns about hormonal and pregnancy-related effects. It's worth noting there is a genuine scientific dispute here: some of the abortifacient concern traces to a 2000 American Herbal Pharmacopoeia monograph, and AHP has since stated publicly that its report was misrepresented and that there is no evidence ashwagandha root causes abortions. But with an unresolved question and no upside worth the risk, the sensible answer during pregnancy is simply no.
They're different products, and the distinction matters more than most people realise. KSM-66 is a root-only extract, standardised to more than 5 percent withanolides, and it's the extract used in several of the best-known stress, sleep and strength trials. Sensoril and Shoden are root-and-leaf extracts standardised to higher withanolide-glycoside levels at much smaller capsule sizes - the Shoden sleep trial used just 120 mg a day. The NIH fact sheet notes that root and leaf differ chemically, that trial doses ranged from 240 mg to 1,250 mg of extract, and that this variation is exactly why it's hard to translate the research into one recommendation. Generic dried root powder is a fourth thing again: far less concentrated, dosed in grams rather than milligrams, and not what the modern trials tested. If you're buying based on the research, buy the kind of product the research used - a named, standardised extract with a stated withanolide percentage.
Across clinical trials ashwagandha has generally been well tolerated for up to about three months. The common side effects are mild: stomach upset, loose stools, nausea and drowsiness. LiverTox notes that large doses in particular can cause gastrointestinal upset, diarrhoea, nausea and vomiting, probably from direct irritation of the gut lining. Drowsiness is common enough that Health Canada requires a label caution about driving and operating machinery. Taking it with food and starting at the lower end of the dose range usually handles the digestive side. The serious but rare effects - liver injury and thyroid changes - are covered separately above, and are worth reading before you start rather than after.
Yes, with several. NCCIH lists potential interactions with diabetes medication, blood pressure medication, immunosuppressants, sedatives, anti-seizure medication and thyroid hormone medication. The sedative interaction is the one people underestimate: ashwagandha can cause drowsiness on its own, so combining it with sleep medication, benzodiazepines, antihistamines or alcohol may compound that. It's also not recommended if you have an autoimmune condition, given its effects on immune signalling, or if you have hormone-sensitive prostate cancer, since it may raise testosterone. And stop it at least two weeks before any planned surgery. If you take prescription medication of any kind, run it past your pharmacist first - it's a two-minute conversation and pharmacists are genuinely good at this.
Less than you'd think from the internet. Health Canada's compendial monograph for Withania somnifera root, dated February 2025, permits five traditional Ayurvedic statements - traditionally used in Ayurveda as a rasayana (rejuvenative tonic); to relieve general debility, especially during convalescence or old age; to help relieve restlessness and/or nervousness (which helps to promote sleep); to balance aggravated Vata (nervine); and for memory enhancement - plus one modern statement: used in Herbal Medicine as an adaptogen to help increase energy and resistance to stress over time. That's the complete list. There is no permitted claim about cortisol, testosterone, muscle, thyroid, depression or anxiety disorders. So if the research on this page discusses cortisol or strength trials, that's us reporting published science, clearly separated from what may lawfully appear on a Canadian label. Anything a brand tells you beyond the wording above isn't a Health Canada-authorised claim.
Side effects & safety
- Liver injury: Rare but documented and occasionally serious. NIH's LiverTox rates ashwagandha 'B' - a likely cause of clinically apparent liver injury - typically appearing 2-12 weeks after starting and resolving 1-4 months after stopping. Most cases were mild to moderate, but deaths and one emergency liver transplant have been reported, mostly in people with pre-existing liver disease. Avoid entirely with cirrhosis or advanced chronic liver disease. Stop and see a doctor for jaundice, dark urine, persistent itching or unusual fatigue.
- Thyroid: Can raise thyroid hormones. In a trial of 50 people with subclinical hypothyroidism, 600 mg/day for 8 weeks lowered TSH and raised T3 and T4 versus placebo. Case reports of thyrotoxicosis exist. Do not start without speaking to whoever manages your thyroid, especially on levothyroxine.
- Pregnancy & breastfeeding: Avoid. NCCIH advises against it in pregnancy and while breastfeeding; Health Canada requires a consult-a-practitioner warning; France's ANSES advised against it in 2024.
- Sedatives & drowsiness: Ashwagandha can cause drowsiness on its own and may compound sedatives, sleep medication and alcohol. Health Canada requires a caution about driving and operating machinery.
- Autoimmune conditions: Not recommended by NCCIH for people with autoimmune disorders, and may interact with immunosuppressant medication.
- Other medications: Possible interactions with diabetes medication, blood pressure medication and anti-seizure medication. Check with your pharmacist.
- Prostate cancer: Because it may raise testosterone, experts advise men with hormone-sensitive prostate cancer to avoid it.
- Surgery: Stop at least two weeks before any planned procedure.
- Nightshade family: Ashwagandha is a Solanaceae, like tomatoes and peppers. Relevant if you know you react to nightshades.
- Long-term use: Safety data extends to roughly three months. Beyond that, genuinely unknown.
The bottom line
Ashwagandha is the rare traditional herb that has actually been put through dozens of randomised, placebo-controlled human trials - and the results lean consistently in one direction: modestly less perceived stress, modestly better sleep, measurably lower cortisol, over about two months of daily use. That's a better evidence base than most of what's sold as an adaptogen.
It is also the rare traditional herb with real cautions attached. Rare liver injury is documented and occasionally serious. It shifts thyroid hormones. It's off the table in pregnancy. One European country has banned it outright and another has advised against it for specific groups.
So the honest position is neither 'everyone should take this' nor 'stay away'. If you're a healthy adult under real stress, not pregnant, with no liver, thyroid or autoimmune issues and no interacting medication - a standardised root extract at 300-600 mg a day for 8 weeks is a reasonable, evidence-informed thing to try. Buy a branded, standardised, third-party-tested extract rather than a cheap generic powder. Give it two months. Stop if anything feels off, and stop immediately for jaundice or persistent itching. And don't ask it to do the job that sleep, daylight and less caffeine should be doing first.
References
Every graded claim links to its source; here they are together.
- Stress & cortisol: Chandrasekhar, Kapoor & Anishetty 2012, Indian J Psychol Med
- Sleep quality: Cheah et al. 2021, PLOS ONE - systematic review and meta-analysis
- Anxiety: Sarris et al. 2022, WFSBP/CANMAT clinician guidelines
- Strength, muscle & testosterone: Wankhede et al. 2015, J Int Soc Sports Nutr
- Thyroid hormones: Sharma, Basu & Singh 2018, J Altern Complement Med
- Traditional use (Ayurveda): Health Canada NNHPD monograph - Ashwagandha, Withania somnifera (February 2025)
- LiverTox, NIDDK/NIH - Ashwagandha
- NCCIH (NIH) - Ashwagandha: Usefulness and Safety
- NIH Office of Dietary Supplements - Ashwagandha
- NIH Office of Dietary Supplements - Ashwagandha
- Health Canada NNHPD Ashwagandha monograph