Ashwagandha sits in an interesting position in the supplement world: it has more quality human RCTs behind it than most adaptogens, but the word "adaptogen" itself means almost nothing on a label. Whether the bottle you are looking at contains an extract that resembles what was used in the trials depends entirely on the extract standard listed, and most labels do not make this clear enough.
This article is structured around the practical decisions involved: which extract to choose, what the best evidence actually shows, when to expect changes, and what the honest caveats are.
Ashwagandha has real bioactive compounds and interacts with hormone systems. If you have a thyroid condition, are pregnant, take immunosuppressants, or have an autoimmune condition, please speak with your healthcare provider before starting it. Rare cases of liver injury associated with certain ashwagandha products have been reported. This is wellness information, not medical advice or a treatment recommendation for any anxiety disorder.
Which form of ashwagandha is in the research?
Withania somnifera is the botanical name. The active compounds most studied are withanolides, a class of steroidal lactones found primarily in the root. The concentration and type of withanolides varies enormously between raw powder (typically 0.1 to 0.5% withanolides) and standardized extracts (5 to 10%).
Two patented extract standards show up in most quality clinical trials:
- Root-only extraction
- Standardized to 5% withanolides
- Typical dose: 300 to 600 mg/day
- Most published RCTs (anxiety, cortisol, strength, thyroid)
- Full-spectrum root phytochemistry preserved
- Root and leaf extraction
- Standardized to 10% withanolides
- Typical dose: 125 to 250 mg/day
- Fewer but still positive trials on stress and cortisol
- Higher withanolide density per gram
If the label on your product does not say KSM-66 or Sensoril (or another named extract), you are buying unstandardized ashwagandha powder. The withanolide content may be anywhere from trace to high depending on the batch and supplier. This is not necessarily unsafe, but you cannot compare your dose to the clinical literature meaningfully.
There are other standardized extracts (Shoden, which standardizes to 35% withanolides, has some emerging trial data, but fewer studies than KSM-66 at the time of writing).
What do the cortisol RCTs actually show?
Two trials stand out for quality and are worth understanding in detail:
64 adults with a history of chronic stress were randomized to 300 mg KSM-66 extract twice daily or placebo for 60 days. The primary outcome was perceived stress scale (PSS) score. Results: the ashwagandha group showed a 44% reduction in PSS score versus 5.5% in placebo. Serum cortisol decreased by 27.9% in the supplement group versus 7.9% in placebo. Secondary outcomes included reductions in self-reported anxiety (using the GHQ-28), fatigue, and improvements in overall well-being. This is a double-blind, placebo-controlled design in a population specifically recruited for chronic stress. It is the most cited ashwagandha trial for stress and anxiety outcomes. doi:10.4103/0253-7176.106022
60 adults with sleep problems and anxiety were randomized to 300 mg KSM-66 twice daily or placebo for 8 weeks. Primary outcomes were sleep quality (Pittsburgh Sleep Quality Index, PSQI) and anxiety (Hamilton Anxiety Rating Scale, HAM-A). The ashwagandha group showed significant improvements on both measures: PSQI score improved by 72% versus 29% in placebo, and HAM-A score decreased by 41.2% versus 24.2% in placebo. Morning cortisol was also significantly lower in the supplement group at 8 weeks. This trial is notable for measuring both anxiety and sleep outcomes simultaneously and finding measurable effects on both. doi:10.7759/cureus.6093
Both trials have limitations: they are relatively small (60 to 64 participants), conducted in India where stress profiles and dietary context differ from, say, a North American population, and the manufacturers of KSM-66 were not involved in these specific trials but the extract is commercially available, which creates a potential commercial ecosystem around positive results. These are real limitations, not reasons to dismiss the studies, but they are worth naming.
A 2021 systematic review by Pratte et al. (referenced below) and a 2021 meta-analysis by Priyanka et al. in Medicine pooled multiple ashwagandha trials and found consistent directional evidence for reductions in stress, anxiety, and cortisol, with the caveat that most trials were small and used different outcome measures.
How does ashwagandha affect cortisol and why does that matter for anxiety?
Cortisol is the primary glucocorticoid released by the adrenal glands as part of the HPA axis stress response. In chronic psychological stress, morning cortisol is often elevated and the normal daily cortisol curve (high in the morning, low in the evening) becomes flatter and dysregulated.
Ashwagandha's proposed mechanism is at the level of the HPA axis. Withanolides appear to modulate the sensitivity of the axis to stress signals, reducing the cortisol output for a given stressor. There is also evidence from animal studies that withanolides inhibit the enzyme CYP11B1, which is involved in cortisol synthesis, though whether this is the primary mechanism in humans is not established.
There are also GABAergic effects: some withanolides appear to bind GABA-A receptors, similar to how benzodiazepines act but much less potently. This provides a possible direct anxiolytic mechanism distinct from the cortisol pathway.
The significance of a 27.9% reduction in serum cortisol (as seen in Chandrasekhar 2012) is context-dependent. If your cortisol is elevated due to chronic stress, reducing it toward the normal range is likely to feel better. If your cortisol is already in a normal range and your anxiety is primarily driven by cognitive or situational factors, the effect may be smaller.
The 8-week window: why response takes time
Both major trials measured their primary outcomes at 8 weeks. This is not arbitrary. The HPA axis adapts slowly, and steroid hormone regulation changes take weeks to months to stabilize.
Some people report noticing reduced physical tension or a calmer baseline within 2 to 3 weeks of daily use. This could reflect early GABA-mediated effects, or it could reflect expectation. The meaningful cortisol changes in the trials were measured at 8 weeks.
If you start ashwagandha and assess it after 2 weeks because you don't feel different, you are likely assessing too early. The evidence suggests setting an 8-week minimum window, with consistent daily dosing.
What dose and how should you take it?
Based on the trials above:
- KSM-66: 300 mg twice daily (600 mg total), taken morning and evening. This is the dose from both Chandrasekhar 2012 and Langade 2019.
- Sensoril: 125 to 250 mg daily, usually once per day, because its higher withanolide concentration means less compound is needed.
- Unstandardized powder: Cannot be meaningfully compared to the research doses.
Taking ashwagandha with food reduces the chance of gastrointestinal discomfort, which is the most common mild side effect reported in trials (and which occurred at similar rates in placebo groups in most trials, suggesting it is not always supplement-related).