Ashwagandha is an Ayurvedic root, used for centuries as a Rasayana and taken today mostly for stress and sleep. The small trials point one way. Over eight to twelve weeks a standardized root extract lowers stress-scale scores and morning cortisol, and nudges sleep. In men who lift, it adds modest strength and testosterone.
The catch is evidence quality: the positive trials are small and short. It also has three real cautions: rare liver injury, raised thyroid hormones, and no pregnancy safety data. A standardized root extract at 300 to 600 mg a day, for eight to twelve weeks, is the studied approach.
Findings & Outcomes
It also has three real cautions: rare liver injury, raised thyroid hormones, and no pregnancy safety data.
What It Is
Ashwagandha (Withania somnifera) is a small shrub. Its root has been used for centuries in Ayurveda, India's traditional medicine, where it is classed as a Rasayana, a rejuvenating tonic. It belongs to the nightshade family, the same botanical group as tomato, potato, and bell pepper.
Modern supplements use the root, usually as a standardized extract concentrated for compounds called withanolides. The two most-studied products are branded extracts sold under their own names. People take it mainly to feel less stressed and to sleep better. That is where the human evidence is best, though even the best of it rests on a few short trials.
Anatomy of the Practice
1A root, standardized to withanolides
The active preparation is a concentrated root extract. Withanolide content is what a product is standardized to, and it differs from brand to brand. So a result for one extract does not automatically apply to another.
2On the stress axis
The best-mapped effect is on the hypothalamic-pituitary-adrenal (HPA) axis, which sets cortisol output. In trials, ashwagandha lowered morning cortisol more than placebo. That drop is the mechanism researchers point to for how it shifts the way stressed people feel.
3Active across several systems
The same broad activity reaches the thyroid, the immune system, and the nervous system. Because it acts on so many systems, several conditions call for care (see Cautions).
What It Does
The best evidence is for stress and sleep, the same uses Ayurveda has always claimed. A 2022 meta-analysis found that a standardized root extract lowered anxiety scores against placebo, though the pooled figure hides wide variation between trials. A separate analysis of five trials in 400 people found a small, steady gain in sleep, strongest in people who already have insomnia.
The evidence for strength and testosterone is far thinner. In 57 untrained young men, ashwagandha paired with eight weeks of resistance training added more strength and testosterone than placebo. In one mixed-sex stress trial, testosterone rose in the men but not the women. That is a thin base in selected male groups, and ashwagandha is not a treatment for low testosterone.
One limitation applies to every finding above. Almost every positive trial is small, runs eight to twelve weeks, and tests a single branded extract, often funded by the company that sells it. The direction is consistent; the size is not settled. Each result holds only for the extract and dose tested, over two to three months, with nothing shown past that.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
Anxiety And Stress
Two 60-day trials cut perceived stress versus placebo
In two two-month trials, people who took a standardized ashwagandha extract reported meaningfully less stress than people who took a placebo.
Chandrasekhar 2012 randomized 64 adults with chronic stress to 300 mg of a high-concentration full-spectrum root extract twice daily or placebo for 60 days, and found significant reductions on all stress-assessment scales (P<0.0001). Lopresti 2019 randomized 60 stressed adults to 240 mg/day of a different standardized extract (Shoden) or placebo for 60 days, with a significant Hamilton Anxiety reduction (P=0.04) and a near-significant Depression Anxiety Stress Scale change (P=0.096). Both trials are small, short, and each tests a single proprietary extract; in the Lopresti trial the lead author had prior study funding from a supplement maker.
Who this may not transfer to:Measured in stressed adults of both sexes; the stress result is not specific to one sex.
The studies · 2
Chandrasekhar 2012, Indian J Psychol Med · Indian J Psychol Med
Lopresti 2019, Medicine (Baltimore) · Medicine (Baltimore)
Counts once: this finding and 2 others here come from the same source, so they are one body of evidence, not separate confirmations.
How it works
Morning cortisol fell more on ashwagandha than placebo
Ashwagandha lowered the stress hormone cortisol in the morning more than a placebo did, which points to an effect on the body's central stress axis.
Both trials measured cortisol as a mechanism marker alongside the stress scales. Lopresti 2019 found a significantly greater fall in morning cortisol (P<0.001) and DHEA-S (P=0.004) versus placebo over 60 days at 240 mg/day; Chandrasekhar 2012 found reduced serum cortisol (P=0.0006) at 600 mg/day. The authors read this as a moderating effect on the hypothalamic-pituitary-adrenal axis. Neither abstract reports the absolute change in nanograms, so the size of the drop in everyday terms is not established from these sources.
Who this may not transfer to:Cortisol was measured in adults of both sexes.
The studies · 2
Lopresti 2019, Medicine (Baltimore) · Medicine (Baltimore)
Chandrasekhar 2012, Indian J Psychol Med · Indian J Psychol Med
Counts once: this finding and 2 others here come from the same source, so they are one body of evidence, not separate confirmations.
Mood & stress
Pooled anxiety dropped a large SMD -1.55, but the trials disagreed sharply (I-squared 93.8%)
Pooling the trials, ashwagandha lowered anxiety and stress by a large amount on average, but the trials disagreed with each other so much that the average number carries less weight than its size suggests.
Akhgarjand 2022 pooled randomized trials of ashwagandha for anxiety and stress using a random-effects model. The pooled standardized mean differences were large (anxiety -1.55, stress -1.75), but the heterogeneity statistics were extreme (I-squared above 83% for both), meaning the underlying trials produced widely different results. Extreme heterogeneity plus a small number of short trials of a few proprietary extracts is why the effect is graded emerging, not moderate: the direction is consistent, the magnitude is not reliable.
Who this may not transfer to:Pooled from trials in adults of both sexes.
The studies · 2
Akhgarjand 2022, Phytother Res · Phytother Res
Chandrasekhar 2012, Indian J Psychol Med · Indian J Psychol Med
Counts once: this finding and 2 others here come from the same source, so they are one body of evidence, not separate confirmations.
Evidence And Methods
Most positive trials are small (under 100 people), short (8 to 12 weeks), and tied to the extract maker
The trials that look good are small, short, and usually tied to the company selling that particular extract. That makes the stress result promising, not proven, and it does not carry across every ashwagandha product on a shelf.
Cheah 2021 (sleep) and Akhgarjand 2022 (anxiety and stress) both note the same structural limits: few trials, small samples, short durations, high heterogeneity, and reliance on branded standardized extracts. A benefit shown for a specific extract at a specific dose does not automatically transfer to a different product with a different withanolide content. This is a statement about the evidence base, not a claim that the effect is absent; larger independent trials would settle the magnitude.
Who this may not transfer to:Describes the randomized-trial literature across both sexes.
The studies · 2
Akhgarjand 2022, Phytother Res · Phytother Res
Cheah 2021, PLoS One · PLoS One
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
Sleep
Five trials in 400 adults show a small sleep gain (SMD -0.59)
Across five trials in 400 people, ashwagandha produced a small but consistent improvement in sleep, and the effect was clearest at 600 mg a day for at least eight weeks and in people who actually had insomnia.
Cheah 2021 pooled five randomized controlled trials (400 participants) and found a small significant benefit on overall sleep (standardized mean difference -0.59, 95% CI -0.75 to -0.42), plus improvements in mental alertness on rising and anxiety, but no significant effect on quality of life. The moderate heterogeneity (I-squared 62%) is far lower than the anxiety meta-analysis, which makes the sleep signal steadier. The review notes that long-term safety data are limited.
Who this may not transfer to:Pooled from adults of both sexes, healthy and with insomnia.
The studies · 2
Cheah 2021, PLoS One · PLoS One
Langade 2021, J Ethnopharmacol · J Ethnopharmacol
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
People with insomnia fell asleep faster and slept more efficiently on 300 mg twice daily
People with insomnia fell asleep faster and spent more of their time in bed actually asleep on ashwagandha, and they gained more than healthy sleepers did.
Langade 2021 ran a double-blind, randomized, parallel-group trial in both healthy volunteers and insomnia patients taking 300 mg of root extract twice daily. Sleep onset latency and sleep efficiency were the most improved parameters (both P<0.0001 within group), with total sleep time and wake-after-sleep-onset also improving, and the effect was more pronounced in the insomnia group. It is a single small trial and the authors call for larger studies to generalize the result.
Who this may not transfer to:Measured in adults of both sexes, healthy volunteers and insomnia patients.
The study · 1
Langade 2021, J Ethnopharmacol · J Ethnopharmacol
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
Muscle And Strength
In 57 untrained young men, bench-press max rose 101 lb (46 kg) versus 57 lb (26 kg) on placebo
Young men new to lifting who took ashwagandha alongside eight weeks of training gained more strength and lost a little more fat than those who trained on a placebo, though it was one small study.
Wankhede 2015 randomized 57 men aged 18 to 50 with little training experience to 300 mg of root extract twice daily or placebo, both groups training for 8 weeks. The ashwagandha group gained more bench-press strength (101 lb (46.0 kg) vs 58 lb (26.4 kg), p=0.001), more leg-extension strength (32 lb (14.5 kg) vs 22 lb (9.8 kg), p=0.04), more arm muscle size, and greater body-fat reduction (p=0.03). It is a single trial in untrained young men using one branded extract, and the effect has not been replicated in women, older adults, or trained lifters.
Who this may not transfer to:Tested only in untrained young men; not shown in women, older adults, or experienced lifters.
The study · 1
Wankhede 2015, J Int Soc Sports Nutr · J Int Soc Sports Nutr
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
Progress Markers
Testosterone rose more in training men (+96.2 vs +18.0 ng/dL), not in women
In a couple of small trials in men, testosterone rose a little more on ashwagandha than on a placebo. The effect was not seen in women, and this is early, small evidence, not a treatment for low testosterone.
Wankhede 2015 found a greater testosterone increase on ashwagandha than placebo in young training men (+96.2 ng/dL vs +18.0, p=0.004). In the mixed-sex Lopresti 2019 stress trial, testosterone rose in males (P=0.038 over time) but not females (P=0.989), and the male change was not significant against placebo (P=0.158). The evidence is a handful of small trials in selected male groups; the size and durability of any effect are not established.
Who this may not transfer to:Testosterone effect seen only in men; no change measured in women.
The studies · 2
Wankhede 2015, J Int Soc Sports Nutr · J Int Soc Sports Nutr
Lopresti 2019, Medicine (Baltimore) · Medicine (Baltimore)
Counts once: this finding and 2 others here come from the same source, so they are one body of evidence, not separate confirmations.
How It Works
For stress, the leading explanation is ashwagandha's effect on cortisol. How much of the calmer feeling that actually accounts for has not been measured.
For sleep, the proposed pathway is activity at GABA, the brain's main calming signal. That fits the mild effect the trials found: people fall asleep somewhat faster, without the heavy sedation of a sleep drug. GABA activity shows up in animal and in-vitro work, but is not confirmed as the human mechanism.
How To Take It
Ways to Do It
For the same stress axis, the free and better-evidenced moves come first: paced breathing, and, for insomnia, the CBT-I protocol. A standardized extract sits alongside them as an adjunct. Whether it fits is a call worth making with your doctor, especially if you have any of the conditions in Cautions. The protocol itself is short. The form (root extract), the dose, and a defined run of use matter more than the brand.
Trials used standardized root extract, usually 300 mg twice daily, or a single 600 mg dose. The sleep benefit was clearest at 600 mg a day or more, for at least eight weeks. Those trials tested root extract; whole-plant and leaf preparations were not studied. Because withanolide content varies between products, a third-party-tested extract that states its standardization is the sensible choice.
The trials ran roughly eight to twelve weeks. There is no good data beyond that, so anything longer is outside what has been studied.
Any effect builds over the eight-to-twelve-week trial window, not in a single dose. The trials measured change against a starting baseline, so compare how you feel now with how you felt before you started. That before-and-after is the only real read on whether it did anything for you. The practices with stronger evidence for the same goals are worth pairing with it.
Go Deeper
- Anxiety: the condition ashwagandha is most often reached for, with the full range of what helps and how the evidence compares.
- Breath and heart-rate variability: a free, well-evidenced way to act on the same stress axis, with no safety signals to weigh.
- The CBT-I protocol: the first-line, best-evidenced approach to insomnia, the place to start for sleep.
- Resistance training: the loading the strength and testosterone finding was built on, since the trial paired the herb with eight weeks of lifting.
- Testosterone therapy: what actually treats low testosterone, and why a handful of small trials do not make ashwagandha that.
Cautions For This Practice
Extra restraintEverything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
Documented liver injury from pure product, fatal in 3 of 3 with existing liver disease
Bjornsson 2020 described 5 cases from Iceland and the US Drug-Induced Liver Injury Network: jaundice after a 2 to 12 week latency, cholestatic or mixed pattern, prolonged itching and high bilirubin, all self-limited with liver tests normalizing in 1 to 5 months. Philips 2023 reported 23 patients with ashwagandha-associated liver injury in India, of whom 8 took single-ingredient ashwagandha with no adulteration on chemical analysis; 5 had underlying chronic liver disease, 3 developed acute-on-chronic liver failure, and all 3 died. The injury is idiosyncratic and uncommon, but it is real and not explained by contamination. LiverTox, the NIH reference on drug-induced liver injury, carries a dedicated ashwagandha entry.Bjornsson 2020, Liver Int (Iceland and US DILIN case series)Philips 2023, Hepatol Commun (India case series and review)LiverTox: Ashwagandha, NIDDK/NCBI Bookshelf
No more side effects than placebo over 8 to 12 weeks, including thyroid labs
Verma 2021 gave healthy volunteers root extract for 8 weeks and found no significant change or abnormality in vital signs, hematology, biochemistry, or thyroid profile versus placebo, and no reported adverse events. Chandrasekhar 2012 reported adverse effects that were mild and comparable between ashwagandha and placebo. These are the reassuring short-term safety data, but the trials run weeks, not years and are far too small to detect the rare idiosyncratic liver injury that post-marketing case reports have found.Verma 2021, Complement Ther MedChandrasekhar 2012, Indian J Psychol Med
In 50 hypothyroid patients T3 and T4 rose and TSH fell: a caution if your thyroid is not underactive
Sharma 2018 randomized 50 people with subclinical hypothyroidism to 600 mg/day of root extract or placebo for 8 weeks and found ashwagandha significantly raised T3 and T4 and lowered TSH, effectively normalizing thyroid indices. The direction of this effect depends entirely on the person: welcome in an underactive thyroid, but a thyroid-stimulating activity that can stack with levothyroxine or worsen an overactive thyroid. Larger short trials in healthy people (Verma 2021) found no significant change in thyroid profile, so the effect is clearest where the thyroid is already underactive.Sharma 2018, J Altern Complement Med
Liver injury: avoid it if you have liver disease
Pure, single-ingredient ashwagandha has caused liver injury in independent case series, occasionally severe, and in people who already have liver disease it has been fatal. Do not take it if you have liver disease. If you develop jaundice, dark urine, or abdominal pain while taking it, stop and seek medical care.
Thyroid disease or thyroid medication
Ashwagandha raises thyroid hormone levels. If you have a thyroid condition or take thyroid medication, it can push your levels too high, so talk to your doctor before starting.
Do not take it in pregnancy, and take care around conception and breastfeeding
Ashwagandha is contraindicated in pregnancy. It has a traditional use for bringing on menstruation, and animal studies have linked it to pregnancy loss. Avoid it while pregnant and while trying to conceive. There is not enough safety data to support use while breastfeeding, so the cautious course is to avoid it then as well.
Autoimmune disease and immunosuppressant medication
Ashwagandha can stimulate immune activity. If you have an autoimmune condition (lupus, rheumatoid arthritis, Hashimoto's thyroiditis, or multiple sclerosis) added immune stimulation could aggravate it. If you take an immunosuppressant, for example after a transplant or for an autoimmune condition, the herb could work against the drug. Raise it with your doctor before starting.
Sedatives, alcohol, and surgery
Its calming effect can add to the effect of sedatives, benzodiazepines, sleep medication, and alcohol, so combining them can leave you more sedated than you intend. Because of that sedative action, stop ashwagandha well before any planned surgery and tell the anesthesiologist you have been taking it.
It is a nightshade
Because ashwagandha is a nightshade (see What It Is), anyone with a known sensitivity or allergy to nightshades may react to it.
Start slow, be smart, read the research, and consult a professional if you have any concerns. This is here to inform your choice, not make it for you.
Common Questions
Does ashwagandha actually reduce stress?
In short trials, most run two months, yes, by a modest amount, measured against placebo and not just against nothing. Whether it reproduces outside those small, industry-linked studies is still open.
Is ashwagandha safe because it is natural?
Natural describes where ashwagandha comes from. It says nothing about how the body handles it. Most people tolerate it well in short trials, with no more side effects than placebo. But it is not risk-free: the liver, thyroid, and pregnancy warnings in Cautions are the reason.
How much should I take, and for how long?
For eight to twelve weeks, the length of the trials. There is no long-term safety data, so a longer run is uncharted rather than shown safe. Splitting the dose across the day or taking it once does not appear to change the result.
Can ashwagandha raise testosterone or build muscle?
The signal comes from two small trials in men. The size and durability of any effect are unknown. That makes it the least certain of ashwagandha's claims.
Who should avoid ashwagandha entirely?
Two groups should avoid it outright: anyone pregnant or trying to conceive, and anyone with liver disease. Thyroid disease, autoimmune conditions, and use of sedatives or immunosuppressants each turn it into a decision to make with your doctor first.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 20 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 9, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.