Cortisol is the number this whole category is sold on, and it's the number you should trust least. Two systematic reviews published within a year of each other, pooling largely the same ashwagandha trials, agree that cortisol drops — and reach opposite conclusions about whether people actually feel less stressed. Reishi has the mirror-image problem: the felt improvement is there and the biomarker has barely been looked at. Whichever endpoint you pick, the other one refuses to follow.
Ashwagandha reliably lowers cortisol. Albalawi's 2025 meta-analysis of seven trials found a reduction of 1.16 µg/dL (95% CI −1.64 to −0.69) but no significant effect on the Perceived Stress Scale (SMD −0.355, 95% CI −1.188 to 0.47, p = 0.40). Arumugam's 2024 review of nine trials in 558 patients found the opposite on that scale: PSS improved by 4.72 points (95% CI −8.45 to −0.99). Reishi has the reverse profile — fatigue fell 28.3% against 20.1% on placebo in 132 patients (Tang 2005), while cortisol has been measured just once, as a secondary urinary endpoint in a 26-person cardiovascular trial.
So this ranking is organised by which question you're asking, because the evidence doesn't let you answer both at once.
Why Does Everyone Measure Cortisol?
Because it's cheap, objective and easy to publish. A blood or saliva sample gives you a number, the number moves, and a moving number makes a paper. Perceived stress needs validated questionnaires, larger samples and longer follow-up to show anything.
That incentive shapes the whole literature. The endpoint that's easiest to measure became the endpoint the category advertises, and nobody checked whether it tracks the thing buyers actually want.
What Happens When You Check?
The two most recent syntheses disagree, and the disagreement is the finding. Albalawi's 2025 review in Nutrition and Health searched four databases for RCTs of at least two weeks at 250 mg/day or more, and pooled seven studies on cortisol and six on perceived stress across 488 participants. Cortisol fell significantly. Perceived stress did not, with a confidence interval that crossed zero.
Arumugam's 2024 review in Explore pooled nine RCTs and 558 patients and found the Perceived Stress Scale improved by 4.72 points, the Hamilton Anxiety Scale by 2.19, and serum cortisol by 2.58.
Same herb, overlapping trials, opposite verdicts on the measure that matters. Both reviews reported moderate heterogeneity around 50%. The reproducible finding in this literature is the biomarker; the unstable one is the human experience.
Does Reishi Fix That Problem?
It has the opposite problem. Tang's 2005 trial randomised 132 patients with neurasthenia to 5.4 grams a day of a polysaccharide extract or placebo for eight weeks. The sense of fatigue fell 28.3% from baseline against 20.1% on placebo, and 51.6% of the treatment group were rated more than minimally improved against 24.6% (p = 0.002).
Reishi's cortisol record, by contrast, is one measurement. Chu's 2012 cross-over trial gave 26 patients with borderline blood pressure or cholesterol 1.44 g/day for 12 weeks and measured urinary catecholamines and cortisol among many secondary outcomes. It wasn't a stress trial, the population wasn't stressed, and no cortisol effect appears among its reported findings.
So reishi moved how people felt without anyone establishing that it moves the hormone. Which is, for a buyer, the more useful direction to fail in.
What About the One Blend Trial?
It exists, and it's recent. Hisamuddin's 2026 trial in Brain and Behavior randomised 50 participants to a five-species mushroom blend or placebo for 12 weeks, measuring sleep quality, fatigue, anxiety, depression and serum cortisol, ACTH, noradrenaline, melatonin and CRP.
Anxiety improved more in the blend group at both timepoints — STAI-S p = 0.025 at six weeks and p = 0.011 at twelve, Hamilton Anxiety p = 0.002 at both — with depression scores down at six weeks.
Its limitation is structural: five species in one capsule means you cannot attribute the result to any of them. Our blend versus single species article runs the dose arithmetic, which comes out to roughly 50 mg of extract per species per day.
Why Do the Two Reviews Disagree?
Three things, none of them scandalous. The trials pooled aren't identical — Albalawi restricted inclusion to studies of at least two weeks at 250 mg/day or more, which filters the set differently. Both reported heterogeneity around 50%, meaning the underlying results genuinely scatter.
And "ashwagandha" isn't one intervention. Root extracts and root-plus-leaf extracts differ in composition, so pooling them assumes an equivalence that doesn't hold — the point our dosage guide makes about withanolide standardisation.
What survives all three explanations is the asymmetry. Nobody disputes the cortisol direction. The disagreement is confined to whether people noticed, and that's the half a buyer is paying for.
The Ranking
| Option | Cortisol evidence | Felt-stress evidence | Verdict |
|---|---|---|---|
| Ashwagandha | Strong and reproducible: −1.16 µg/dL across 7 trials | Contested — two 2024–25 reviews disagree | Best biomarker, unreliable feeling |
| Reishi | One secondary urinary measurement, no effect reported | Fatigue −28.3% vs −20.1%; 51.6% vs 24.6% improved | Best felt result, no biomarker |
| Five-species blend | Measured in one 50-person trial | Anxiety improved at 6 and 12 weeks (p = 0.011–0.002) | Real result, unattributable |
| Lion's mane | Not measured | 28-day stress trend only, p = 0.051 | Not a stress supplement |
| Cordyceps, chaga, Trametes | None | None | No basis |
| Amanita muscaria | None | No controlled data of any kind | Anecdote only |
What Should You Actually Buy?
- If a doctor is tracking your cortisol, ashwagandha is the one with the reproducible biomarker effect. Our dosage guide covers why root and root-plus-leaf extracts can't be dose-converted.
- If you want to feel less worn down, reishi has the better result on that specific outcome — and it's the outcome you can notice.
- If sleep is the actual problem, treat it as a sleep question, not a cortisol one. See reishi and sleep and our sleep supplement guide.
- Don't buy anything on a cortisol claim alone. The evidence above is exactly why: the hormone can move while nothing you'd notice does.
- Pick one and give it eight weeks. Tang measured at eight, Hisamuddin at six and twelve, and the ashwagandha trials ran from two weeks up.
Why Blends Are Popular Here and Shouldn't Be
Stress is the use case where blends sell hardest, because the promise is vague enough that no single ingredient has to carry it. That's also why they're the worst choice if you want to learn anything.
A five-species blend at two capsules a day divides its extract five ways. If reishi's result came from 5.4 grams and your blend delivers tens of milligrams of it, you're not running a smaller version of Tang's trial — you're running a different experiment. Our Forest Power Blend guide does that arithmetic on our own product and doesn't flatter it.
Frequently Asked Questions
Which mushroom lowers cortisol the most?
No mushroom has a reproducible cortisol result. Ashwagandha, which isn't a mushroom, does: seven pooled trials show a reduction of 1.16 µg/dL. Reishi's cortisol has been measured once, as a secondary urinary endpoint in a 26-person cardiovascular trial that reported no effect.
Does lowering cortisol make you feel less stressed?
Not dependably. Albalawi's 2025 meta-analysis found significant cortisol reduction alongside a null result on the Perceived Stress Scale (p = 0.40), while a 2024 review of overlapping trials found the scale did improve. The biomarker and the experience come apart.
Is reishi good for stress?
It has the category's best result on a felt outcome: fatigue down 28.3% against 20.1% on placebo across 132 patients, with 51.6% rated more than minimally improved versus 24.6% (p = 0.002). That's fatigue rather than stress, and in a clinical population.
How long until it works?
Eight weeks is the reasonable floor. Tang's reishi trial measured at eight weeks, the blend trial at six and twelve, and ashwagandha trials ran from two weeks upward. Judging in the first fortnight tells you very little.
Should I take a stress blend instead?
The one blend trial did show reduced anxiety at both six and twelve weeks. But five species in one capsule means no result can be attributed, and each species lands far below the doses its own trials used.
Shop Stress and Relaxation
Reishi capsules (120 caps €39, 2 × 120 €70) and dried reishi (100 g €49, 1 kg €349) cover the felt-fatigue evidence. Ashwagandha capsules (120 caps €29) are the cheapest option here and the one with the reproducible cortisol effect. Forest Power Blend (120 caps €69) is the blend route — read the dose arithmetic first. The sleep and relaxation collection holds the range. Free shipping on orders over €50.
Related Articles
- Ashwagandha vs Reishi for Stress
- Reishi as an Adaptogen: Stress Resilience
- Ashwagandha Benefits and Research
- Best Mushroom Supplement for Anxiety
- Best Mushroom Supplement for Sleep
- Reishi, Sleep Quality and GABA Signalling
- Mushroom Blend vs Single Species
Sources
- Albalawi AA. Dual impact of Ashwagandha: significant cortisol reduction but no effects on perceived stress — a systematic review and meta-analysis. Nutr Health. 2025;31(4):1395-1408. PubMed 40746175
- Arumugam V, Vijayakumar V, Balakrishnan A, et al. Effects of Ashwagandha (Withania somnifera) on stress and anxiety: a systematic review and meta-analysis. Explore (NY). 2024;20(6):103062. PubMed 39348746
- Tang W, Gao Y, Chen G, et al. A randomized, double-blind and placebo-controlled study of a Ganoderma lucidum polysaccharide extract in neurasthenia. J Med Food. 2005;8(1):53-58. PubMed 15857210
- Chu TT, Benzie IF, Lam CW, Fok BS, Lee KK, Tomlinson B. Study of potential cardioprotective effects of Ganoderma lucidum (Lingzhi): results of a controlled human intervention trial. Br J Nutr. 2012;107(7):1017-1027. PubMed 21801467
- Hisamuddin AS, Ramli F, Leo TK, et al. Adaptogenic effects of mushroom blend supplementation on stress, fatigue, and sleep: a randomised, double-blind, and placebo-controlled trial. Brain Behav. 2026;16(1):e71193. PubMed 41540766