The short version
- A 2026 GRADE-assessed meta-analysis of 34 randomised trials in 1,769 adults found saffron reduced Beck Depression Inventory scores by 4.39 points and Beck Anxiety Inventory scores by 5.06 points versus placebo, with certainty of evidence rated moderate.
- The same analysis found no significant effect on the clinician-administered Hamilton Depression and Hamilton Anxiety rating scales, or on the Profile of Mood States.
- Heterogeneity in those pooled results was 92.3% for depression and 94.8% for anxiety, meaning the underlying trials disagreed with each other enormously.
- A 2025 meta-analysis comparing saffron with SSRIs found non-significant differences for depression (SMD 0.10) and anxiety (SMD 0.04) across 8 and 4 studies respectively, which reflects limited statistical power rather than demonstrated equivalence.
- In 56 healthy adults with subclinical low mood, 30mg a day of standardised saffron extract for eight weeks reduced depression scores, and urinary crocetin levels rose in step with that change.
A cooking spice with 34 randomised controlled trials behind it is an odd thing to find in the literature, and saffron genuinely has them. It is one of the better-studied botanicals in the mood category — better studied than holy basil, better studied than reishi — and the results have a strange internal split that almost nobody selling it mentions.
The pooled numbers
Mahmoudi and colleagues, in Nutritional Neuroscience in 2026, ran a PRISMA-compliant systematic review and meta-analysis of randomised placebo-controlled trials in adults, lasting at least four weeks, with saffron as the only difference between arms. Thirty-four trials, 1,769 participants, certainty of evidence graded using GRADE.
On the Beck Depression Inventory, pooled across 14 trials and 817 participants, saffron reduced scores by a weighted mean difference of 4.39 points (95% CI −6.64 to −2.15, p < 0.001). On the Beck Anxiety Inventory, across 6 trials and 339 participants, by 5.06 points (95% CI −8.44 to −1.68, p = 0.003). GRADE certainty: moderate. For a botanical, that is a strong file.
Then the other half of the same analysis. No significant effects were observed on the Hamilton Depression Rating Scale, the Hamilton Anxiety Rating Scale, or the Profile of Mood States.
Why that split matters
Beck inventories are self-report. The participant fills them in. Hamilton scales are clinician-administered — a trained rater interviews the person and scores them. Both types were measured in the same body of trials, often in the same studies, and the effect appeared on one and not the other.
There are two readings and we do not know which is right. The generous one: self-report captures a real internal change that a structured clinical interview is too blunt to pick up. The unflattering one: participants who believed they were taking an active treatment reported feeling better, and a trained observer could not detect a difference. The second reading is what expectancy effects look like in data, which is why the reviewers explicitly call for high-quality trials on the back of it.
Heterogeneity is the other thing to hold. I² was 92.3% for the depression outcome and 94.8% for anxiety. Those are extreme values; they mean the trials disagreed with each other enormously and a pooled average across them is a summary of a very scattered picture.
The comparison people quote wrongly
A 2025 meta-analysis in Nutrition Reviews compared saffron directly against SSRIs. Across 8 studies on depression, the difference was non-significant (SMD 0.10, 95% CI −0.09 to 0.29). Across 4 studies on anxiety, also non-significant (SMD 0.04, 95% CI −0.22 to 0.29). Participants on saffron reported fewer adverse events (risk difference −0.06, 95% CI −0.09 to −0.04, I² = 0%).
This gets circulated as "saffron works as well as an antidepressant". It does not say that. A non-significant difference across eight small trials is a statement about statistical power at least as much as about saffron, and these are not equivalence trials — they were not designed or sized to demonstrate that two treatments are interchangeable. If you are on an SSRI, nothing here is a reason to change anything without your prescriber.
What happens in people who are not unwell
Jackson and colleagues, in Frontiers in Nutrition, ran the trial closest to the average reader: 56 healthy adults aged 18 to 54 with subclinical low mood, anxiety or stress, randomised to 30mg a day of a standardised saffron extract or placebo for eight weeks.
The saffron group reported reduced depression scores and improved social relationships by the end. Urinary crocetin — a marker confirming they had actually absorbed the compound — rose significantly and correlated with the change in depression scores, which is a nice piece of internal validation. And on acute intake, the usual stress-induced drop in heart rate variability during a laboratory stressor was attenuated.
Fifty-six people, eight weeks, and several authors were employed by the extract manufacturer, which the paper discloses. It is a good small trial, not a definitive one.
What we'd actually tell you
We do not sell saffron and have no plans to, so there is nothing at the bottom of this page to click. If you want to try it, 30mg a day of a standardised extract for eight weeks is the protocol with a trial behind it, and it is expensive relative to what it delivers — this is the most costly spice in the world and the effect size is moderate at best.
The reason we wrote this up anyway is that saffron's evidence base is a useful teaching case. It is better than most of the stress shelf, it is graded moderate rather than low, and it still splits down the middle depending on who was holding the questionnaire. If that is what a good botanical file looks like, calibrate accordingly for the rest of them — and remember that slow breathing produced its effect without anybody paying for a gram of stigma.
Good questions
How much saffron do the mood trials use?
30mg a day of a standardised extract is the figure that recurs, taken for at least eight weeks. Meta-analysis inclusion required a minimum of four weeks. Cooking with saffron does not approach a trial dose in any practical quantity, and the price of the spice makes that an expensive way to try.
Is saffron as effective as an antidepressant?
The trials do not establish that, despite how they get quoted. Comparisons against SSRIs found non-significant differences across 8 depression studies and 4 anxiety studies, which reflects small samples rather than proven equivalence. These were not designed as equivalence trials. Do not change or stop a prescription on the strength of them.
Why did saffron work on some scales and not others?
The scales it moved are self-report, filled in by the participant. The scales it did not move are clinician-administered, scored by a trained interviewer. Both were measured in the same literature. That could mean self-report is more sensitive, or it could mean people who expected to feel better said so and an observer saw no difference.
Does saffron help with stress rather than mood?
Less clearly. The pooled analysis found no significant effect on the Profile of Mood States, which is the closest thing to a general mood-state measure in the set. One trial in healthy adults did find that acute saffron blunted the drop in heart rate variability during a laboratory stressor, but that is a single physiological outcome in 56 people.
Is saffron worth the money?
For most people, probably not, and we do not sell it so there is nothing in it for us either way. It is the most expensive spice in the world, the effect size is moderate, and the effect is clearest on questionnaires the participant filled in themselves. If you have not yet tried the free interventions in this category, try those first.
Sources
- Mahmoudi R, Mohammadi-Sartang M, Servatyari K, Rafieipour N. Effect of saffron on depression, anxiety and mood disorder: a GRADE assessed systematic review and meta-analysis of 34 randomized controlled trials. Nutr Neurosci, 2026. View study
- Shafiee A, Jafarabady K, Seighali N, et al. Effect of Saffron Versus Selective Serotonin Reuptake Inhibitors (SSRIs) in Treatment of Depression and Anxiety: A Meta-analysis of Randomized Controlled Trials. Nutr Rev, 2025. View study
- Jackson PA, Forster J, Khan J, et al. Effects of Saffron Extract Supplementation on Mood, Well-Being, and Response to a Psychosocial Stressor in Healthy Adults: A Randomized, Double-Blind, Parallel Group, Clinical Trial. Front Nutr, 2020. View study
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.
