What are adaptogens, and which have real evidence?

The word was first defined for substances that enhance the "state of nonspecific resistance" during stress. Proposed mechanisms involve the stress-hormone axis, cortisol, nitric oxide and stress-protein pathways, but most of that evidence comes from animals and isolated cells⁴. For readers, the useful question is what each herb has shown in people.

A herb-by-herb grade

• Ashwagandha (best supported). Twelve randomized trials (1,002 adults) pooled to lower anxiety and stress scores, but results varied widely and the certainty was low¹. Five trials (400 adults) found better sleep². Case reports describe liver injury³. See our ashwagandha evidence page.
• Rhodiola (mixed). Two systematic reviews of 11 trials disagree: one finds contradictory results with every trial at risk of bias⁸, the other finds possible benefit but no independent replication⁵. See rhodiola.
• Panax ginseng (modest). Pooled trials showed a small drop in fatigue but no gain in physical performance, and the authors called the evidence insufficient⁶. See ginseng.
• Eleuthero (weak). A 96-person trial in chronic fatigue found no overall benefit⁹. See eleuthero.
• Schisandra (preliminary). A review of older studies reported better endurance and mental performance in mild fatigue¹⁰; a modern 54-person trial tested muscle strength instead¹¹. See schisandra.
• Holy basil (promising, small). A review of 24 studies reported favorable outcomes in all of them⁷, and one 100-person, 8-week trial lowered perceived stress¹². See holy basil.

What to do with this. Pick one herb, one standardized product, and one measurable goal. Basics such as sleep, movement and daylight come first, and none of these herbs replaces care for depression, anxiety disorders or exhaustion that does not lift.

Ashwagandha

Akhgarjand and colleagues pooled 12 randomized trials (1,002 adults aged 25 to 48). Ashwagandha lowered anxiety (standardized mean difference -1.55, 95% CI -2.37 to -0.74) and stress (-1.75, -2.29 to -1.22) against placebo, but heterogeneity was very high (I-squared 94% and 83%) and the authors graded the certainty of evidence low. Dose-response analysis suggested stress benefit at 300 to 600 mg a day¹. Effects this large, with this much inconsistency, usually point to small-study bias and should be read cautiously. A separate meta-analysis of five trials (400 adults) found a significant sleep improvement (SMD -0.59, 95% CI -0.75 to -0.42), clearer in people with insomnia, at 600 mg a day or more, and after eight weeks or longer. Safety data were limited². Björnsson and colleagues described five patients (mean age 43) with jaundice 2 to 12 weeks after starting ashwagandha products. Injury was cholestatic or mixed, and liver tests normalized within one to five months in four; one was lost to follow-up, and one case also involved rhodiola³.

Rhodiola

Ishaque and colleagues found 11 trials of rhodiola for fatigue (10 randomized). Two of six physical-fatigue trials and three of five mental-fatigue trials were positive, and all had a high or unclear risk of bias, so they called the evidence contradictory⁸. Hung and colleagues reviewed 11 placebo-controlled trials and judged the quality of most to be moderate or good, with only mild adverse events, but noted the lack of independent replication⁵. Both conclusions can be true: promising but unreplicated.

Eleuthero and ginseng

Hartz and colleagues randomized 96 adults with unexplained fatigue for six months or longer to Siberian ginseng or placebo; 76 gave two-month data. Fatigue fell in both arms, and the difference was not significant overall. A subgroup with less severe fatigue (n=45) improved at P=0.04, unadjusted for multiple comparisons, so this is a hint only⁹. A meta-analysis of 12 trials with 630 participants found ginseng reduced fatigue (SMD 0.34, 95% CI 0.16 to 0.52, four trials) but did not improve physical performance (SMD -0.01, 95% CI -0.29 to 0.27, eight trials). The authors judged the evidence insufficient because trials were few and small⁶.

Schisandra

The strongest claims come from a narrative review by long-time adaptogen researchers, which rated the evidence "good" for schisandra and eleuthero in mild fatigue and "strong" for one rhodiola extract¹⁰. A narrative review is not a pooled analysis, and the eleuthero verdict sits beside Hartz's null trial. A 12-week trial of 54 adults over 50 found 1 g a day of schisandra extract raised knee-extension strength by 10.2 Nm on the right (95% CI 3.7 to 16.8) and 6.7 Nm on the left (0.3 to 13.1) compared with placebo, but not muscle mass¹¹. That is not a stress outcome.

Holy basil

Jamshidi and Cohen reviewed 24 human studies and reported favorable outcomes in all and no significant adverse events⁷. A body of research in which every study is positive deserves caution about publication bias. In a randomized, double-blind trial, 100 stressed adults took 125 mg twice daily for eight weeks: perceived stress (P=0.003), insomnia scores (P=0.025) and hair cortisol (P=0.025) improved against placebo, and salivary cortisol was lower after a lab stress test¹². It is one short, single-product trial.

Where the evidence disagrees

Ashwagandha trials agree in direction but not in size. Rhodiola reviews disagree on quality. Ginseng helps fatigue scores but not performance. Mechanistic claims rest mostly on animal and cell work⁴.

Bottom line (confidence 2 of 5). Adaptogens are a convenient grouping of traditional stress herbs, not a proven pharmacological class. Ashwagandha has the most consistent human evidence, with low certainty and a liver-injury signal from case reports. For the others, expect small, uncertain effects.