Open a supplement aisle, or a wellness feed, and ashwagandha turns up everywhere: in “cortisol” gummies, sleep blends, protein powders, even coffee. The pitch attached to it is usually a compact version of the same idea — this root lowers stress.

Three papers published in recent weeks approach that idea from three different directions. Read together they are less a verdict than a map of the gaps: what a two-week trial can show in healthy adults, why “ashwagandha” on a label does not describe one substance, and where people who take it are actually getting their information.

What a two-week trial can and cannot show

The most direct evidence comes from a randomised, double-blind, placebo-controlled trial published on 4 October in the Journal of the International Society of Sports Nutrition. It is not an ashwagandha trial in the usual sense: it tested two blends — a morning product of 600 mg ashwagandha plus 300 mg Rhodiola rosea, and an evening product combining magnesium L-threonate, L-theanine and apigenin — separately, combined, and against placebo.

The 103 participants were healthy and physically active, averaging 34.8 years old. They took their assigned products for two weeks, bracketed by a baseline week before and a washout week after.

The morning group reported a reduction in weekly perceived stress (p = 0.034). The reported effect size is the part worth reading twice: a partial eta squared of 0.020, roughly 2% of the variance in the measure. Daily stress ratings did not differ between groups. That same group reported fewer sleep-related impairments and showed a small reduction in wake after sleep onset measured by a wearable (p = 0.047). Combining the morning and evening products produced no additional benefit on sleep quality over the morning product alone.

The authors are candid about the ceiling. “Minimal effects on other outcomes,” they write, “may reflect limited benefits in healthy individuals over a short supplementation period.” Fourteen days in people who were already well is a narrow window. It says nothing about someone with a clinical sleep or anxiety problem, and it does not test ashwagandha on its own.

A label that covers more than one plant part

Then there is the definitional problem, which got its own paper on 5 October in Food Science & Nutrition. Ashwagandha is Withania somnifera, and the review argues that root and leaf preparations differ enough chemically to deserve separate regulatory treatment. Root is dominated by withanolides A and D and glycol-withanolides; leaf is richer in withaferin A, a compound that appears in far fewer human studies.

The review says recent European risk assessments and Denmark’s ban discussed “ashwagandha” as a single category without consistently separating plant parts, and that analytical surveys have documented root material cut with leaves and stems. Its conclusion is that authenticated, standardised root preparations have the stronger evidence base, while human data on leaf remains sparse.

One detail belongs next to that conclusion: the review’s authors are affiliated with India’s Ministry of Ayush and the World Ashwagandha Council, an industry body for the botanical. They declare no competing interests and state that the views are their own. That is not grounds to dismiss the chemistry, but an industry-authored review of a commercial herb that lands in favour of one standardised form of it is worth reading with that context in view — the discipline this site applies to every evidence check, including our earlier look at probiotic supplements.

What the federal consumer guidance says

The least promotional summary of the evidence sits on NCCIH’s NIH page, last updated in March 2023. It notes that many clinical trials have examined ashwagandha but that many were small and used a variety of preparations. Its read: some preparations may be effective for insomnia and stress, the evidence on anxiety is unclear, and there is not enough evidence to determine whether ashwagandha helps with asthma, athletic performance, cognitive function, diabetes, menopause or female infertility.

On safety, NCCIH says ashwagandha may be safe short term — up to three months — with not enough information to reach conclusions about longer use. Possible effects include drowsiness, stomach upset, diarrhoea and vomiting. The page also names people who should avoid it: during pregnancy and breastfeeding; before surgery; with autoimmune or thyroid disorders; and alongside medicines for diabetes or high blood pressure, immunosuppressants, sedatives, anticonvulsants or thyroid hormone. Because ashwagandha may raise testosterone, NCCIH advises people with hormone-sensitive prostate cancer to avoid it too.

The regulatory frame matters as much as the evidence here. Unlike drugs, dietary supplements are not approved by the FDA before they go on sale. The agency’s own dietary supplements page puts that responsibility on the seller: manufacturers and distributors are “responsible for evaluating the safety and labeling of their products before marketing.”

The rare safety signal, made concrete

The NCCIH line about liver injury — “there have been a number of cases that link liver injury to ashwagandha supplements” — is easy to skim past. A case report published on 29 July 2026 in Case Reports in Hepatology makes it specific. A previously healthy 27-year-old woman developed jaundice and generalised itching after taking ashwagandha for three days. Other causes were excluded, imaging was normal, and a liver biopsy showed a classic cholestatic pattern without fibrosis. Scored with the updated RUCAM causality tool, the case reached 7 points, which the authors classify as probable herb-induced liver injury. She improved with treatment.

One case cannot tell anyone their own risk, and three days is unusually quick. But it undercuts the assumption that a plant product is automatically gentle, and it lines up with the caution in the federal guidance.

Where the advice is coming from

A mixed-methods study published on 18 September in Frontiers in Pharmacology surveyed 508 adults in Saudi Arabia and interviewed 20 of them. Nearly half scored low on knowledge about ashwagandha. Social media and informal networks were the primary sources of information; healthcare professionals were rarely consulted. Positive views tracked with previous use and with self-reported anxiety, and some participants took ashwagandha alongside other herbal or dietary supplements — a combination the authors flag as an interaction risk.

The sample is one country’s adults, so the figures are a prompt rather than a global measurement. The pattern they describe — enthusiasm moving faster than information — is a familiar one in the supplement aisle.

What none of this settles

Nothing above says ashwagandha does nothing, or that it is unsafe for everyone. The narrower reading is this: the evidence for stress and sleep is mixed, mostly short-term and often small in effect; the word on the label does not reliably tell you which part of the plant is inside; and the liver reports and the interaction list are specific enough to matter to particular people.

Anyone who takes prescription medicine, has a liver condition, or is pregnant or breastfeeding has a concrete reason to raise it with a clinician before adding a supplement. As our medical disclaimer sets out, nothing here is individual medical advice.

For sleep itself, the better-evidenced levers remain unglamorous and free — consistent timing above all. Our guide to things not to do before bed and our coverage of sleep regularity research cover that ground in more detail.