Saffron for low mood: what the studies show and where their limits lie

Hosseini and a team at Mashhad University of Medical Sciences collected what actually happens after swallowing. Their finding: crocin is barely detectable in the bloodstream after oral intake. It is converted to crocetin in the gut.

Crocetin in turn distributes into various tissues, because it binds only weakly to albumin. And it can cross the blood brain barrier by passive diffusion. The authors write explicitly that it could therefore be effective in disorders of the nervous system.

For you this means: the substance that arrives in the brain is not the substance you swallow. Your gut stands in between.

Hausenblas and a team pooled five randomised controlled trials of saffron in people with a diagnosed depressive disorder, two of them against placebo, three against an antidepressant.

Against placebo they found a large mean effect size of 1.62. Against the antidepressants the effect size was minus 0.15, so practically no difference between the treatments. The mean Jadad score of the studies was 5 and therefore high.

The authors themselves wrote the decisive sentence about it: larger studies by research teams outside Iran with long follow-up are needed before firm statements about efficacy and safety can be made.

Tóth and a team at the University of Szeged and the University of Pécs repeated the exercise with stricter methodology. Eleven randomised trials entered the qualitative analysis, nine were pooled statistically.

Result: saffron was significantly more effective than placebo, with a Hedges g of 0.891 and a confidence interval from 0.369 to 1.412. Against the antidepressants examined, the g was minus 0.246 with a p value of 0.053, so no inferiority, but also just at the edge of significance.

For you this means: an independent European working group also arrived at a similar picture. That is worth more than a repetition by the same authors.

Marx and a team at the Food and Mood Centre of Deakin University included 23 randomised controlled trials, both with saffron alone and as an addition to existing medication.

Against placebo they found a large g of 0.99 for depressive symptoms and a g of 0.95 for anxiety symptoms. As an addition to antidepressants, the g was 1.23.

And then comes the sentence I have never read in any promotional text about saffron: the Egger test indicated publication bias. The authors therefore explicitly call for further studies and criticise the lack of regional diversity.

Noorbala and a team at the Psychiatric Research Center of Roozbeh Hospital at Tehran University of Medical Sciences compared a saffron extract with fluoxetine over six weeks. Forty adults with mild to moderate depression according to DSM-IV took part.

At the end there was no statistically meaningful difference between the groups in the improvement of the depression scale. The groups also did not differ in the side effects observed.

Important for context: the authors call their own work a pilot study and call for a large trial. That is not a modesty formula, it is a methodological statement.

Akhondzadeh and a team from the same clinic compared saffron with imipramine, an older tricyclic antidepressant, over six weeks. Thirty people with mild to moderate depression took part.

The difference between the groups was not statistically meaningful. Something else stood out: in the imipramine group, anticholinergic side effects such as dry mouth and tiredness occurred more often.

For you this means: where studies found a difference, it lay more often in tolerability than in improvement.

Shafiee and a team at Alborz University of Medical Sciences pooled eight randomised trials in which saffron was tested directly against SSRIs, that is, against the antidepressants commonly used today.

For depressive symptoms the standardised mean difference was 0.10 with a confidence interval from minus 0.09 to 0.29, so no meaningful difference. For anxiety symptoms the difference across four studies was 0.04. Adverse events occurred less often in the saffron group, with a risk difference of minus 0.06.

The authors conclude cautiously: saffron could be an alternative, and larger studies in more diverse populations are needed.

Kell and a team at the University of Southern Queensland gave 128 healthy adults with self reported low mood either 28 milligrams of saffron extract daily, 22 milligrams daily or placebo for four weeks.

In the group taking 28 milligrams, negative mood fell clearly, with a difference to placebo of d equals minus 1.10. In the group taking 22 milligrams there was no effect. Two co-authors of the paper work for the company that manufactures the extract tested.

For you this means two things. First: the amount appears to play a large role. Second: with studies on a specific branded product it is always worth looking at the author list.

Lopresti and a team gave 202 adults between 18 and 70 years of age with subthreshold depressive symptoms 28 milligrams of saffron extract daily or placebo for twelve weeks. It is the largest work on this topic.

On the primary outcome, the depression scale of the DASS-21, saffron was superior to placebo, with a difference of 2.92 points and a Cohen's d of 0.39. A clinically meaningful improvement was reached by 72.3 percent in the saffron group compared with 54.3 percent under placebo. For all secondary outcomes there was no difference between the groups.

And then there is a sentence I value greatly: the authors explicitly name the large placebo response in their study and call for it to be taken into account in future work. More than half of the placebo group improved in a clinically meaningful way.

Lopresti and Drummond at Murdoch University systematically evaluated the clinical studies available at the time and additionally collected the possible mechanisms of action.

Their conclusion on the mechanisms: the antidepressant effects of saffron could rest on serotonergic, antioxidant, inflammation dampening, neuroendocrine and neuroprotective properties. The evidence for this comes predominantly from cell experiments and animal models, not from measurements in people with depression.

For you this means: from studies we know reasonably well that something moves on the scales. Why, is a well founded assumption and not established knowledge.

Osimo and a team at the University of Cambridge evaluated 37 studies with 13,541 people with depression and 155,728 control participants.

Result: 27 percent of the people with depression had a CRP above 3 milligrams per litre, that is, low grade inflammation. 58 percent had a CRP above 1 milligram per litre. Compared with healthy controls, the odds ratio was 1.46.

For you this means: in roughly a quarter to a half of those affected, inflammation is a contributing factor. In the majority it is not. Both belong in the picture.

Liao and a team evaluated 26 double-blind randomised trials with 2,160 participants on omega 3 fatty acids in depression.

Across all studies there was a favourable overall effect with a standardised mean difference of minus 0.28. The decisive finding, however, is in the detail: the benefit showed up with pure EPA formulations and with formulations containing at least 60 percent EPA, and at an EPA amount of up to one gram per day. For DHA weighted preparations this advantage did not appear.

For you this means: omega 3 is not simply omega 3. Which composition makes sense in an individual case belongs in a medical conversation and not in a blog recommendation.

Okereke and a team at Harvard Medical School and Brigham and Women's Hospital examined 18,353 people aged 50 and over within the VITAL trial. One half received 2,000 International Units of vitamin D3 daily, the other placebo, over a median of 5.3 years.

The result was clearly negative. The risk of depression or clinically relevant depressive symptoms did not differ, with a hazard ratio of 0.97. Mood scores did not change either; the mean difference was 0.01 points.

The authors write in so many words that these results do not support the use of vitamin D3 in adults for the prevention of depression. That does not mean an existing deficiency should go untreated. It means: vitamin D as a blanket mood prophylaxis for everyone did not hold up.

Jacka and a team ran the SMILES trial, a twelve week randomised study with 67 adults with moderate to severe depression. One group received seven dietary counselling sessions with a clinical dietitian, the other an equally long programme of social support.

Important: 55 of the 67 participants were in treatment in parallel, with psychotherapy, medication or both. The dietary intervention came on top, it replaced nothing.

The dietary group improved considerably more on the MADRS, with a Cohen's d of minus 1.16. Remission was reached by 32.3 percent compared with 8.0 percent. The number needed to treat was 4.1.

Falsini and a team at the Università Cattolica del Sacro Cuore in Rome gave 25 people with early age related macular degeneration 20 milligrams of saffron daily or placebo for three months and then switched the groups.

After saffron, the amplitude in the focal electroretinogram rose compared with baseline and placebo, and thresholds fell. A follow-up by the same working group over an average of 14 months described that this improvement remained stable.

The authors themselves stress that these results need to be replicated and that the clinical significance is still open. I do not quote them as an eye recommendation, but as an indication that carotenoids may change something in several tissues.

Agha-Hosseini and a team at the universities of Tehran and Zanjan examined women between 20 and 45 years of age with a regular cycle and premenstrual complaints over at least six months. Over two cycles they received saffron or placebo.

In the cycles examined there was a meaningful difference in favour of saffron, both in the daily symptom diary and on the depression scale.

For you this means: if your low mood follows a cycle pattern, that is a question of its own with its own diagnostic workup. It belongs in a gynaecological and medical conversation, not covered over with a preparation.

Modaghegh and a team at Mashhad University of Medical Sciences gave 30 healthy volunteers either placebo, 200 milligrams or 400 milligrams of saffron daily for one week. That is a multiple of the amount used in depression studies.

Under the higher dose, standing systolic blood pressure and mean arterial pressure fell clearly. In addition, red blood cells, haemoglobin, haematocrit and platelets fell slightly, while sodium, urea and creatinine rose.

The authors stress that all changes stayed within the normal range and were not clinically meaningful. For practice I still draw a clear consequence from it: with clotting disorders, under blood thinners and with low blood pressure, caution is appropriate.

Lopresti and a team examined 160 adults with persistent depressive symptoms despite ongoing antidepressant medication. For eight weeks they additionally received saffron extract or placebo.

The result is split. On the clinician rated MADRS, symptoms fell by 41 percent under saffron compared with 21 percent under placebo. On the self rated version of the same scale the difference disappeared: 27 versus 26 percent. For quality of life there was no difference.

The authors themselves call the findings contradictory and call for further research. What matters for you: this study took place under medical supervision. Combining on your own initiative is something entirely different.