That's the general answer. Your stack is different.
Check your whole stack
Withania somnifera · KSM-66 / Sensoril / Shoden
CONDITIONALGenuinely reduces cortisol and improves sleep in stressed populations — but most of what's sold is inert root powder, and the safety signals are serious enough to rule it out for large swaths of the population.
Ashwagandha genuinely reduces cortisol 23–28% and improves sleep quality (SMD −0.59) in stressed populations via HPA axis modulation and GABAergic pathways. Testosterone benefit is real — but only in stressed or overtrained males. Women: no androgen effect. Healthy, low-stress individuals: near-zero benefit (the adaptogenic ceiling is the actual pharmacology).
Extract standardization is the critical quality variable. Shoden has 118–267× higher AUC than generic 2.5% extracts. Most retail products cannot reach therapeutic plasma concentrations from any clinical trial.
Idiosyncratic hepatotoxicity (dozens of DILI cases 2017–2023, one liver transplant) and thyrotoxicosis risk in healthy individuals prevent unconditional recommendation. SEVERE interactions with thyroid medications and immunosuppressants.
Ashwagandha is marketed as a complete adaptogen — a botanical that simultaneously lowers stress, boosts testosterone, builds muscle, improves sleep, sharpens focus, and balances hormones. Brands position it as a safe, plant-based alternative to prescription anxiolytics, suitable for daily long-term use by virtually everyone.
These claims aren't fabricated — they're stripped of critical context. The word marketers never use: homeostatic. Ashwagandha normalizes dysregulated systems. If your system isn't dysregulated, it does very little.
| Claimed Benefit | Effect / Study | Verdict |
|---|---|---|
|
Cortisol reduction
STRONG
What would change this: Large RCT (N>300) in low-stress healthy adults showing ≥15% cortisol reduction would remove the population gate. |
−23 to −28% serum cortisol Lopresti 2019 (N=60); Majeed 2023 (N=141) |
Works — in stressed populations |
|
Sleep quality
STRONG
What would change this: Significant PSQI improvement in healthy sleepers (not just insomnia cohort) would make this unconditional. |
SMD −0.59 (CI −0.75 to −0.42) Cheah 2021 meta (N=400, 5 RCTs) |
Works — especially stress-induced insomnia |
|
Anxiety / subjective stress
STRONG
What would change this: Replication failure in ≥3 independent trials using HAM-A or PSS. |
HAM-A and PSS significant reductions Majeed 2023 (N=141, 3-arm RCT) |
Works consistently |
|
Testosterone (stressed men)
MODERATE
What would change this: Significant testosterone increase in unstressed men with normal baseline levels. |
+96.2 ng/dL vs +18.0 placebo Wankhede 2015 (N=57, trained males) |
Works — stressed/trained males only |
|
Testosterone (women)
DEBUNKED
This claim is refuted. No androgen effect in women — p=0.989. Mechanism doesn't apply. |
No significant change (p=0.989) Lopresti 2019 (N=60, women) |
Does not apply |
|
Muscle strength (RT synergy)
MODERATE
What would change this: Significant strength gains in sedentary individuals without resistance training. |
Bench press +46.0 vs +26.4 kg Wankhede 2015 (N=57, requires RT) |
Conditional — requires resistance training |
|
VO2max / aerobic performance
MODERATE
What would change this: Null meta-analysis from pre-registered multi-site RCTs with objective VO2max measurement. |
Favorable Bayesian probability Bonilla 2021 meta (N=615, 12 trials) |
Promising |
|
Thyroid (subclinical hypothyroidism)
MODERATE
What would change this: Replication in larger RCT (N>200) with 12-month follow-up under medical supervision. |
TSH −17.4%, T4 +19.6% Sharma 2018 (N=50) |
Works — supervised medical use only |
|
Cognition / memory
EMERGING
What would change this: Consistent, pre-registered improvement across ≥3 validated cognitive battery trials. |
Mixed results across multiple cognitive batteries Multiple small RCTs |
Insufficient evidence |
|
Weight loss
WEAK
What would change this: Dedicated RCT with weight loss as primary endpoint showing ≥3kg greater loss vs placebo over 12 weeks. |
Secondary finding in stress cohorts only | Unproven as primary endpoint |
Withanolides structurally mimic endogenous corticosteroids and exert mild negative feedback on the hypothalamus and pituitary — reducing the signals that trigger cortisol release. Stressed people see cortisol drop 23–28%. Unstressed people don't. This isn't a side effect of the research — it's the core pharmacology. Ashwagandha normalizes dysregulated systems. You cannot normalize what is already normal.
Specific withanolides and triethylene glycol bind to GABA-A receptors — the same receptors targeted by benzodiazepines — dampening CNS excitability and facilitating deeper NREM sleep with shorter sleep onset latency. This is why ashwagandha can improve stress-induced insomnia without the dependency profile of prescription sedatives, and also why it carries additive sedation risk when combined with benzodiazepines.
Two routes: (a) Cortisol and testosterone share pregnenolone as a precursor. Chronic HPA overdrive steals pregnenolone toward cortisol synthesis, suppressing testosterone production. Reducing cortisol frees the precursor pool. (b) Withanolide antioxidant activity protects testicular Leydig cells, enhancing their sensitivity to luteinizing hormone. Both routes explain why benefit only appears in men whose testosterone is already suppressed — there is nowhere to pull a healthy baseline higher.
| Population | Dose | Timing | Form | Source |
|---|---|---|---|---|
| Chronically stressed adults | 300–600 mg/day | Split: AM + PM | KSM-66 (5%) or Sensoril (10%) | Lopresti 2019; Majeed 2023 |
| Sleep-specific (insomnia / stress) | 250–600 mg | 1–2 hours pre-bed | Sensoril (10%) or KSM-66 | Cheah 2021 meta |
| Athletes (strength / hypertrophy) | 600 mg/day | Pre- or post-workout | KSM-66 root-only | Wankhede 2015; Bonilla 2021 |
| Older adults 50+ (hormonal support) | 240–600 mg/day | Morning | Shoden (35%) or KSM-66 | Lopresti 2019 |
| Low-dose high-bioavailability | 120–240 mg/day | Morning | Shoden (35% withanolide glycosides) | Lopresti 2019; PK 2025 |
| Subclinical hypothyroidism | 600 mg/day | Morning | Root extract — medical supervision ONLY | Sharma 2018 |
CYP3A4 induction increases immunosuppressant clearance — risks organ rejection or autoimmune flare. Ashwagandha is also immunostimulatory — direct pharmacodynamic conflict with transplant immunosuppression. Absolute contraindication.
Ashwagandha stimulates T3/T4 production. With levothyroxine: iatrogenic thyrotoxicosis risk. With antithyroid medications: directly opposed mechanisms — ashwagandha promotes thyroid activity while methimazole suppresses it. If used, mandatory TSH monitoring, but avoidance is the safer default. Documented thyrotoxicosis cases in previously healthy individuals confirm this is a real clinical risk.
Additive CNS depression via GABA-A agonism — excessive daytime somnolence, potential respiratory depression at higher sedative doses. Avoid concurrent use or reduce sedative dose under medical supervision.
Additive hypoglycemic effect — ashwagandha independently reduces blood glucose. Monitor blood glucose on initiation; antidiabetic dose may need adjustment.
Potential additive hypotensive effects — symptomatic blood pressure drops possible on initiation. Monitor blood pressure.
Putative CYP3A4 induction may alter anticoagulant metabolism — altered INR, increased thrombotic or bleeding risk. INR monitoring required; consult prescriber before use.
| Side Effect | Incidence | Dose-Related? | Management |
|---|---|---|---|
| GI distress (nausea, loose stools) | 5–10% at 600mg/day | Yes — worse >1,000mg | Take with food; reduce dose; divide dosing |
| Somnolence / excessive drowsiness | Uncommon at standard doses | Yes | Move dose to evening; avoid concurrent sedatives |
| Drug-induced liver injury (DILI) | Rare — dozens of cases 2017–2023, one liver transplant | No — idiosyncratic | Discontinue immediately on jaundice, dark urine, pruritus, or unexplained fatigue. Urgent medical review. |
| Thyrotoxicosis (palpitations, weight loss, fever) | Rare — case reports in healthy individuals | Unknown | Discontinue; urgent medical assessment; usually self-resolves |
| Headache | Uncommon | Unclear | Usually resolves within first week of use |
' width='800' height='400'/%3E%3C/svg%3E)
| Form | Effective Daily Dose | Monthly Cost | Food Alternative |
|---|---|---|---|
| Shoden (35%) | 120–240 mg/day | £20–35 | None — no food source |
| KSM-66 (5%) | 600 mg/day | £15–25 | None |
| Sensoril (10%) | 250–500 mg/day | £12–20 | None |
| Prolanza (5% SR) | 300 mg once daily | £25–40 | None |
| Generic root powder | 600 mg/day (ineffective) | £5–10 | N/A — therapeutic dose unreachable |
No food alternative exists — withanolides are not present in meaningful quantities in any dietary source. Supplement is the only delivery vehicle.
Cortisol reduction, sleep quality in stressed populations, and anxiety/stress reduction — unanimous HPA axis dampening across diverse RCTs, distinct extracts, and multiple validated biomarkers.
Testosterone in stressed/trained men, strength with RT synergy, aerobic performance — real effects with clear population gates and solid mechanistic rationale.
Idiosyncratic hepatotoxicity (dozens of cases, one liver transplant), thyrotoxicosis in healthy individuals, zero long-term RCT data beyond 12 weeks, complete dependence on pharmaceutical-grade extracts most consumers never purchase.
A 12-month, multi-center, double-blind, placebo-controlled RCT (N=500) tracking KSM-66 600mg/day continuously vs cycled, with 4-week interval ALT/AST/ALP/bilirubin monitoring (to quantify DILI incidence), systematic TSH/T3/T4 tracking in healthy adults, and longitudinal HPA habituation assessment. Negative results on hepatic and thyroid endpoints would push conviction to HIGH. Significant incidence rates would push to LOW or AVOID. Until that data exists, MODERATE is the correct call — strong efficacy, unresolved long-term safety.
How strong is the evidence for the claims in this review? Higher = more confidence the claims are supported. This does not measure how large the effect is or how important it is compared with other levers.
Is this worth your time, money, effort, risk, and trust for this goal? Different from Verdict Score (evidence strength) and Leverage Map (relative importance) — Action ROI is the worth-it call once friction is priced in.
Evidence-scored dosing, timing, forms, and who should skip it. One page, no fluff.
Get the protocolConviction-scored verdicts on supplements, nutrition, training, physio, and recovery.