Ashwagandha: cortisol, mood, skin, hair, heart, kidneys and liver
Does ashwagandha lower cortisol and stress?
The answer
In short trials ashwagandha lowered cortisol and stress scores, and a review of 14 trials in mental disorders (mostly anxiety) found less anxiety, with depression less certain. We found no human trial for tension headaches, stress acne or gut. It has been linked to liver injury, deaths in people with liver disease, and one kidney-transplant rejection. Ask a pharmacist first.
Do
- Ask a pharmacist or doctor before starting if you have liver disease, or a thyroid or autoimmune condition, or take diabetes, blood-pressure, sedative, seizure, thyroid or immune-suppressing medicines: NCCIH (US) says it might interact with those medicines and advises against it with thyroid or autoimmune conditions.Clinical guidance NCCIH (US), "Ashwagandha: Usefulness and Safety"
- Get medical advice if you notice signs of liver trouble such as yellow skin or eyes (NCCIH, US, notes rare liver injury cases).Clinical guidance NCCIH (US), "Ashwagandha: Usefulness and Safety"
- Tell your doctor or pharmacist about any herbal product you take: the UK's NHS says herbal medicines can stop medicines working or cause harm.Clinical guidance NHS (UK), "Herbal medicines and complementary therapies"
Avoid
- Taking it if you already have liver disease: in an Indian case series, five of eight had existing liver disease, and the three who developed acute-on-chronic liver failure all died.Tested Philips 2023
- Taking it while pregnant or breastfeeding: NCCIH (US) says to avoid it in pregnancy and not to use it while breastfeeding.Clinical guidance NCCIH (US), "Ashwagandha: Usefulness and Safety"
- Taking it after an organ transplant without asking your transplant team: one report describes acute rejection that needed the transplanted kidney removed after ashwagandha use, and NCCIH (US) says it might interact with immune-suppressing medicines.Tested Sriperumbuduri 2020
- Taking it for more than about three months on the assumption that it is safe: NCCIH (US) says there is not enough information on long-term safety.Clinical guidance NCCIH (US), "Ashwagandha: Usefulness and Safety"
What families do
Tradition and common belief, not evidence.
Ashwagandha is sold as an adaptogen that calms stress hormones and is said to help mood, stress eating, belly fat, headaches, brain function, immunity, skin, hair, gut and heart.
The evidence
Nine short trials reviewed in 2023 found ashwagandha lowered cortisol in stressed adults, though they differed a lot and none checked long-term or adrenal effects. A 2025 meta-analysis of 14 trials in people with mental disorders found better anxiety and stress scores, with depression less certain. Single small trials in other populations reported fewer food cravings (52 and 60 people), better recall (130), and, from one clinic in India with a manufacturer-supplied product, better skin barrier and hair measures. A 40-day Mexican pilot found no change in weight or waist but lower triglycerides. On safety, a 2026 review of 23 trials in healthy adults found normal liver, kidney and cardiovascular markers and no serious events. Case reports describe liver injury (three deaths in people with existing liver disease in India), a kidney-transplant rejection and an overactive thyroid. Our search found no human trial of ashwagandha for tension headaches, stress acne, gut health, kidney or heart disease outcomes, or anything in Black participants.
This tier rests on: That ashwagandha lowers cortisol and stress scores in short trials: one systematic review of 9 differing studies, and single small trials (52 to 130 people) for cravings, memory, skin and hair. None in Black participants, and no trial for tension headaches, stress acne or gut.
Why this tier and strength: Tier B because the trials are randomised but in other populations, several of them in India, with no ethnicity reported. Strength thin because the cortisol review pools studies that differ in design, dose and duration, and the other claims rest on one small trial each, several with manufacturer-supplied products. Would move if a trial enrolled Black participants or reported by ancestry.
What has been tested
B Stressed adults in 9 clinical trials (country and ethnicity not stated in the abstract) (n=9 trials)
Systematic review of trials of Withania somnifera and cortisol, from 4,788 records screened, trials lasting 30 to 112 days.
Supplementation decreased cortisol with no significant adverse effects. The studies differed in design, dose and duration. None looked at the effect of lowering cortisol on adrenal function or at long-term effects.
Della Porta M, Maier JA, Cazzola R (2023), Nutrients, [review] Open source Study page
Note: No pooled figure in the abstract. Short trials. Ethnicity not reported.
B 360 people treated and 353 controls in 14 randomised trials, people with mental disorders (mostly anxiety; inclusion was any age), median dose 600 mg a day, median follow-up 8 weeks (n=713)
Systematic review and meta-analysis, frequentist and Bayesian.
Ashwagandha improved anxiety, depression, stress and sleep quality compared with comparators. The Bayesian analysis confirmed the anxiety result but not the depression result. The authors list few studies, large differences between studies and outlier effects as limits. No significant difference in safety or tolerability.
Marchi M, Grenzi P, Travascio A, Uberti D et al (2025), BJPsych Open, [review] Open source Study page
Note: Depression finding is the weakest part. Ethnicity not reported in the abstract.
B 52 adults under chronic stress (country and ethnicity not stated in the abstract) (n=52)
Randomised, double-blind, placebo-controlled, 300 mg root extract twice a day for 8 weeks.
Perceived stress, food cravings, eating behaviour, serum cortisol, body weight and BMI all improved compared with placebo. The extract was reported safe and tolerable.
Choudhary D, Bhattacharyya S, Joshi K (2017), Journal of Evidence-Based Complementary & Alternative Medicine, [primary] Open source Study page
Note: Small. The size of the weight change is not in the abstract. Waist circumference was not reported.
B 60 healthy college students, Colorado Springs, US (ethnicity not stated in the abstract) (n=60)
Randomised, double-blind, placebo-controlled, 700 mg full-spectrum root extract a day for 30 days.
Restorative sleep scores were better on ashwagandha (58.4 against 48.2, adjusted for starting sleep) and food cravings were lower than on placebo by the end. The overall test for sleep and stress together was not significant; the authors say stress differences usually need about six weeks.
O'Connor J, Lindsay K, Baker C, Kirby J et al (2022), Journal of Medicinal Food, [primary] Open source Study page
Note: Short trial in young adults. A correction notice was published (erratum 2023).
B 130 healthy stressed adults aged 20 to 55 (125 completed); registered in the Clinical Trials Registry-India; ethnicity not stated (n=130)
Randomised, double-blind, placebo-controlled, 300 mg sustained-release extract once a day for 90 days.
Recall memory scores improved and pattern recall errors fell compared with placebo. Perceived stress, serum cortisol and sleep quality scores improved and happiness scores rose.
Gopukumar K, Thanawala S, Somepalli V, Rao TSS et al (2021), Evidence-Based Complementary and Alternative Medicine, [primary] Open source Study page
Note: One branded product. Healthy stressed adults, not people with memory problems.
B Healthy adults (number not stated in the abstract), 30-day blinded period then 30-day open-label extension (n=not stated in the abstract)
Randomised, double-blind, placebo-controlled, 60 mg Withania somnifera extract.
Blood levels of several immunoglobulins, cytokines and immune cell counts (including NK cells) rose in the ashwagandha group. No adverse events reported. The abstract reports lab markers, not fewer infections.
Tharakan A, Shukla H, Benny IR, Tharakan M et al (2021), Journal of Clinical Medicine, [primary] Open source Study page
Note: Lab markers only, no infection outcome, sample size not in the abstract. A rise in a marker is not shown to mean better protection.
B Adults aged 18 to 60 with hair loss or skin conditions (number not stated in the abstract), one clinic in Hyderabad, India (n=not stated in the abstract)
Randomised, double-blind, placebo-controlled, 300 mg standardised root extract twice a day for 75 days.
Greater improvement than placebo in transepidermal water loss (skin barrier), in hair density, growth and anagen:telogen ratio, and in skin and hair quality-of-life scores. No adverse events. The extract was supplied by the manufacturer, Ixoreal Biomed.
Jillella A, Keerthi N, Venkateswar Reddy J (2026), Phytotherapy Research, [primary] Open source Study page
Note: One clinic, manufacturer-supplied product, sample size not in the abstract. Ethnicity and hair type not stated, so nothing is known about afro-textured hair or skin of colour.
B Healthy adults aged 18 to 45 with mild to moderate hair loss (61 analysed), India (n=61)
Randomised, double-blind, placebo-controlled, topical ashwagandha root-extract serum for 75 days.
Hair shedding fell in the 60-second comb test and hair density, growth and thickness improved compared with placebo at day 75.
Yerram C, Jillella A, Reddy V (2023), Journal of Ayurveda and Integrative Medicine, [primary] Open source Study page
Note: A serum on the scalp, not a capsule. Hair type not stated.
B 43 adults with overweight or obesity, Mexico (the abstract reports 17 control and 21 intervention) (n=43)
Double-blind, randomised, placebo-controlled pilot, 500 mg a day for 40 days with a monitored diet.
No significant change in weight, BMI or waist circumference. Triglycerides and VLDL cholesterol fell. The authors call for longer studies.
Ballesteros-Torres JM, Vazquez-Aguilar A, Luzardo-Ocampo I, Escalante-Aburto A et al (2025), Nutrition & Metabolism, [primary] Open source Study page
Note: Pilot, 40 days. Participant numbers in the abstract do not add up to 43.
B 80 healthy adults (40 men, 40 women), Lucknow, India (n=80)
Randomised, double-blind, placebo-controlled safety study, 300 mg root extract twice a day for 8 weeks.
No untoward effects in any treated volunteer. Blood counts, biochemistry including liver tests, vital signs including blood pressure, and thyroid hormones showed no significant abnormal change.
Verma N, Gupta SK, Tiwari S, Mishra AK (2021), Complementary Therapies in Medicine, [primary] Open source Study page
Note: Healthy people, 8 weeks. Ethnicity not stated.
B 2,317 healthy adults in 23 randomised trials of standardised, single-ingredient root extract (125 to 600 mg a day, up to 180 days) (n=2317)
Systematic review of safety and tolerability.
Liver, kidney, blood, hormone and cardiovascular markers stayed within normal ranges. No serious adverse events attributable to the extract. Mild effects (stomach discomfort, headache, drowsiness) were infrequent and occurred in both treatment and comparison groups. Long-term data beyond 180 days are limited.
Coope OC, Willems MET, Levington A, Tallon MJ et al (2026), Pharmaceuticals (Basel), [review] Open source Study page
Note: Authors say case reports of harm mostly involved non-standardised or multi-ingredient products. Some authors are affiliated with supplement companies. Healthy adults only, so it says little about people with liver or kidney disease.
B 8 patients with liver injury from single-ingredient ashwagandha products (of 23 with ashwagandha-related injury), multiple centres in India, 2019 to 2022 (n=8)
Retrospective case series with work-up to exclude other causes and chemical analysis of the product.
Cholestatic hepatitis was the commonest presentation. Five had existing chronic liver disease; three of them presented with acute-on-chronic liver failure and all three died. In the others the injury was prolonged but resolved on its own, and one progressed to chronic injury. The product contained only natural plant chemicals, with no adulteration found.
Philips CA, Valsan A, Theruvath AH, Ravindran R et al (2023), Hepatology Communications, [primary] Open source Study page
Note: A harm signal from a handful of cases; it cannot say how common this is. The existing ashwagandha card describes a smaller series in which no one developed liver failure.
B One 69-year-old male kidney transplant recipient, Ottawa, Canada (n=1)
Case report.
He developed acute rejection requiring removal of the transplanted kidney after using ashwagandha, while taking tacrolimus, mycophenolate and prednisone. The authors recommend warning transplant recipients about herbal products including ashwagandha.
Sriperumbuduri S, Umar MS, Lajoie-Starkell G, Fairhead TR et al (2020), Kidney International Reports, [primary] Open source Study page
Note: One case. It cannot show that ashwagandha caused the rejection. I read the introduction and case presentation, not the whole discussion.
B One 47-year-old previously healthy Japanese man (n=1)
Case report.
Painless thyroiditis with an overactive thyroid two months after starting ashwagandha; symptoms and thyroid blood tests improved after he stopped.
Hayashi M, Hamada H, Azuma SI, Hayashi K (2024), Cureus, [primary] Open source Study page
Note: One case. The 2018 India trial in the existing ashwagandha card reported better thyroid tests in people with an underactive thyroid, so the thyroid evidence points both ways.
B 9,685 US adults in the NHANES survey, 2017 to March 2020 (n=9685)
Nationally representative cross-sectional survey of herbal and supplement use.
Ashwagandha-containing products were used by 28 of the survey adults. An estimated 15.6 million US adults had used at least one of six botanicals with a liver-injury link (turmeric, green tea extract, ashwagandha, Garcinia cambogia, red yeast rice, black cohosh) in the past 30 days, similar to the number taking the cholesterol drug simvastatin (about 14 million) or NSAID painkillers (about 14.8 million).
Likhitsup A, Chen VL, Fontana RJ (2024), JAMA Network Open, [primary] Open source Study page
Note: Shows how common use is, not how often harm happens. I did not read any breakdown by race.
Harm
Signal: Case series link ashwagandha to cholestatic liver injury; in an Indian series three people with existing liver disease developed acute-on-chronic liver failure and died. One kidney-transplant rejection and one overactive-thyroid case are reported after use. NCCIH (US) calls liver injury rare, says it might interact with several medicines, and advises against it in pregnancy, breastfeeding, thyroid or autoimmune conditions and before surgery. A 2026 review of 23 trials in healthy adults found no serious adverse events with standardised extract.
Measured variables
- Serum cortisol (measured)
- Perceived stress, food-craving and happiness scores (self-rated)
- Memory test scores (CANTAB)
- Skin water loss and hair density (measured, one clinic)
- Liver, kidney and thyroid blood tests (harm outcomes)
- No ancestry variable in any source
Not tested, and gaps
Searches 17 Nov 2026 (PubMed via E-utilities, titles read, abstracts opened for everything cited). (1) (ashwagandha OR "Withania somnifera") AND ("African American" OR Black OR African OR Caribbean): 40 records, top 8 titles read, none a study in Black participants (matches were on African plants and hepatotoxic botanical exposure); the existing ashwagandha card's narrower search returned 0. Control for the method: vitamin D AND African American AND randomised returned 153 records. (2) Topic searches with ashwagandha: cortisol 28 records; depression 34; eating, craving or weight 31; headache or migraine 12 (none on tension headache; "tension headache" alone returned 0); cognition 46; gut, irritable bowel, gastrointestinal 52 and a narrower gut search 9 (only dog, constipation-blend and safety papers in the titles read; the constipation blend trials were listed and not opened); hair 11; skin, acne, eczema 45 (the one oral skin and hair trial, Jillella 2026, was opened); heart, blood pressure, lipids 13; immune 175 and a narrower immune search 21; kidney 37; liver 110; acne 5 (titles only, none a trial of acne). Untested in my search: any benefit or harm in Black participants; ashwagandha for tension headaches, stress acne, gut, heart-disease outcomes or kidney disease; effect on afro-textured hair or dark skin; long-term (over about six months) safety; the cravings and skin findings have not been repeated in my search. Not opened: a 2026 review of ashwagandha-associated liver injury, a 2024 retrospective chart review of adaptogen and antidepressant interactions, the 2025 dose-response meta-analysis on mental health (its reported effect sizes looked implausibly large, so I did not use it), and the NIH Office of Dietary Supplements fact sheet. The thyroid and transplant findings are single case reports. ClinicalTrials.gov (API v2, intervention ashwagandha AND African American OR Black): 5 records, titles only read, none looked like an ashwagandha trial in Black participants; the same intervention search with no population term returned 88 trials, so the method works.
All sources
- Della Porta M, Maier JA, Cazzola R (2023), Nutrients, [review] Open
- Marchi M, Grenzi P, Travascio A, Uberti D et al (2025), BJPsych Open, [review] Open
- Choudhary D, Bhattacharyya S, Joshi K (2017), Journal of Evidence-Based Complementary & Alternative Medicine, [primary] Open
- O'Connor J, Lindsay K, Baker C, Kirby J et al (2022), Journal of Medicinal Food, [primary] Open
- Gopukumar K, Thanawala S, Somepalli V, Rao TSS et al (2021), Evidence-Based Complementary and Alternative Medicine, [primary] Open
- Tharakan A, Shukla H, Benny IR, Tharakan M et al (2021), Journal of Clinical Medicine, [primary] Open
- Jillella A, Keerthi N, Venkateswar Reddy J (2026), Phytotherapy Research, [primary] Open
- Yerram C, Jillella A, Reddy V (2023), Journal of Ayurveda and Integrative Medicine, [primary] Open
- Ballesteros-Torres JM, Vazquez-Aguilar A, Luzardo-Ocampo I, Escalante-Aburto A et al (2025), Nutrition & Metabolism, [primary] Open
- Verma N, Gupta SK, Tiwari S, Mishra AK (2021), Complementary Therapies in Medicine, [primary] Open
- Coope OC, Willems MET, Levington A, Tallon MJ et al (2026), Pharmaceuticals (Basel), [review] Open
- Philips CA, Valsan A, Theruvath AH, Ravindran R et al (2023), Hepatology Communications, [primary] Open
- Sriperumbuduri S, Umar MS, Lajoie-Starkell G, Fairhead TR et al (2020), Kidney International Reports, [primary] Open
- Hayashi M, Hamada H, Azuma SI, Hayashi K (2024), Cureus, [primary] Open
- Likhitsup A, Chen VL, Fontana RJ (2024), JAMA Network Open, [primary] Open
- NCCIH (US), "Ashwagandha: Usefulness and Safety", [official], practical advice (clinical guidance), opened 2026-11-17 Open
- NHS (UK), "Herbal medicines and complementary therapies", [official], practical advice (clinical guidance), opened 2026-11-17 Open
Common questions
Does ashwagandha lower cortisol and stress?
In short trials ashwagandha lowered cortisol and stress scores, and a review of 14 trials in mental disorders (mostly anxiety) found less anxiety, with depression less certain.
Does ashwagandha help with stress eating or stress belly?
In short trials ashwagandha lowered cortisol and stress scores, and a review of 14 trials in mental disorders (mostly anxiety) found less anxiety, with depression less certain.
Is ashwagandha good for your skin, hair, gut or immune system?
We found no human trial for tension headaches, stress acne or gut.
Is ashwagandha safe for your liver, kidneys or heart?
It has been linked to liver injury, deaths in people with liver disease, and one kidney-transplant rejection.
Does ashwagandha help with depression or tension headaches?
We found no human trial for tension headaches, stress acne or gut.
Last checked against its sources: . Next check due: 17 May 2027.
This is information, not medical advice. If you are worried, see a doctor or pharmacist. How we grade the evidence.