Can ashwagandha lower cortisol? What the 2026 evidence says

Nobody buys ashwagandha to optimise a hormone chart. They buy it for the nights when the body stays switched on, for the lag between stress and sleep, for the hope that one supplement might quiet the whole cycle without starting a different problem. That is the useful way into the 2026 evidence, because the cortisol data are tidier than the lived experience they are supposed to represent.

On paper, the headline is straightforward. The 2026 systematic review and meta-analysis in Planta Medica pooled 23 randomized trials involving 1,706 participants and found that ashwagandha lowered cortisol overall. That is the clearest signal in the literature right now. It still does not mean every ashwagandha product lowers cortisol, or that a lower cortisol reading automatically produces a calmer day.

Placed next to that result, a 2025 meta-analysis indexed on PubMed sounds more restrained: cortisol fell, but perceived stress did not clearly improve. That tension drives the piece. If the endocrine marker moves and the questionnaire score does not, the real question is where the benefit shows up, in whom, and under what trial conditions.

What the 2026 evidence actually found

What makes Fornalik’s review useful is its discipline. The authors were not summarising vibes or influencer anecdotes. They pooled randomized data on hormonal outcomes, including cortisol, testosterone and thyroid markers, across different adult populations and different preparations of Withania somnifera.

In Planta Medica, Fornalik et al. stated the core result plainly: “Ashwagandha significantly reduced cortisol (SMD =- 1.18, p < 0.04).”

On its own, though, a standardized mean difference of -1.18 does not settle much. The trials ran from short interventions to multi-week programs. Doses ranged from modest standardized extracts to gram-level root preparations. Some studies enrolled otherwise healthy but stressed adults; others focused on narrower clinical groups. A meta-analysis can show that a signal exists. It cannot make the interventions interchangeable.

Taken at face value, the 2026 review is biomarker first and product second. It supports the claim that ashwagandha can lower cortisol under trial conditions. It does not support the stronger retail claim that any capsule labelled ashwagandha will reproduce the same effect.

Lower cortisol is not the same as feeling less stressed

Biology and lived experience rarely line up so neatly. Cortisol is not a mood score. It follows a daily rhythm, shifts with sleep, exercise and blood-draw timing, and can improve before a participant would say the week felt materially easier.

That is why one line in the recent literature matters so much. Albalawi et al. wrote in BJPsych Open: “No significant impact was observed on perceived stress.”

In the 2025 meta-analysis in BJPsych Open, the authors covered 488 participants across seven cortisol studies and six perceived-stress studies. Their conclusion was not that ashwagandha does nothing. It was that the cortisol signal looks more repeatable than the perceived-stress signal once the studies are pooled. For readers used to hearing the herb pitched as a broad stress fix, that is the more sober answer.

Single trials add some texture. A 2025 randomized double-blind trial led by Pradeep Kumar Prajapati reported improvements in stress, anxiety, sleep and serum cortisol with an ashwagandha root powder preparation. A 2026 three-arm AshwaSR trial on PMC also found benefits on stress, anxiety, sleep quality and serum cortisol in healthy stressed adults using 150 mg and 300 mg sustained-release capsules. Those are not trivial results. They also narrow the likely beneficiary: not the generic wellness shopper, but adults who already look stressed or sleep-disrupted at baseline.

Sleep data lean the same way. In a 2021 sleep meta-analysis on PMC, Cheah et al. found the clearest effects in adults with insomnia, in studies lasting at least eight weeks, and at doses of 600 mg a day or more of the studied extracts. The authors wrote that “the effects on sleep were more prominent in the subgroup of adults diagnosed with insomnia, treatment dosage >=600 mg/day, and treatment duration >=8 weeks.” That still is not dosing advice. It is a sign that the positive trials clustered in more specific use cases than the supplement market usually admits.

Extract standardization is the real translation problem

Here the translation problem is hard to ignore. Ashwagandha is a botanical umbrella, not a single intervention. The trials behind the stronger claims often used branded or standardized extracts such as KSM-66, Shoden or AshwaSR, each with its own manufacturing method and withanolide profile. A powder, gummy or blend on a store shelf may share the plant name without sharing the clinical evidence.

Within the 2026 review, dose exposures ranged from very small standardized amounts to gram-level preparations. That alone shows why extract-to-extract comparison remains the field’s biggest weakness. If one study uses a root-only extract, another uses a sustained-release capsule, and a third uses a different standardization target, readers cannot simply average the numbers and treat them as a buying guide.

For readers, the practical takeaway is duller than the label copy but more accurate. Trial-backed formulations matter more than the front-of-label promise. Anyone considering supplementation should talk with a clinician before starting, especially people with thyroid disease, liver disease, autoimmune conditions, or medications with sedating or endocrine-active effects.

Who may benefit, and who should be cautious

The best-supported answer is narrower than the sales pitch but stronger than blanket dismissal. Ashwagandha probably can lower cortisol in some adults, especially those who begin a study stressed, sleep-impaired or both. The evidence is weaker for the broader promise that it will make most people feel noticeably less stressed in everyday life.

Safety follows the same pattern: somewhat reassuring in the short term, but not clean enough to ignore. A 2026 safety review in Pharmaceuticals found standardized root-only extracts were generally well tolerated in healthy adults, with biomarkers staying within expected ranges across the included trials. But a 2026 clinical review in Nutrients also noted that the supplement is not risk-free, especially once thyroid signalling, drug interactions and rare liver injury case reports enter the picture. Natural is not a safety category.

The cleanest answer to the headline is a qualified yes. The 2026 evidence suggests ashwagandha can lower cortisol. It does not yet support a generic claim that it reliably lowers felt stress across products and populations. The literature fits a tighter frame: biomarker-first benefits, clearer signals in stressed or insomnia-affected adults, real uncertainty around product equivalence, and enough safety caveats that self-experimentation should not be treated as casual.

References

1. Fornalik M, et al. Hormonal modulation with Withania somnifera: systematic review and meta-analysis of randomized-controlled trials. Planta Medica. 2026. Full text
2. Albalawi R, et al. Dual impact of Ashwagandha: significant cortisol reduction but no effects on perceived stress, a systematic review and meta-analysis. BJPsych Open. 2025. PubMed
3. Prajapati PK, et al. A randomized double-blind study on the efficacy of Ashwagandha root powder ghan (AF-43). International Journal of Research in AYUSH. 2025. Full text
4. Thanawala S, et al. Efficacy and safety of Ashwagandha root extract sustained-release capsules in healthy stressed adults: a randomized, double-blind, placebo-controlled, parallel-group trial. Medicine. 2026. PMC
5. Cheah KL, et al. Effect of ashwagandha extract on sleep: a systematic review and meta-analysis. PLOS One. 2021. PMC
6. Back to the roots: safety and tolerability of standardised Ashwagandha root extract in healthy adults, a systematic review of biomarkers and adverse events. Pharmaceuticals. 2026. Full text
7. The clinical implications of Ashwagandha (Withania somnifera L.) with a special reference to side effects: a review. Nutrients. 2026. Full text