Rhodiola rosea grows at high altitude and high latitude across Scandinavia, Siberia and the mountains of Asia, and its root has a long history in Russian and Scandinavian folk medicine for stamina and low mood. Soviet-era research built the adaptogen label; Western trials began around 2000, several with one standardized extract, SHR-5.

The literature is small, old and uneven. The 2012 systematic review that remains the best summary found 11 controlled trials, with two of six on physical fatigue and three of five on mental fatigue reporting benefit, and judged every one of them at high or unclear risk of bias. Newer sports trials add short-term results in trained people. It is a plausible herb with no trial of a size that would settle anything.

The constituents credited are rosavins and salidroside, and extracts are often standardized to both. A 2016 review of the antidepressant hypothesis catalogs actions on neuroendocrine, immune and neurotransmitter systems in animal and cell work, including monoamine effects and modulation of the stress axis. The one human mechanistic signal is cortisol: in the burnout trial, the cortisol response to awakening changed in the treatment group against placebo. A 2016 drug-interaction study found a commercial product reduced CYP2C9 activity by 21% over 14 days, which is a mechanism of a different kind.

Stress-related fatigue. The phase III SHR-5 trial randomized 60 people aged 20 to 55 meeting Swedish criteria for fatigue syndrome to 576 mg of extract a day or placebo for 28 days. Both groups improved on most measures (the placebo effect was large), but the extract beat placebo on the Pines burnout scale and on three attention indices, and the cortisol awakening response differed between groups. No serious side effects.

Night-shift physicians. A 2000 crossover in 56 young physicians found one low-dose tablet a day improved a composite of mental performance tests after night duty during the first two-week period, with no side effects.

Anxiety and mood. Eighty mildly anxious people were randomized to 400 mg a day of an extract or no treatment for 14 days: anxiety, stress, anger, confusion and depression scores fell against the untreated group, with no cognitive difference. There was no placebo, which the authors acknowledge.

Sport. A 2025 crossover in 27 resistance-trained adults found a week of either dose improved bench press and leg press maximums, set volume and Stroop test scores against a no-capsule control, with Wingate cycling unaffected. A 2010 crossover in 14 trained men found four weeks lowered blood lactate and creatine kinase after exhaustive exercise without changing VO2 max or time to exhaustion.

Depression. The 2016 review counted two placebo-controlled trials in 146 people with major depression and seven open-label studies in 714 with milder stress-related depression and called the antidepressant action possible.

The verdict that still stands. The 2012 systematic review: contradictory evidence, methodological flaws that limit assessment, and a need for one rigorously designed trial that has not yet been run.

Well tolerated in every trial, with no serious side effects attributed to the extract, and the 2016 review calls its short-term safety profile favorable in contrast to conventional antidepressants.

The documented interaction matters more than any side effect. A 14-day course of a commercial product reduced CYP2C9 activity by 21% in healthy volunteers, more in extensive metabolizers, which the authors flag as potentially relevant for narrow-therapeutic-index CYP2C9 drugs such as warfarin and phenytoin. Anyone on those should not add rhodiola without the prescriber knowing. Long-term safety beyond a few months is unstudied.

A dietary supplement in the United States, sold without review of its claims. In several European countries rhodiola products are registered as traditional herbal medicines for temporary relief of stress symptoms, a status based on tradition of use rather than demonstrated efficacy. Not licensed anywhere as a treatment for depression or fatigue syndrome.