Ashwagandha: what the sleep trials actually measured
A traditional botanical with a modern trial literature that is smaller, shorter and more preparation specific than the market suggests.
Key takeaways
- The trials have been small and short, mostly in adults already reporting stress or poor sleep.
- Results do not transfer between preparations, because standardised extracts differ chemically from one another.
- The evidence for anxiety is weaker than the evidence for subjective sleep and stress scores.
- It is not a treatment for depression, insomnia, thyroid disease or burnout, and there is no authorised EU health claim for it.
The randomised trials of ashwagandha that are usually cited share a shape. They recruited modest numbers of adults, often people already reporting stress or poor sleep, and they ran for weeks rather than years. Their primary outcomes were, in most cases, questionnaire scores: self reported sleep quality, perceived stress, sometimes an anxiety scale. Those instruments are legitimate research tools. They are also self reports collected in populations that had volunteered because they felt unwell, which is the setting in which expectation does most of its work.
The second complication is that these trials did not test one substance. They tested particular standardised root extracts, and standardisation describes a manufacturing target rather than equivalence between products. Different extracts differ chemically, in solvent, in plant part and in the profile of withanolides they carry. A result obtained with one preparation does not automatically transfer to another sold under the same common name. The literature is preparation specific in a way the shelf is not.
Safety is documented rather than theoretical. Case series and national regulatory reviews have described liver injury associated with ashwagandha containing products, which is why several authorities have issued advisories and why thyroid disease and pregnancy appear routinely among the listed cautions. Those cautions exist because effects on thyroid hormone measures and on pregnancy have been reported, not because caution is a formality.
In the European Union there is no authorised health claim for ashwagandha, so any specific benefit stated on a label is unlawful whatever the trial literature contains. That gap between evidence and permission is often read as regulatory obstruction. It is closer to an admission that the evidence has not been assessed as sufficient.
The quieter question sits underneath all of it. A change in a questionnaire score can be statistically detectable and still be small enough that the person filling in the questionnaire would not have noticed it in an ordinary week.
A shorter version of this piece first appeared as a LinkedIn post in July 2026.
Notes and sources
Each regulatory statement in this article is set against the document it comes from, so you can read the original rather than our summary of it.
1. No health claim for ashwagandha appears on the EU register of authorised claims
2. Liver injury has been reported in association with ashwagandha containing products
LiverTox: Clinical and Research Information on Drug Induced Liver Injury, Ashwagandha
3. A national authority has advised caution over liver safety
Danish Food Agency and Nordic assessments of Withania somnifera
