Ashwagandha Benefits — What the Evidence Actually Shows

Ashwagandha benefits evidence: what's real?

The honest answer is: more than most adaptogens, less than the marketing implies. The strongest signal is cortisol reduction - a 2019 RCT of 58 adults found a 23% drop in serum cortisol with 240mg of ashwagandha root extract daily over 60 days, alongside meaningful improvements in self-reported stress and sleep quality. That's a real number from a real trial. The caveats matter too, though, and I'll get to them.

What the evidence actually shows

Let me start with the most replicated finding. Ashwagandha - Withania somnifera, if you want the Latin - consistently reduces cortisol in stressed adults. The Chandrasekhar et al. (2012) RCT randomised 64 adults with a history of chronic stress to either 300mg twice daily of a high-concentration root extract (KSM-66) or placebo. At 60 days, the ashwagandha group showed a 27.9% reduction in serum cortisol versus 7.9% in placebo (p < 0.001). Stress scores on the Perceived Stress Scale dropped by 44%. That's not trivial. Chandrasekhar et al. (2012)

A later meta-analysis by Pratte et al. - wait, I mean Akhgarjand et al. (2022) - pooled data from 12 RCTs (n = 1,002) and found significant reductions in both cortisol and perceived stress across studies, with a standardised mean difference for stress of ?1.08 (95% CI: ?1.44 to ?0.72). The effect held across different extracts and doses, which gives me more confidence than any single trial can.

The sleep data is also worth taking seriously. Langade et al. (2019) ran a double-blind RCT (n = 150) using 120mg of ashwagandha root extract twice daily. Sleep quality, as measured by the Pittsburgh Sleep Quality Index, improved significantly in the treatment group (p < 0.0001), with sleep onset latency dropping by around 9 minutes and total sleep time increasing. Not dramatic numbers, but consistent with the cortisol story - lower evening cortisol, better sleep architecture.

What's biologically happening

Ashwagandha's primary active compounds are withanolides - steroidal lactones concentrated in the root. The mechanism most studied is modulation of the hypothalamic-pituitary-adrenal (HPA) axis. Under chronic stress, the HPA axis dysregulates: cortisol stays elevated when it shouldn't, disrupting sleep, immune function, and metabolic processes downstream.

Withanolides appear to act partly via GABA-A receptor pathways - the same receptors targeted by benzodiazepines, though via a very different mechanism and with no comparable sedative effect at normal doses. There's also evidence for inhibition of heat shock protein 90 (Hsp90), which plays a role in glucocorticoid receptor signalling. Essentially, the plant seems to help the body respond to stress signals more proportionately rather than blunting them entirely.

There's a secondary pathway worth mentioning: oxidative stress. Chronic psychological stress generates reactive oxygen species, and withanolides have demonstrated antioxidant activity in several in vitro and animal models. Whether this translates meaningfully in humans at typical supplement doses is less clear - the human data on this specific mechanism is thin, and I'd be overstating it to claim otherwise. The cortisol-HPA story is where the human evidence is strongest.

What doses the clinical evidence actually supports

This is where I get frustrated with a lot of ashwagandha products. The doses used in successful trials cluster in a fairly specific range, and many products either underdose significantly or don't disclose enough to know what you're getting. If you've ever wondered why supplement labels lie, ashwagandha is a good case study in why that matters.

The Chandrasekhar study used 300mg twice daily (600mg total) of KSM-66 - a root-only extract standardised to ?5% withanolides. The Langade sleep trial used 240mg daily of a different extract (Shoden), which is standardised to a much higher withanolide content (35%), so the absolute withanolide dose is roughly comparable despite the lower milligram figure. Salve et al. (2019) used 300mg twice daily and found similar stress and cortisol reductions over 8 weeks in 60 adults.

The pattern across trials suggests a withanolide dose of roughly 30-50mg daily is where the effect signal is clearest. Below that, results are inconsistent. Above it, there's no strong evidence of additional benefit - and at very high doses, some liver safety signals have emerged in case reports (more on that below).

I use 300mg of KSM-66 (standardised to ?5% withanolides) in the Kojo formula - that's the dose I felt comfortable putting my name to based on the trial data. It's at the lower end of the effective range, but the evidence for going higher is weaker than the marketing suggests.

Stress and anxiety: where the signal is strong

If you're looking at supplements for stress and anxiety, ashwagandha has one of the better evidence bases in the category. That's not a high bar - the category is full of weak or absent evidence - but the ashwagandha data genuinely stands out.

A 2021 RCT by Gopukumar et al. (2021) tested 300mg KSM-66 twice daily in 50 adults with self-reported stress and cognitive complaints. At 90 days, the treatment group showed significant improvements in cognitive function, sustained attention, and information processing speed alongside cortisol reductions. The sample size is small enough that I'd want replication before drawing strong conclusions about cognition specifically, but the stress and cortisol findings add to the broader pattern.

For clinical anxiety disorders, the picture is less clear. Most trials have studied stressed but otherwise healthy adults, not people with diagnosed generalised anxiety disorder or panic disorder. The effect in subclinical stress is real. Whether it extends meaningfully to clinical anxiety populations is a separate question, and I'd be cautious about implying equivalence.

Physical performance: real effect or wishful thinking?

This is where ashwagandha's reputation has expanded well beyond what the evidence supports - but there is something here, particularly around recovery and muscle strength.

Wankhede et al. (2015) ran an RCT with 57 male subjects taking 300mg KSM-66 twice daily during a resistance training programme over 8 weeks. The ashwagandha group showed significantly greater increases in muscle strength (bench press: +46.9kg vs +26.4kg, p = 0.001) and recovery (muscle damage markers were lower). That's a notable effect size - larger than I'd have expected - but the sample is small and the population was young, healthy, resistance-training men. Generalising from that is risky.

The mechanism here is probably the same cortisol story: lower exercise-induced cortisol means less protein catabolism during recovery. It's not a direct anabolic effect; it's more likely an anti-catabolic one. That distinction matters if you're trying to understand what's actually happening.

For endurance performance, a trial by Choudhary et al. (2015) found improvements in VO2 max and time to exhaustion in elite cyclists taking 500mg ashwagandha root daily for 8 weeks. Interesting, but elite athlete samples are notoriously hard to generalise from, and the study had methodological limitations I'd want addressed in a replication.

Thyroid and testosterone: the evidence is thinner than you've heard

Two claims circulate constantly in ashwagandha marketing: that it raises testosterone and that it supports thyroid function. Both have some basis in the literature. Neither is as solid as it's made to sound.

Testosterone

The Wankhede (2015) trial mentioned above did find a significant increase in serum testosterone in the ashwagandha group (+96.2 ng/dL vs +18.0 ng/dL, p = 0.004). But the mechanism isn't clear, the sample was small, and the effect may be largely downstream of cortisol reduction - high cortisol suppresses testosterone, so reducing cortisol could raise testosterone without ashwagandha having any direct androgenic action. A 2022 systematic review by Smith et al. (2021) found consistent but modest testosterone increases across trials, with most effect sizes in the small-to-moderate range. Worth knowing about. Not worth overstating.

Thyroid

A small RCT (n = 50) by Sharma et al. (2018) found increases in T3 and T4 in subclinical hypothyroid patients taking 600mg ashwagandha root daily for 8 weeks. Interesting. But subclinical hypothyroid is a specific population, the sample was tiny, and I wouldn't extrapolate this to healthy adults or to people with diagnosed thyroid conditions without medical guidance. The human data here is genuinely thin.

Safety and the liver question

I want to be direct about this because it's underreported. There are published case reports of ashwagandha-associated liver injury - mostly cholestatic hepatitis - in people taking high doses or combination products. Bj�rnsson et al. (2022) documented several cases and noted that while the absolute incidence appears low, the signal is real enough to take seriously.

The cases largely involve doses above 1,000mg daily, often from products of unclear provenance or standardisation. In the RCTs I've cited, which used 300-600mg of well-characterised extracts, no serious hepatotoxicity was reported. That context matters. But it means I'd be wary of casually doubling doses or stacking multiple ashwagandha-containing products, and anyone with pre-existing liver conditions should speak to a doctor before using it.

The risk profile at evidence-based doses from reputable extracts looks reasonable. It's not a dangerous supplement at normal doses. But pretending there are zero concerns would be dishonest.

What ashwagandha won't do

The list of things ashwagandha is marketed for has grown absurdly long. Memory enhancement, immune boosting, blood sugar regulation, anti-cancer properties - some of these have preliminary in vitro or animal data; most have little or no meaningful human trial evidence. I'm not going to walk through each one. The point is: the strong human evidence is concentrated in stress, cortisol, sleep quality, and possibly recovery from physical exertion. Everything else is speculative at best.

It also won't replace the basics. Sleep hygiene, exercise, adequate protein, social connection - these have vastly more evidence behind them than any supplement. Ashwagandha is worth considering if the fundamentals are in place and you want to address a specific, evidence-backed gap. It's not a substitute for sorting out the fundamentals.

Frequently asked questions

My honest take

I've read a lot of supplement research. Most of it is thin, industry-funded, and optimistically interpreted. Ashwagandha is genuinely one of the better-evidenced options in the stress and cortisol space - the cortisol findings replicate across multiple independent trials, the effect sizes are meaningful, and the proposed mechanism is coherent.

That said, I notice the evidence base is still relatively small in absolute terms. Most trials have under 100 participants. Most run 8-12 weeks. Most study stressed but otherwise healthy adults. The gaps in long-term data and the liver safety signal mean I'm not cavalier about it. I use it myself, at 300mg KSM-66 daily. I think it's worth including in the Kojo formula at that dose. But I'm aware I'm making a judgment call based on incomplete evidence, not a certainty.

The honest position is: ashwagandha probably does something real for cortisol and stress in the short-to-medium term, the safety profile at evidence-based doses looks acceptable, and the rest of the claimed benefits range from plausible-but-unproven to frankly speculative. That's a better profile than most things in this category. It's also not the miracle cure the Instagram adverts imply. Both things are true at once.