The Research on Ashwagandha and Stress: A Closer Look

Ashwagandha (Withania somnifera) is one of the most heavily studied adaptogens sold today, which means there are now enough individual clinical trials for researchers to pool them into larger, combined analyses — a more reliable way to look for a real, repeatable effect than reading any single trial in isolation. This entry looks at three of the most recent pooled analyses on cortisol and everyday stress, plus a fourth on sleep, and stays specifically in that lane: what the biomarker and questionnaire data actually show, not a broader wellness claim.

What three recent pooled analyses actually found

A 2025 systematic review and meta-analysis in BJPsych Open pooled 15 randomized controlled trials (873 participants total), searching PubMed, Web of Science, Scopus, and Cochrane from inception through September 2024. At the 8-week mark, ashwagandha supplementation was associated with a statistically significant drop in cortisol (p < 0.0001), a significant drop in Perceived Stress Scale scores (p = 0.0013), and a significant drop in Hamilton Anxiety Scale scores (p = 0.0053) compared with placebo. Notably, the same analysis found no statistically significant improvement in a separate quality-of-life measure (p = 0.37) — a reminder that "less stressed on a symptom scale" and "reports a better quality of life overall" aren't automatically the same result.

A second 2024 meta-analysis, published in the journal Explore, pooled 9 randomized controlled trials (558 participants) using doses that ranged fairly widely — from 125 mg to 600 mg per day, for 30 to 90 days, across both root-only and root-and-leaf extracts. Compared with placebo, it found reductions in Perceived Stress Scale scores, an anxiety-scale measure, and serum cortisol, all favoring ashwagandha.

A third, more recent 2025 systematic review and meta-analysis, published in Nutrition and Health, took a narrower, more skeptical look specifically at cortisol versus perceived stress. It pooled 7 studies reporting cortisol and 6 reporting perceived-stress scores (488 participants combined), searching four databases for trials published between January 2012 and February 2024. Cortisol came down by a statistically significant margin (p < 0.001) — but the pooled perceived-stress result was not statistically significant (p = 0.40), meaning this particular analysis could not confirm that people, on average, reported feeling less stressed even though their measured cortisol had dropped.

Where sleep fits into the picture

A separate systematic review and meta-analysis in PLOS ONE pooled 5 randomized controlled trials (400 participants) looking specifically at ashwagandha extract and sleep. It found a small but statistically significant improvement in overall sleep compared with placebo, with the effect more pronounced in the subset of participants who had a diagnosed sleep difficulty, who took at least 600 mg per day, and who continued for at least 8 weeks. Shorter courses, lower doses, and healthy sleepers without an existing sleep complaint showed a smaller effect. Stress, cortisol, and sleep are obviously connected physiologically, but it's worth noting this is a distinct body of trial data, not simply a restatement of the stress findings above.

Why the results don't all agree

A few things explain why these pooled analyses land in slightly different places rather than one clean, unanimous number. First, the underlying trials use different products: doses across the studies feeding these analyses range from 125 mg to 600 mg per day, using root-only extracts in some trials and root-and-leaf blends in others, standardized to different withanolide percentages — so "ashwagandha" in this literature is really a family of related but non-identical products, not one single substance. Second, most trial durations are short by the standards of chronic-condition research — commonly 8 to 12 weeks, occasionally as brief as 30 days — so none of this speaks to what happens over a year or more of use. Third, cortisol is a lab-measured biomarker while perceived stress and anxiety scores are self-reported questionnaires, and the two don't always move in lockstep: a self-report scale is more exposed to expectation and placebo response than a blood or saliva cortisol reading, which may partly explain why the newer, more tightly screened pooled analysis found a split result where earlier ones didn't. Finally, sample sizes in the individual trials feeding all of these meta-analyses are still fairly small, which widens the confidence intervals around any single pooled number and leaves room for the picture to shift again as more trials are published.

Practical takeaways

Across this set of pooled analyses, the doses associated with a measured effect mostly cluster in the 240 mg to 600 mg per day range of a standardized extract, taken daily for at least several weeks — none of this literature supports a single dose producing an immediate change. Because the newest and narrowest analysis couldn't confirm a perceived-stress benefit even though cortisol moved, it's reasonable to expect a measurable change in a lab value without necessarily feeling dramatically different day to day, and to treat "I feel calmer" as a genuinely separate outcome worth tracking on its own rather than assuming it automatically follows a lower cortisol reading. As with any supplement in this category, checking a label for a standardized withanolide percentage and a stated extraction method (root versus root-and-leaf) makes it possible to at least roughly compare a product to what was actually studied. For a look at how one specific, widely-sold standardized extract compares to the trial doses discussed here, see our entry on NOW Foods Ashwagandha. Anyone with a thyroid condition, anyone pregnant or nursing, and anyone on regular prescription medication should check with a healthcare provider before starting it, since ashwagandha can interact with thyroid, sedative, and immunosuppressant medications.

Nothing on this site is intended to diagnose, treat, cure, or prevent any disease, and none of it is individualized medical advice. Talk to a healthcare provider before starting any new supplement, especially if you are pregnant, nursing, have a thyroid condition, or take regular prescription medication.