In plain terms: Ashwagandha (standardized root extract) has some human trial evidence supporting it, but the evidence is not yet strong enough for EBL to call it established. It is worth knowing about and worth discussing with your clinician.
Ashwagandha (standardized root extract)
The evidence as the record states it
Stress: Chandrasekhar et al. (Indian J Psychol Med, 2012; PMID 23439798): 64 adults, a full-spectrum root extract at 300 mg twice daily (600 mg/day) for 60 days; all stress scales and serum cortisol fell versus placebo. Lopresti et al. (Medicine (Baltimore), 2019; PMID 31517876): 60 adults, a DIFFERENT standardized extract at 240 mg/day, below this record's dose range, for 60 days; HAM-A fell (P = .040) and morning cortisol fell, but the DASS-21 did not differ significantly from placebo (P = .096) and the testosterone change was not significant versus placebo. Pooled: Arumugam et al. (Explore, 2024; PMID 39348746), 9 trials, 558 patients, Perceived Stress Scale, HAM-A and cortisol all improved; Albalawi et al. (Nutr Health, 2025; PMID 40746175), 7 trials on cortisol and 6 on perceived stress, cortisol fell 1.16 micrograms/dL but the Perceived Stress Scale did not change (SMD -0.36, P = 0.40). Sleep: Cheah et al. (PLoS One, 2021; PMID 34559859), 5 trials, 400 participants, small effect on sleep (SMD -0.59), larger in diagnosed insomnia, at 600 mg/day or more and 8 weeks or longer; Langade et al. (J Ethnopharmacol, 2021; PMID 32818573), 80 adults, sleep onset latency and sleep quality improved, more in those with insomnia. Strength: Wankhede et al. (J Int Soc Sports Nutr, 2015; PMID 26609282), 57 men, strength and testosterone rose; Ziegenfuss et al. (Nutrients, 2018; PMID 30463324), 38 men, a root-and-leaf extract at 500 mg/day, squat and bench press gains larger than placebo, funded by that extract's maker; Lee et al. (Nutrients, 2026; PMID 42738987), meta-analysis of the resistance-training trials, found the strength effect lost significance under conservative adjustment and rated the certainty low. Safety: Björnsson et al. (Liver Int, 2020; PMID 31991029), five cases of cholestatic liver injury with jaundice attributed to ashwagandha supplements in Iceland and the US Drug-Induced Liver Injury Network, resolving in 1 to 5 months. Grade B: replicated randomized evidence on self-reported and biomarker endpoints, with the pooled analyses split on perceived stress. Funding: the extract in Chandrasekhar et al. and Wankhede et al. was supplied by its maker; Lopresti et al. was funded by the maker of the extract it tested, which also supplied it, with the two Indian principal investigators declaring no competing interest and the first author declaring past study funding from the same company; Ziegenfuss et al. was funded by its extract's maker. No trial behind the letter was run without an extract maker's money or product.
Why Grade B and not Grade A
Grade B reflects moderate human evidence -- typically a single strong trial, multiple smaller trials, or surrogate endpoint data without hard clinical outcome replication. To reach Grade A, Ashwagandha (standardized root extract) would require at least one additional independent randomized trial demonstrating a hard clinical endpoint or objectively measured functional outcome in a comparable population.
What would move this grade
An independent randomized trial demonstrating a hard clinical endpoint -- mortality, major adverse event, or objectively confirmed functional improvement -- in an adequately powered sample, with results independently replicated, would support Grade A.
How this was studied
Form: Standardized ashwagandha (Withania somnifera) extracts taken orally as capsules: a high-concentration full-spectrum root extract (Chandrasekhar 2012, PMID 23439798), a root extract (Wankhede 2015, PMID 26609282; Langade 2021, PMID 32818573), a different standardized extract (Lopresti 2019, PMID 31517876) and an aqueous root-and-leaf extract (Ziegenfuss 2018, PMID 30463324). The withanolide content of these extracts is not stated in the cited abstracts. · Dose: 300 mg twice daily, 600 mg/day (Chandrasekhar 2012, PMID 23439798; Wankhede 2015, PMID 26609282); 240 mg once daily (Lopresti 2019, PMID 31517876); 500 mg/day (Ziegenfuss 2018, PMID 30463324). Pooled sleep effects were larger at 600 mg/day or more (Cheah 2021 meta-analysis, PMID 34559859); the cortisol meta-analysis included oral doses of 250 mg/day or more (Albalawi 2025, PMID 40746175). The Langade dose is not stated in the cited abstract (Langade 2021, PMID 32818573). · Population: Stress: 64 adults with chronic stress (Chandrasekhar 2012, PMID 23439798) and 60 stressed, healthy adults (Lopresti 2019, PMID 31517876); pooled, 9 trials with 558 patients (Arumugam 2024, PMID 39348746) and 7 cortisol and 6 perceived-stress trials with 488 participants (Albalawi 2025, PMID 40746175). Sleep: 5 trials, 400 adults (Cheah 2021, PMID 34559859); 80 adults, 40 healthy and 40 with insomnia (Langade 2021, PMID 32818573). Strength: 57 men aged 18 to 50 new to resistance training (Wankhede 2015, PMID 26609282), 38 recreationally active men, mean age 26.5 (Ziegenfuss 2018, PMID 30463324), and 3 pooled trials with 161 healthy adults (Lee 2026 meta-analysis, PMID 42738987). · Duration: 60 days (Chandrasekhar 2012, PMID 23439798; Lopresti 2019, PMID 31517876); 8 weeks (Wankhede 2015, PMID 26609282; Langade 2021, PMID 32818573); 12 weeks (Ziegenfuss 2018, PMID 30463324). Pooled sleep effects were larger at 8 weeks or longer (Cheah 2021, PMID 34559859); the cortisol meta-analysis required trials of 2 weeks or longer (Albalawi 2025, PMID 40746175).