Ashwagandha for Anxiety: What the Trials Show, Dose and Safety

Ashwagandha for anxiety has been tested in more than a dozen randomized trials, but most were small and short^[1]. The largest review of people with diagnosed conditions, published in BJPsych Open in 2025, pooled 14 randomized trials with 360 people given ashwagandha and 353 controls^[1].

Most of those participants had an anxiety disorder, and ashwagandha reduced anxiety symptoms with an outlier-corrected standardized mean difference (SMD) of -1.13 (95% CI -1.65 to -0.60)^[1]. The authors flagged three limits: few studies, substantial differences between them and the pull of outlier results^[1].

Prevalence figures by age, sex and state are on the anxiety statistics and generalized anxiety disorder pages.

Other reviews and the NIH view

A 2022 meta-analysis in Phytotherapy Research combined 12 trials with 1,002 adults aged 25 to 48 and found an anxiety SMD of -1.55 versus placebo^[2]. Its heterogeneity was very high (I² = 93.8%), and the authors rated the certainty of evidence as low^[2].

A 2024 review in Explore pooled 9 trials with 558 participants and reported scores on the Hamilton Anxiety Rating Scale (HAM-A) 2.19 points lower than with placebo (95% CI -3.83 to -0.55)^[6]. The same review found lower Perceived Stress Scale (PSS) scores and lower serum cortisol^[6].

The U.S. National Center for Complementary and Integrative Health (NCCIH) reads the same literature more cautiously. Its fact sheet, last updated in 2023, says some preparations may help insomnia and stress but that the evidence on anxiety is unclear^[5].

Much of the ashwagandha research measures chronic stress, not an anxiety disorder. NIMH defines stress as a response to an external cause, and anxiety as the body's reaction to stress that can occur even when there is no current threat^[8]. Generalized anxiety disorder (GAD) requires hard-to-control worry on most days for at least 6 months plus at least three other symptoms^[8].

The cortisol findings are more consistent than the self-report findings. A 2025 meta-analysis found serum cortisol fell by 1.16 µg/dL, yet the change in Perceived Stress Scale scores across 6 trials was not statistically significant^[7]. The author called for longer trials and more diverse participants to explain the gap^[7].

One widely cited trial shows how the research is usually designed. It enrolled 64 adults with chronic stress, gave 300 mg of root extract twice a day for 60 days, and found lower scores on every stress scale and lower serum cortisol than with placebo^[9]. Participants had a history of chronic stress rather than a diagnosed anxiety disorder^[9].

Some trials found no stress effect at all. In a 12-week trial of 120 adults with fatigue, ashwagandha root extract reduced fatigue but not perceived stress compared with placebo^[4].

Doses varied widely. In a 2021 review of 7 trials with 491 adults, participants took 240 to 1,250 mg a day of extract, or 12,000 mg a day of whole-root granules in one study^[4]. Several of those studies saw larger benefits at 500 to 600 mg a day than at lower doses^[4].

The 2022 WFSBP and CANMAT taskforce provisionally recommended 300 to 600 mg a day of root extract standardized to 5% withanolides for generalized anxiety disorder^[4]. The 2025 review of patients with diagnosed disorders found a median oral dose of 600 mg a day^[1].

Dose-response results are less tidy. The 2022 meta-analysis found a favorable effect on anxiety at doses up to 12,000 mg a day, while the stress benefit centered on 300 to 600 mg a day^[2].

KSM-66, Sensoril, Shoden and other extracts

Most trials tested branded extracts, and they are not interchangeable. Of the 7 trials in the 2021 review, three used KSM-66 (root only), two used Sensoril or Shoden (root and leaf), one used an unspecified extract and one used root granules^[4]. NIH notes that root and leaf differ in chemical composition, and that most commercial products contain root extract^[4].

Withanolides, a group of steroidal lactones, are thought to drive many of ashwagandha's effects, though lab studies suggest other compounds may contribute^[4]. The WFSBP and CANMAT taskforce named the quality and standardization of herbal products as a key reason it could not make firmer recommendations^[3].

No trial has pinned down when effects begin. The trials in the 2021 review lasted 6 to 8 weeks, and the 2025 review of patients with diagnosed disorders had a median follow-up of 8 weeks^[4][1].

Some shorter trials reported benefits at 30 days, including a Florida study of 60 stressed adults and a 60-student trial at the University of Colorado, Colorado Springs^[4]. Others ran 60 or 90 days^[4].

Standard medications also take time. NIMH says SSRIs and SNRIs may take several weeks to start working and buspirone needs 3 to 4 weeks to be fully effective^[8]. No trial reviewed here compared ashwagandha's onset with those drugs.

Ashwagandha appears well tolerated for up to about 3 months, but its safety over many months or years is unknown^[4]. Common side effects are usually mild: stomach upset, loose stools, nausea and drowsiness^[4]. The 2025 patient meta-analysis found no significant difference from comparators in safety or tolerability^[1].

Liver injury

NCCIH describes liver injury from ashwagandha supplements as rare but documented in a number of cases^[5]. In one case series, 5 people aged 21 to 62 who took 450 to 1,350 mg a day for 1 week to 4 months developed signs such as jaundice, itching and high bilirubin^[4].

People in these reports improved after stopping the supplement, and some needed medical treatment^[4]. A 2017 report from Japan described a 20-year-old man who developed liver dysfunction after taking ashwagandha together with several antianxiety drugs^[4].

Thyroid and hormones

In a trial of 50 people with subclinical hypothyroidism, 300 mg of root extract twice a day for 8 weeks lowered TSH and raised T3 and T4 compared with placebo^[4]. Three case reports describe thyrotoxicosis in women aged 32 to 73 taking ashwagandha, which resolved after they stopped^[4].

Ashwagandha may also raise testosterone, which is why NCCIH says people with hormone-sensitive prostate cancer should avoid it^[5].

Who is advised to avoid it

NCCIH says ashwagandha should be avoided in pregnancy and while breastfeeding, and is not recommended before surgery or for people with autoimmune or thyroid disorders^[5]. It may interact with drugs for diabetes and high blood pressure, immunosuppressants, sedatives, anticonvulsants and thyroid hormone^[5].

Regulators abroad have gone further. Denmark banned ashwagandha in 2023, and France's food safety agency advised against it in 2024 for pregnant and breastfeeding women and people with endocrine disorders^[4]. The American Herbal Pharmacopoeia, whose 2000 monograph was one source for the pregnancy concern, says its report was misrepresented and that there is no evidence ashwagandha root causes pregnancy loss^[4].

The 7 trials in the 2021 review summarized by NIH compared ashwagandha with placebo, not with an antidepressant or with psychotherapy^[4]. That design shows whether ashwagandha outperforms placebo; it does not show how it ranks against standard care.

NIMH says GAD treatment typically involves psychotherapy, medication or both, and calls cognitive behavioral therapy (CBT) the gold-standard psychotherapy^[8]. It lists SSRIs, SNRIs, buspirone and benzodiazepines among medications that can be effective^[8]. The site's guides to therapy for anxiety, SSRIs for anxiety and anxiety medication cover those trials.

The WFSBP and CANMAT taskforce graded ashwagandha and lavender as provisionally recommended for anxiety disorders and did not recommend kava for GAD^[3]. It advised that herbal and nutrient products be used mainly as add-ons within standard care, especially for more severe illness^[3].

Combining products raises interaction questions. NIMH describes benzodiazepines as sedative medications, and NCCIH lists sedatives among the drugs ashwagandha may interact with^[8][5]. Neither NIH fact sheet names SSRIs or SNRIs as known interactions, which reflects a lack of data rather than proof of safety.

Ashwagandha stands apart from nutrient supplements in that guideline. The taskforce found that no nutraceutical, magnesium included, had enough evidence for any anxiety-related disorder apart from N-acetylcysteine for OCD-related disorders^[3]. The site's review of magnesium for anxiety covers those trials.