Benefits
Better sleep quality in adults with poor sleep
In a four-week trial in 165 adults with moderate insomnia, insomnia scores fell about one point more than on placebo on a 24-point scale when the 20 mg and 30 mg groups were pooled, though neither dose was significant alone; rated sleep quality improved from week three. In 52 adults aged 55 to 85, sleep scores and time to fall asleep on a headband sensor improved, partly because the placebo group slept worse.
Lower perceived stress alongside better sleep
In the same insomnia trial, Perceived Stress Scale scores fell about two points more than on placebo with either dose, on a 0 to 40 scale, and a brief anxiety and low mood screen improved on 30 mg only. Mood ratings, sleepiness and quality of life did not differ. Stress was a secondary outcome, and an eight-week trial in adults with low mood found no change on the same scale, so the two trials give mixed results.
Low mood and emotional well-being: modest, inconsistent findings
Two trials in healthy adults with subclinical low mood were null on their main endpoints. After eight weeks, total mood disturbance matched placebo, though a depression subscale and social relationship scores improved. After six weeks, combined depression, anxiety and fatigue scores matched placebo and only a mental health quality-of-life subscale improved. Saffron as a class has stronger mood data than this brand's own trials.
Acute stress response: a later cortisol peak, not a smaller one
In a small study of 19 young men given one 30 mg dose before a lab stress test, self-rated stress and anxiety differed across the three study arms, but only pure safranal, not the extract, beat placebo head to head. The extract delayed the cortisol and cortisone peak without lowering the peak or total cortisol output. In an eight-week trial, a single dose kept heart rate variability from dropping during a stressor.
Standardized, full-spectrum saffron with measured safranal
Batches are assayed for crocins, picrocrocin derivatives, safranal and kaempferol derivatives, and the maker says its patented process encapsulates the extract to protect volatile safranal. Crocetin, the main breakdown product of crocins, rises in users' urine and blood, including a newly identified cis form. That shows what is in the capsule and that it is absorbed, not that it works.
Mechanism of action
Serotonin and dopamine signaling (lab and animal evidence)
In mice, oral Safr'Inside reduced immobility in the forced swim test, a standard screen for antidepressant-like activity, alongside changes in brain serotonin and dopamine. Serum taken from volunteers after a dose increased serotonin and dopamine release from cultured neurons and lowered serotonin transporter expression. Neither shows the same thing happens in the human brain.
Stress chemistry and the kynurenine pathway (animal evidence)
Given before acute restraint stress, the extract blunted depressive-like behavior in mice and shifted enzymes of the kynurenine pathway, a tryptophan route linked to inflammation and mood, away from potentially neurotoxic products. A six-week human trial that measured inflammatory markers and stress-hormone reactivity found no change in either.
Safranal as a fast-acting volatile
Safranal is the aroma compound the maker standardizes most tightly. In the acute stress study, a pure safranal dose matched to 30 mg of extract lowered self-rated stress and anxiety more clearly than the extract itself, which supports safranal as one active part but leaves the extract's own acute effect uncertain.
Gut microbiota (animal and exploratory human data)
In mice with low-grade inflammation, the extract raised Akkermansia and related bacteria and reduced anxiety-like behavior. In 26 older adults from one sleep trial, saffron users gained short-chain fatty acid producing bacteria such as Faecalibacterium. These are associations in small samples, not proof that gut changes drive any sleep effect.
Clinical trials
Decentralized three-arm randomized, double-blind, placebo-controlled trial of 20 mg or 30 mg Safr'Inside taken each evening for 4 weeks (Schuster J, Mundhenke C, Nordsieck H, et al. 2025, Sleep Med X 10:100147, PMID 40698027). Run at Leibniz University Hannover; the study was financially supported by Activ'Inside, which also gave editorial feedback on the manuscript, and two co-authors are its employees.
165 adults in Germany with moderate insomnia symptoms (mean age about 45, 77% women); 150 completed.
On the primary Athens Insomnia Scale (0 to 24), scores fell 1.96 points on 30 mg, 1.54 on 20 mg and 0.71 on placebo; the pooled saffron groups beat placebo (adjusted difference -0.95 points), but neither dose was significant on its own. Rated sleep quality and perceived stress improved on both doses, a brief anxiety and depression screen improved on 30 mg only, and sleepiness, affect and quality of life did not change. Stomach discomfort was reported 10, 6 and 4 times on 30 mg, 20 mg and placebo.
Randomized, double-blind, placebo-controlled pilot trial of one 30 mg Safr'Inside gummy 30 minutes before bed for 4 weeks (Lang L, Ditton A, Stanescu A, et al. 2025, Food Funct 16(17):6817-6832, PMID 40762630). Funded by UK Research and Innovation at the University of East Anglia; Activ'Inside supplied the extract, two co-authors are its employees and one senior author receives company funding.
52 older adults aged 55 to 85 with self-reported sleep complaints, 26 per group; gut bacteria analyzed in 26.
Global sleep questionnaire (PSQI) scores improved 21% on saffron while worsening 9% on placebo (interaction p = 0.02), with similar gains in self-rated sleep efficiency. On an EEG headband, latency to persistent sleep and sleep onset latency shortened versus placebo. Insomnia Severity Index and sleepiness scores did not differ, and REM sleep share fell on saffron while rising on placebo. The placebo group's decline contributes to the difference, and no side effects were reported.
Randomized, double-blind, placebo-controlled parallel-group trial of 15 mg Safr'Inside twice daily for 8 weeks, with lab stress tests on dosing days (Jackson PA, Forster J, Khan J, et al. 2020, Front Nutr 7:606124, PMID 33598475). Run at Northumbria University and funded by Activ'Inside; five co-authors are company employees.
56 healthy adults aged 18 to 54 with subclinical low mood, anxiety or stress, analyzed per protocol.
The primary outcome, Profile of Mood States total mood disturbance, did not differ from placebo, nor did the Perceived Stress Scale. The depression subscale fell more on saffron (p = 0.05), social relationship scores were higher at week eight, and a single dose kept heart rate variability from dropping during the stressor. Urinary crocetin rose and tracked the change in depression scores. Adverse events were equal (20 per group), but stomach upset was reported 5 times on saffron and never on placebo.
Randomized, double-blind, placebo-controlled trial of 15 mg saffron extract twice daily for 6 weeks (Amadieu C, Leyrolle Q, Farneti M, et al. 2025, Am J Clin Nutr 122(6):1625-1635, PMID 41047129). Publicly funded (PRIMA and the French National Research Agency) and run at INRAE NutriNeuro in Bordeaux; the extract came from Activ'Inside, which employs two co-authors, but the paper does not call it Safr'Inside and lists its own specification.
51 healthy adults with subclinical depressive, anxiety and fatigue symptoms.
The primary outcome, a combined depression, anxiety and fatigue score, did not differ from placebo, and neither did each symptom score on its own; both groups improved over time. Only the mental health subscale of a quality-of-life questionnaire improved more on saffron. Inflammatory markers and stress-hormone reactivity did not change. The authors concluded the extract did not affect subclinical depressive symptoms.
Randomized, double-blind, placebo-controlled three-way crossover trial of a single 30 mg dose of Safr'Inside, 0.06 mg synthetic safranal or placebo given under the tongue before the Maastricht Acute Stress Test (Pouchieu C, Pourtau L, Brossaud J, et al. 2023, Nutrients 15(13):2921, PMID 37447245). Funded by Activ'Inside; three co-authors are company employees.
19 healthy men aged 18 to 25, with 28-day washouts between visits.
Self-rated stress and anxiety differed across the three arms, but in pairwise tests only safranal, not the extract, was significantly better than placebo. The extract delayed the time to peak salivary cortisol and cortisone; peak levels and total cortisol output did not differ. The product was well tolerated.