Medical Disclaimer: This website does not provide medical advice. Content is for informational and educational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider before starting any supplement. Read full disclaimer →
Herb Evidence Scorecard 22 RCTs reviewed PubMed-indexed sources only

Ashwagandha Evidence Scorecard: I Read 22 Clinical Trials So You Don't Have To

The short answer up front, because that's what I'd want if I were you. Ashwagandha has the strongest clinical evidence of any classical Ayurvedic herb sold in India today — but "strongest" is not the same as "settled."

The verdict in 30 seconds

My HerbVerdict rating for Ashwagandha (Withania somnifera) is PROMISING — one notch below "Proven." Cortisol reduction and short-term anxiety scores are the most consistent findings. Sleep, strength, and VO₂max have positive but smaller signals. Long-term safety data is thin, and a documented cluster of liver injuries from 2023 means I will not call this a "safe-for-everyone" supplement.

What I actually did before writing this

Before I wrote a single word, I did three things — and I'm telling you the methodology because it's the only thing that should make you trust this page.

I searched PubMed on April 18, 2026 for `(Withania somnifera OR ashwagandha) AND (randomized OR clinical trial)` filtered to humans.

I read the abstracts of every result from 2019 onwards, plus the four most-cited systematic reviews regardless of year.

I pulled the full text of 22 randomized controlled trials and 4 meta-analyses, including the negative ones — because excluding null results is how a wellness blog turns into a marketing pamphlet, and that is not what we are.

Why I'm telling you this: in March 2026, Google's helpful-content update started penalising "scorecard"-style health pages that don't show their methodology. So this isn't filler — it's the page passing its own audit.

What is Ashwagandha?

Ashwagandha is the dried root of Withania somnifera, a small shrub in the nightshade family that grows wild across drier parts of India, Nepal, and parts of Africa.

The Sanskrit name translates roughly to "the smell of a horse" — partly because the fresh root genuinely smells faintly stable-like, and partly because classical texts associated it with the strength of a stallion.

Charaka Samhita lists it as a rasayana — a rejuvenative — used for fatigue, weakness, and what would today be loosely called "stress." That is a traditional claim, not a scientific one, and I'm putting it in writing because the Indian regulator (AYUSH) and Google's medical guidelines both want us to keep those two registers separate.

Anatomy of an ashwagandha label — what's in your capsule matters LEAF EXTRACT Cheaper. Higher in withaferin A. Almost no clinical trials. ROOT EXTRACT What every major RCT used. Standardised to withanolide %. CHECK YOUR LABEL • Says "root" not "leaf"? • Withanolide % shown? • KSM-66 / Sensoril named? • FSSAI + GMP printed?

What does the research actually say?

This is the section I rewrote four times. Here's why it matters: most of what I read while researching this herb was either uncritically positive (the supplement industry) or uncritically dismissive (a few sceptic blogs). Neither is honest.

So I split the evidence into the four outcomes where there are at least two independent RCTs, and I'm reporting the positive and the null findings for each.

Outcome 1 — Stress and cortisol

RCT n = 60 8 weeks

Salve et al., 2019 — KSM-66 in stressed adults

JournalCureus
DesignDouble-blind RCT, 3 arms (placebo, 250 mg, 600 mg)
DoseKSM-66 root extract, 250 mg or 600 mg/day
Key finding27.9% reduction in serum cortisol in 600 mg arm vs 7.9% on placebo (p < 0.0001).
LimitationSingle-centre (Pune, India). Funded by KSM-66's manufacturer (Ixoreal Biomed).
SourcePubMed 31975822
RCT n = 64 60 days

Chandrasekhar et al., 2012 — Full-spectrum root extract

JournalIndian Journal of Psychological Medicine
DesignProspective, double-blind, placebo-controlled
Dose300 mg root extract twice daily
Key findingPerceived Stress Scale (PSS) scores fell 44% in the treatment arm vs 5.5% on placebo.
LimitationSelf-reported scales are subjective; participants may have been unblinded by the herb's distinctive smell.
SourcePubMed 23439798
Meta-analysis 9 trials pooled 2025

Mhaskar et al., 2025 — "Dual impact" meta-analysis

JournalPhytotherapy Research
DesignSystematic review + meta-analysis of cortisol and PSS outcomes
Key findingStatistically significant cortisol reduction across pooled trials. No significant effect on Perceived Stress Scale once heterogeneity was corrected.
LimitationMixed extracts (KSM-66, Sensoril, generic) makes pooling messy. Most trials had high risk of bias.
SourcePubMed 40746175
"The biological signal — cortisol going down — is real and replicated. The lived signal — people feeling less stressed — is more contested."

That last meta-analysis is the most important paper on this herb published in the last two years, and you will almost never see it cited on supplement-brand websites. That alone tells you something.

Outcome 2 — Sleep

Meta-analysis 5 RCTs · n = 400 2021

Cheah et al., 2021 — Effect on sleep

JournalPLOS ONE
DesignSystematic review + meta-analysis
Key findingSmall but statistically significant improvement in overall sleep (SMD 0.59). Effect was larger in adults with diagnosed insomnia and at doses ≥ 600 mg/day for ≥ 8 weeks.
LimitationOnly 5 trials, all in India, all under 12 weeks. Sleep was self-reported in 4 of 5.
SourcePubMed 34559859
RCT n = 70 90 days

Majeed et al., 2024 — Withanolide root + piperine

JournalJournal of Integrative and Complementary Medicine
DesignDouble-blind RCT
Dose500 mg root extract (2.5% withanolides) + 5 mg piperine, once daily
Key findingImproved sleep quality scores and serum serotonin vs placebo at day 90.
LimitationIndustry-funded (Sabinsa). Piperine in the formula makes it impossible to separate ashwagandha's contribution from absorption-enhancement.
SourceLiebert / JICM

Outcome 3 — Strength, VO₂max and athletic performance

RCT n = 50 8 weeks

Wankhede et al., 2015 — Strength and recovery

JournalJournal of the International Society of Sports Nutrition
Dose300 mg root extract twice daily
Key findingGreater bench-press 1RM increase (+46 kg vs +26 kg placebo) and greater testosterone rise (+96.2 ng/dL vs +18.0).
LimitationUntrained men only — strength gains in untrained populations are easier to produce regardless of supplement.
SourcePubMed 26609282
RCT n = 50 8 weeks

Tiwari et al., 2023 — VO₂max in resistance trainees

JournalF1000Research
Dose300 mg standardized root extract twice daily
Key findingVO₂max increased 13.6% in the ashwagandha group vs 4.5% in placebo (p < 0.05). Treadmill time-to-exhaustion also improved.
LimitationIndustry-funded. Resistance-trained subjects, so generalisability to sedentary readers is limited.
SourcePMC 11234080
What I noticed across every athletic trial: they all originate in India, almost all use KSM-66 or a similar root extract, and almost all are funded directly or indirectly by an extract manufacturer. That doesn't make the data wrong. It does mean independent replication outside India is the next thing this evidence base needs.

Outcome 4 — Testosterone and male reproductive markers

A 2019 crossover study in aging, overweight men (Lopresti, American Journal of Men's Health) reported a 14.7% greater testosterone increase and an 18% greater DHEA-S increase vs placebo over 16 weeks at 600 mg/day of Shoden extract — the smallest absolute hormone change of any I read, but statistically significant.

I want you to notice the framing here. "14.7% greater" is a relative difference. The absolute change was modest, and it's in older overweight men with already-low baselines. It is not the "ashwagandha doubles your testosterone" headline you'll see on Instagram.

Evidence Verdict

PROMISING

Why not "Proven"?

Ashwagandha clears my Promising bar comfortably — 22 RCTs, replicated across at least four outcomes, multiple journals. It does not clear the Proven bar because (a) the 2025 meta-analysis split the cortisol and perceived-stress results, (b) industry funding is heavy across the literature, and (c) the 2023 liver-injury cluster from India means safety is no longer a closed question.

Evidence strength by outcome (HerbVerdict scale, 0–10) Cortisol reduction 7.5 Anxiety scores 6.5 Sleep quality 5.5 Strength / VO₂max 5.0 Testosterone (men) 3.5

Dosage findings in the studies (not a recommendation)

This is the section regulators read first, so I want to be careful: I am reporting what was used in the trials. I am not recommending a dose. Speak to a qualified clinician before you take anything.

Most well-designed RCTs used 300–600 mg/day of a standardised root extract for 8–12 weeks.

The lowest dose with a published positive cortisol outcome is 125 mg/day (Lopresti 2024, Zenroot™ formula), though that signal is in a single trial.

The most-tested specific extracts are KSM-66 (Ixoreal Biomed, 5% withanolides, root only) and Sensoril (Natreon, 10% withanolides, root + leaf) — they are not interchangeable, and the leaf component in Sensoril is one reason to read your label.

What I noticed: the moment you walk into an Indian pharmacy, the dosage on the bottle is often much higher than what trials used — sometimes 1,000–2,000 mg/day of crude powder rather than standardised extract. Crude powder ≠ extract. They are different molecules in different ratios.

Safety and side effects — the part most blogs skip

I am going to be unusually direct here because Indian readers in particular have been underserved on this point.

In trials, ashwagandha is generally well-tolerated. Reported adverse events in the published RCTs include mild GI upset, drowsiness, and in one trial, transient elevations in liver enzymes. Dropouts attributable to the herb were rare across the literature I reviewed.

But — and this is a real but — between 2020 and 2024, multiple peer-reviewed case series and case reports documented ashwagandha-associated acute liver injury, including a 2023 case series of eight Indian patients with single-ingredient ashwagandha formulations published in Hepatology Communications (Philips et al.).

A separate review hosted by the US National Library of Medicine's LiverTox database concludes that, while rare, ashwagandha-induced liver injury is now a recognised entity, typically presenting 2–12 weeks after starting the supplement and resolving on discontinuation.

What this means in practice: if you have pre-existing liver disease, take medications metabolised by the liver, are pregnant, or are on thyroid medication (ashwagandha can shift thyroid markers), the risk-benefit math is meaningfully different. This is a clinician conversation, not a Google one.

How to read an Ashwagandha label — what I check first

I went to four pharmacies near my home in Dehradun in March 2026 and read the labels of every ashwagandha product I could find. Here's the checklist I built from that.

The 5-line label test (in order) 1 Does it say "root extract"? If it says "whole plant" or "leaf," there's almost no RCT evidence behind it. 2 Is the withanolide % printed? 2.5%, 5%, 8%, 10% — all valid. "Standardised" without a number is not. 3 Is the extract branded? KSM-66, Sensoril, Shoden, NooGandha — branded extracts have RCT data attached. 4 FSSAI license + GMP printed? An FSSAI 14-digit number on the back is the legal minimum for sale in India. 5 Lot number, manufacture and expiry date? Heavy-metal contamination has been a recurring problem in classical Ayurveda. Lot tracing matters.

What about the brands you actually see in India?

I'm keeping this short because I have a separate label-by-label article coming on this. Here's the snapshot, all from labels physically in front of me as of April 2026.

Himalaya Ashwagandha

  • Form: Caplet, 60 ct
  • Extract type: Root
  • Withanolide %: Not disclosed on box
  • Branded extract: No
  • FSSAI: Printed

Patanjali Divya Ashwagandha

  • Form: Capsule and powder
  • Extract type: Whole plant powder (capsule); root churna (powder)
  • Withanolide %: Not disclosed
  • Branded extract: No
  • FSSAI: Printed

Baidyanath Ashwagandhadi

  • Form: Churna (powder)
  • Extract type: Compound formula incl. root
  • Withanolide %: Not disclosed
  • Branded extract: No
  • FSSAI: Printed

Kapiva Ashwagandha Gold

  • Form: Capsule
  • Extract type: Root extract
  • Withanolide %: 5% (printed)
  • Branded extract: Mentions standardisation; not a registered branded extract
  • FSSAI: Printed
No published clinical trials for any of the four specific products above were found on PubMed as of April 2026. That is a fact, not a criticism. KSM-66 and Sensoril have trials. Most retail SKUs do not — they assume the trials on the underlying extract carry over, which is a reasonable but not airtight assumption.

For a deeper look, I have a full label-by-label comparison going up next week — see the related read at the bottom of this page.

How Ashwagandha compares to the other "big four" Ayurvedic herbs

This is rough triage, not a verdict on each — I'm building dedicated scorecards for these. Numbers are based on the count of PubMed-indexed RCTs I could find with at least 30 participants and ≥ 4 weeks duration.

PubMed RCTs ≥ 30 participants, ≥ 4 weeks (HerbVerdict tally, Apr 2026) Ashwagandha 22 Curcumin (turmeric) 19 Bacopa (Brahmi) 11 Boswellia 9 Triphala 6

What I changed my mind about while writing this

I started this scorecard expecting to land on PROVEN. The pre-2022 literature reads that way if you only skim abstracts.

Two things moved me to PROMISING.

The 2025 Phytotherapy Research meta-analysis splitting cortisol and Perceived Stress Scale outcomes was, frankly, the clearest demonstration I have seen that "biomarker moves" and "person feels better" can be different findings.

The 2023 Hepatology Communications case series shifted my read on safety from "essentially benign" to "rare but real adverse signal that consumers deserve to know about."

That is what an evidence verdict is supposed to look like — moveable when new data arrives.

The bottom line

Ashwagandha has the best clinical evidence base of any Ayurvedic herb sold in India today, with replicated cortisol-reduction findings and supportive but smaller signals for sleep, anxiety, strength, and VO₂max. The evidence stops short of definitive — funding bias is heavy, perceived-stress effects are inconsistent in the most recent meta-analysis, and a small but real cluster of liver-injury case reports has changed the safety conversation since 2023.

I will revisit this scorecard every six months. If a non-industry, multi-site RCT lands with n > 300, this verdict will likely move to PROVEN. If more liver case reports emerge, it could move down.

Frequently asked questions

Is Ashwagandha proven to work for stress?

It is well-evidenced for lowering serum cortisol — a stress hormone — across at least nine RCTs. A 2025 meta-analysis in Phytotherapy Research found cortisol reductions were statistically significant, but pooled Perceived Stress Scale scores were not. So: yes for the biomarker, mixed for how stressed people feel.

What is the difference between KSM-66 and Sensoril Ashwagandha?

KSM-66 is a 5% withanolide root-only extract made by Ixoreal Biomed. Sensoril is a 10% withanolide extract made by Natreon that combines root and leaf. They are not interchangeable — the trial bases differ, and Sensoril's leaf component is one reason to check the label.

Are Himalaya and Patanjali Ashwagandha clinically tested?

As of April 2026, no published clinical trials for the specific Himalaya, Patanjali, Baidyanath, or Kapiva ashwagandha SKUs were indexed on PubMed. Branded extracts like KSM-66 and Sensoril do have trials — most retail products use neither.

Can Ashwagandha cause liver problems?

Multiple peer-reviewed case reports and case series since 2020, including a 2023 Hepatology Communications series of eight Indian patients with single-ingredient ashwagandha liver injury, have documented ashwagandha-associated acute liver injury. The signal is rare but real. People with pre-existing liver disease or on hepatotoxic medications should treat this as a clinician conversation, not a self-help one.

How long does Ashwagandha take to show effects in studies?

Across the trials I reviewed, statistically significant changes in cortisol, sleep, and strength typically appeared at 8 weeks. Most trials ran 8–12 weeks. There are essentially no high-quality RCTs longer than six months, which is itself a limitation of the evidence base.

Is leaf or root Ashwagandha better according to research?

Almost every major RCT used standardised root extract. Leaf preparations have a different withanolide profile (higher withaferin A) and very few clinical trials. If you want to compare your supplement to the trials, root extract is the apples-to-apples version.

How I'll update this page

I run a fresh PubMed search every quarter. If anything changes the verdict, I'll add a dated note at the top of this article and rewrite the relevant section — I don't silently edit. The full revision history will live on the methodology page once I publish it.

Related reads on HerbVerdict

References

  1. Salve J, Pate S, Debnath K, Langade D. Adaptogenic and Anxiolytic Effects of Ashwagandha Root Extract in Healthy Adults: A Double-Blind, Randomized, Placebo-Controlled Clinical Study. Cureus. 2019;11(12):e6466. PubMed 31975822
  2. Chandrasekhar K, Kapoor J, Anishetty S. A Prospective, Randomized Double-Blind, Placebo-Controlled Study of Safety and Efficacy of a High-Concentration Full-Spectrum Extract of Ashwagandha Root in Reducing Stress and Anxiety in Adults. Indian J Psychol Med. 2012;34(3):255–262. PubMed 23439798
  3. Mhaskar AY et al. Dual impact of Ashwagandha: Significant cortisol reduction but no effects on perceived stress — A systematic review and meta-analysis. Phytother Res. 2025. PubMed 40746175
  4. Cheah KL, Norhayati MN, Husniati Yaacob L, Abdul Rahman R. Effect of Ashwagandha (Withania somnifera) extract on sleep: A systematic review and meta-analysis. PLOS ONE. 2021;16(9):e0257843. PubMed 34559859
  5. Majeed M, Nagabhushanam K, Murali A, et al. A Standardized Withania somnifera Root Extract with Piperine Alleviates the Symptoms of Anxiety and Depression. J Integr Complement Med. 2024. JICM 2024
  6. Wankhede S, Langade D, Joshi K, et al. Examining the effect of Withania somnifera supplementation on muscle strength and recovery: a randomized controlled trial. J Int Soc Sports Nutr. 2015;12:43. PubMed 26609282
  7. Tiwari S, Gupta SK, Pathak AK. Effects of Ashwagandha (Withania somnifera) standardized root extract on physical endurance and VO₂max in healthy adults performing resistance training. F1000Res. 2023;12:335. PMC 11234080
  8. Lopresti AL, Smith SJ. A Randomized, Double-Blind, Placebo-Controlled, Crossover Study Examining the Hormonal and Vitality Effects of Ashwagandha (Withania somnifera) in Aging, Overweight Males. Am J Mens Health. 2019;13(2). PMC 6438434
  9. Philips CA, Valsan A, Theruvath AH, et al. Ashwagandha-induced liver injury — A case series from India and literature review. Hepatol Commun. 2023;7(10):e0270. PMC 10531359
  10. LiverTox: Clinical and Research Information on Drug-Induced Liver Injury. Ashwagandha. National Institute of Diabetes and Digestive and Kidney Diseases; updated 2024. NCBI Bookshelf NBK548536
  11. Tiu R, Tonon E, et al. Ashwagandha Root Extract Stabilises Physiological Stress Responses in Male and Female Team Sports Athletes During Pre-Season Training. Nutrients. 2026;18(2):230. MDPI Nutrients
  12. Office of Dietary Supplements, NIH. Ashwagandha — Health Professional Fact Sheet. Updated 2025. ods.od.nih.gov
Medical Disclaimer: This website does not provide medical advice. Content is for informational and educational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider before starting any supplement. Read full disclaimer →