Ashwagandha: effects, studies and a safe assessment
Clinical review pending
Stress, sleep, hormones and training: what ashwagandha studies show, which amounts were tested and which risks matter.
Ashwagandha draws interest mainly for its possible effects on stress and sleep, and small clinical studies show encouraging signals there. At the same time, the plant can affect hormones and, in rare cases, damage the liver. Whether taking it makes sense therefore depends as much on the product, the symptoms and any existing conditions as on the hoped-for benefits.
What is ashwagandha?
Withania somnifera is an evergreen shrub whose root has been used in Ayurvedic medicine for centuries. Other names include winter cherry and Indian ginseng. Supplements are marketed especially for stress, sleep and restlessness.
The common label “adaptogen” describes plants that are credited with helping the body cope with strain. It is not in itself a guarantee of effectiveness or safety. Being sold as a dietary supplement also does not mean a product has met the same standard of proof as an approved medicine. The research consists of many small studies using different preparations. Overview
What are withanolides?
Withanolides are naturally occurring steroidal lactones. Withaferin A and withanolide A are among the most studied compounds, and the plant contains other groups of substances as well. The plant part used and the growing conditions affect the composition. So two packs labeled “ashwagandha” will not necessarily contain the same thing. A standardized content figure helps describe a product, but it does not replace testing of the product as a whole. Plant constituents
What does ashwagandha do in the body?
Human studies have mainly observed changes in perceived stress and in cortisol. How exactly these changes come about is less well understood. Lab experiments show that an aqueous root extract can activate certain GABA receptors. Interestingly, the individual withanolides tested separately did not do this in the same way. That argues against overly simple explanations based on a single “active ingredient,” and for now it remains a finding from an experimental system. Receptor study
Hormonal changes are also possible. In a study of people with elevated TSH, 600 mg a day for eight weeks changed TSH, T3 and T4. A measurable change, however, is not proof that self-treatment with it is beneficial. Thyroid study
Which benefits have clinical evidence?
The findings are most interesting for stress and sleep. Other areas range from physical performance and male hormones to memory. What the studies have in common is limited size and duration; some were funded by manufacturers. Different extracts and widely varying results make it hard to put a reliable number on the size of the benefit. Evidence review
Stress and cortisol
A meta-analysis of nine randomized trials with 558 participants reported an average improvement of 4.72 points on the Perceived Stress Scale compared with placebo. For serum cortisol, a mean difference of −2.58 was given. Without the unit used in the specific analysis method, this second figure is not a usable everyday target; in particular, it does not mean “2.58% less stress.” Meta-analysis
A study of 64 chronically stressed adults tested 300 mg of a concentrated root extract twice a day for 60 days. Stress scores and cortisol fell statistically significantly compared with placebo, with reported values of p < 0.0001 and p = 0.0006 respectively. Mild side effects were similarly common in both groups. Another study of 60 adults compared 125 mg twice a day, 300 mg twice a day or placebo over eight weeks. The change in stress scores was larger at a total of 600 mg than at 250 mg a day. That does not justify a blanket dose increase for every other product. Study of 64 adults, Dose comparison
Sleep quality
In a review, five randomized trials with a total of 400 people showed a favorable overall effect on sleep, with a standardized mean difference of −0.59. This statistical measure is not a number of extra hours of sleep. Larger effects appeared in subgroups with insomnia, at doses of at least 600 mg a day and in studies lasting at least eight weeks. Subgroup findings are no guarantee of an optimal dose, however. Sleep review
An eight-week study of 80 people reported improvements especially in the time it took to fall asleep and in sleep efficiency, as well as in total sleep time and time spent awake. People with existing sleep problems benefited more than those who already slept well. Single study
Anxiety and mood
A meta-analysis of twelve studies with 1,002 people found standardized effects of −1.55 for anxiety and −1.75 for stress. The differences between studies were very large: I² was 93.8% and 83.1% respectively. The certainty of the evidence was rated as low. Positive signals justify further research; they do not make ashwagandha a substitute for treating an anxiety disorder. Meta-analysis
Strength, endurance and recovery
A Bayesian meta-analysis of twelve studies reported benefits on various measures of performance and recovery but called for further research in trained athletes. Four studies with 142 people together showed an increase in maximal oxygen uptake of about 3.00 ml/kg/min. Performance review, VO2max analysis
A training study of 57 young men with little previous training used 300 mg twice a day for eight weeks. The gain in maximum bench press weight was about 46.0 kg, compared with 26.4 kg on placebo. These striking results come from a small group of beginners, whose training gains can be especially large. They cannot be read as the progress experienced strength athletes should expect. Training study
Testosterone and male fertility
A crossover study of 57 overweight men aged 40 to 70 used 21 mg of withanolide glycosides a day for eight weeks. Salivary testosterone rose by 14.7% and DHEA-S by 18% compared with placebo. Even so, there was no significant difference in fatigue, vitality or sexual well-being. A changed hormone level, in other words, is not automatically an improvement you can feel. Hormone study
A pilot study of 46 men with low sperm counts tested 675 mg a day, split into three doses, for 90 days. Within the treated group, sperm count rose by 167%, semen volume by 53% and motility by 57%. These figures are not matching increases in the likelihood of pregnancy, and they do not automatically apply to healthy men. Pilot study
Blood sugar, inflammation and memory
One review of metabolic effects included five clinical, 13 preclinical and six in vitro studies. Although changes in glucose, HbA1c, insulin, blood lipids and oxidative stress were described, the authors did not consider the clinical data solid enough to support use for treating diabetes. Review
For memory, a systematic review found five suitable clinical studies with very different groups of participants. One specific pilot study looked at 50 adults with mild cognitive impairment: 300 mg twice a day for eight weeks improved several memory and attention tests. The result is promising, but it does not answer whether healthy people can expect a reliable boost in performance. Review, Pilot study
How much was studied, and when?
Many studies used 300 to 600 mg of root extract a day, sometimes with about 5% withanolides. The NIH fact sheet also describes a provisional recommendation from a WFSBP/CANMAT working group for one specific standardized extract in generalized anxiety. An expert assessment of this kind is not a general guide to self-treatment. NIH fact sheet
| Outcome studied | Study amount per day | Dosing pattern | Observation period |
|---|---|---|---|
| Stress and cortisol | 300 mg twice a day | Morning and evening | 60 days |
| Sleep | At least 600 mg in one subgroup analysis | No optimal timing established | At least eight weeks in this analysis |
| Strength and recovery | 300 mg twice a day | Alongside a training program | Eight weeks |
| Memory | 300 mg twice a day | Two doses | Eight weeks |
| Male fertility | 675 mg in total | Three doses | 90 days |
The table describes research, not an individual prescription. Before taking ashwagandha, contraindications, medications and the specific product matter more than borrowing a number. If a trial makes sense after talking it over with a professional, the goal and the point at which it will be reassessed should be set in advance. A lab schedule from a safety study is not a precaution that everyone automatically needs.
Which form matches the studies?
| Product form | Plant part or standardization | Assessment |
|---|---|---|
| Standardized root extract | Root, often about 5% withanolides | Appears in studies at 300 to 600 mg a day; products still differ. |
| Extract with withanolide glycosides | 21 mg of glycosides a day in the hormone study | This amount of active compounds is not the same as 21 mg of ordinary powder. |
| Concentrated full-spectrum root extract | Several root constituents | Used in stress and fertility studies. |
| Non-standardized root powder | No guaranteed withanolide content | Cannot simply be converted from extract studies. |
The plant part and the stated constituent content belong together. Even identical percentages do not guarantee an identical extraction method or effect. NIH fact sheet
How long until a possible effect?
Many studies ran for eight weeks, and sleep studies often for six to twelve weeks. That allows an assessment over this period, but it does not promise an effect will kick in on any particular day. If symptoms or side effects appear, no one should push through to the end of a supposedly required trial period.
For a personal assessment, concrete goals help: less distressing rumination, more restful sleep or coping better with everyday life. If nothing improves, reassessing makes more sense than assuming that all that is missing is a higher dose.
Who should not take ashwagandha on their own?
Particular caution, or avoiding it altogether, is called for during pregnancy and breastfeeding, and with existing liver disease, thyroid disease, autoimmune disease, hormone-sensitive prostate cancer and upcoming surgery. Sedating medications and immunosuppressants should also be checked beforehand. NCCIH
The risk is especially relevant in advanced liver disease. In an Indian case series of eight cases of liver injury caused by a single-ingredient ashwagandha product, three patients had acute-on-chronic liver failure; all three later died. A case series cannot give a frequency for all users, but it shows why existing liver problems have to be taken seriously. Case series
Side effects and interactions
The most frequently reported effects are drowsiness, upper abdominal discomfort, loose stools or diarrhea, and nausea or vomiting. A review of 30 human studies names drowsiness, epigastric discomfort and loose stools as the events reported in more than 5% of participants. That is not a precise prediction of frequency for every product. Safety review
In a controlled eight-week study of 80 healthy adults taking 300 mg twice a day, there were no significant abnormalities in the blood, liver and thyroid values that were tested. A small study cannot rule out rare harms, however. Safety study
Reported liver injuries often appeared two to twelve weeks after starting, frequently with a cholestatic or mixed pattern, meaning bile flow was affected. In a series of five patients, jaundice occurred, in some cases with nausea, itching, exhaustion and abdominal discomfort. In those who were followed up, values returned to normal within one to five months; no liver failure was described in this series. Case series, LiverTox
Yellowish skin or eyes, dark urine, itching, pain under the right rib cage, and unusual tiredness or loss of appetite are reasons to stop taking it and to seek medical advice promptly.
Interactions mainly involve sedating drugs, immunosuppressants and thyroid medications. One case report describes a 73-year-old woman with thyrotoxicosis and supraventricular tachycardia after two years of treating her own underactive thyroid with ashwagandha. Her symptoms improved after she stopped. A single case does not prove the risk for every person, but it makes the possible hormonal effect clinically relevant. Case report
Which lab values can be relevant in individual cases?
The original version recommends a baseline panel and a repeat after eight weeks. That schedule was used in one study; it does not amount to a general guideline recommendation. Which values make sense depends on existing conditions, medications, symptoms and the reason for taking ashwagandha.
| Value | Why it can be relevant | Notes on timing |
|---|---|---|
| Cortisol | Changed as an outcome measure in studies | Not a routine test that proves benefit or “stress normalization” |
| TSH and free T4 | Possible thyroid effects | Plan with a doctor if there is a relevant history or symptoms |
| ALT and AST | Possible liver injury | Get checked right away if symptoms appear; do not wait for an eight-week check |
| Testosterone and DHEA-S | Changes in small studies of men | Only for a specific clinical question |
| Fasting glucose and HbA1c | Metabolic effects are being studied | Based on an existing indication; not an established test of ashwagandha’s effect |
Weighing benefit and tolerability together
A blood value alone does not tell you whether taking it is improving your life. Likewise, feeling good does not rule out an unwanted hormonal change. A sensible assessment combines symptoms, daily life, medications and, where appropriate, carefully chosen tests. An extensive commercial panel is not automatically needed for this.
Frequently asked questions
What does ashwagandha actually do?
The strongest positive clinical signals are for perceived stress and sleep. The mechanisms and the reliable size of the benefit are less clear than some product claims suggest.
How much should I take each day?
Studies often tested 300 to 600 mg of specific extracts. That is not a universal dose: composition, contraindications and medications all have to be taken into account.
When can I expect an effect?
Many studies assessed results after about eight weeks. That is no guarantee of success and no reason to put up with side effects until then.
Does ashwagandha help with sleep?
There is evidence of a limited benefit, especially for people who already have trouble sleeping. Breathing pauses during sleep or other specific sleep disorders still need to be looked into on their own.
Does it raise testosterone?
Small studies found changes. In the training study, for example, levels rose by 96.2 ng/dl, compared with 18.0 ng/dl on placebo. Other hormonal changes did not go along with better vitality or sexual well-being. Training study
What side effects are possible?
Tiredness and digestive complaints are the most commonly reported. The good short-term tolerability seen in many study participants does not rule out rare serious events.
Can it damage the liver?
Yes, such cases are documented. Jaundice, dark urine or pronounced new symptoms while taking it should be checked promptly. People with existing liver disease need to be especially cautious.
Who should avoid it?
Pregnant and breastfeeding women in particular, as well as people with relevant liver, thyroid or autoimmune disease, should not start a trial on their own. Hormone-sensitive prostate cancer, surgery and certain medications are also reasons to avoid it or to have it checked by a professional first.