Ashwagandha and Cortisol: What Was Measured

The measurement decides half the answer

Cortisol can be measured as total hormone in serum, as free hormone in saliva, as the fraction excreted over 24 hours in urine, or as a months-long average deposited in hair. Serum and saliva aren't interchangeable. Saliva carries the unbound fraction, which is the part that reaches receptors, while a serum assay reports total cortisol including what's bound to corticosteroid-binding globulin. A single morning blood draw adds its own problem, since cortisol peaks shortly after waking and swings widely from day to day, so one sample at one time point is a noisy estimate of a person's typical output. Diurnal slope, area under the curve across a day, or hair cortisol would all be steadier. Almost none of the ashwagandha trials used them. They took one morning sample and compared group means at day 60.

Most of the evidence rests on a single morning draw.

The two trials everyone cites

The most quoted study enrolled 64 adults with a history of chronic stress, gave them 300 mg of a root extract twice daily for 60 days, and measured serum cortisol alongside three stress questionnaires. Cortisol fell 27.9% from baseline in the extract group and 7.9% in the placebo group, a between-group difference with p=0.002. In absolute terms that was a move from about 26.2 µg/dL to 9.0 µg/dL, or roughly 720 nmol/L down to 250 nmol/L, starting from a value above the usual morning reference range. The paper declared no funding source and no conflict of interest, and also stated that the extract was supplied by Ixoreal Biomed, the company that makes it. The second trial enrolled 60 stressed adults, gave 240 mg a day for 60 days, and found morning cortisol down 23% against a 0.5% rise on placebo (p<.001). It was funded by Arjuna Natural, which manufactures the extract used.

Sixty-four people in one. Sixty in the other.

The trial that found nothing

A third study from the same research group used the same supplier's extract in a different population and got a null result. Fifty-seven overweight men aged 40 to 70 with mild fatigue were randomised to 8 weeks of extract or placebo, then crossed over to the other arm for 8 weeks, with 43 completing the full 16 weeks. Cortisol was measured in saliva. The extract group finished 7.8% lower than placebo, and the difference wasn't significant (p=0.319). Fatigue, vigour and sexual wellbeing improved in both arms with no between-group difference. The authors offered two explanations: saliva instead of serum, and a population recruited for fatigue rather than stress, some of them shift workers, whose cortisol rhythms are already disrupted.

Same extract, same lead author, different fluid, no effect.

Who paid, and how big the trials were

Follow the money and the sample sizes together, because in this literature they travel as a pair. The 60-person trial was funded by Arjuna Natural, the extract's manufacturer, and the 57-man crossover came from the same funder. The 64-person trial declared no support, while using extract donated by Ixoreal Biomed. Enrollment across the field clusters between 50 and 70 people, which means 25 to 35 per arm, which means only a large effect could be detected reliably. A 2026 meta-analysis pooled 23 randomized trials and 1,706 participants and came out at SMD -1.18 for cortisol, with a 95% interval of -2.33 to -0.03 and heterogeneity of 90.4%. Individual trials reported falls between 11% and 66% of baseline. The Egger test for publication bias came back at p=0.01, and a leave-one-out analysis showed the pooled result depended heavily on one study.

An SMD above 1 is a very large effect for a botanical. That size, in trials this small, is a warning rather than a triumph.

A lower number isn't a calmer person

Cortisol is a hormone with a diurnal rhythm, a role in blood pressure and glucose control, and a relationship to stress that runs in both directions. Perceived stress is a score on a questionnaire about how the last month felt. They correlate, weakly, and a trial can move one without moving the other. That happened inside a single study here. In the 60-person trial, cortisol dropped 23% with p<.001, the Hamilton anxiety rating improved with p=.040, and the DASS-21 total, which combines depression, anxiety and stress items, came in at p=.096. Same participants, same capsules, three outcomes, and only two of them cleared the bar. Anyone quoting the cortisol number as proof of a felt benefit is skipping the part of the same paper where the felt benefit didn't reach significance.

A number moved. That's all a number does.

Frequently asked questions

Does ashwagandha lower cortisol?

In several small trials, yes, by roughly 20% to 30% from an elevated baseline. The pooled estimate is large but comes with 90.4% heterogeneity and evidence of publication bias, so the true average effect is probably smaller than the headline.

Why did one trial find nothing?

It measured cortisol in saliva instead of serum, and it enrolled men with fatigue rather than adults selected for stress. Either difference could explain a null result, and the trial wasn't powered to tell which.

How long do the trials run?

Most run 60 days, with a range of 2 to 16 weeks across the meta-analysis. Nothing in the literature covers a year of daily use.

Is a lower cortisol reading a good thing?

Not automatically. Cortisol isn't a toxin to be minimised, and the trials enrolled people with elevated baseline values. In someone with a normal morning cortisol, a further fall isn't obviously desirable.

Who funded this research?

Mostly the companies selling the extracts. Two of the three largest trials here were manufacturer-funded, and the third used donated extract while declaring no support.

---

This article is for general information only and does not constitute medical advice. Statements about dietary supplements have not been evaluated by the Food and Drug Administration and are not intended to diagnose, treat, cure, or prevent any disease. Talk with your health care provider before using ashwagandha, especially if you take prescription medication.

Sources

• Chandrasekhar K et al., Ashwagandha root extract in reducing stress and anxiety in adults, Indian J Psychol Med 2012
• Lopresti AL et al., Stress-relieving and pharmacological actions of an ashwagandha extract, Medicine 2019
• Lopresti AL et al., Hormonal and vitality effects of ashwagandha in aging, overweight males, Am J Mens Health 2019
• Fornalik M et al., Hormonal modulation with Withania somnifera: systematic review and meta-analysis, Planta Med 2026