This may be the largest meta-analysis of its kind to date
STUDY: Fornaro M et al, Molecular Psychiatry 2026
STUDY TYPE: Systematic review and network meta-analysis (163 randomized trials)
FUNDING: Independent
Background
From omega-3 to St. John’s wort, supplements are widely used for depression but lack regulatory review. This network meta-analysis steps into that role, pooling the largest evidence base yet to see which hold up.
The Study
- 163 randomized trials, 15,757 adults with major depressive disorder, testing 137 compounds vs. placebo or usual antidepressant care.
- Outcomes: depressive symptoms, response, remission, and dropout rates over a median 8 weeks.
Results
Here are the supplements that had the most reliable efficacy across the analyses:
- Omega-3 EPA (fish oil) (effect size 1.7)
- Vitamin D3 (effect size 1.6)
- St.John’s wort extract ZE117 (effect size 1.9)
Those are large effect sizes, which surprised the authors. “It is remarkable that those compounds had effect sizes larger than 1, which is highly unusual in any high-quality double-blind, placebo-controlled trial for any pharmaceutical compound, including depression.”
To arrive at that conservative list, they limited the analysis to studies that are 1) High quality and 2) Not industry sponsored. That’s a little unfair, as it usually requires industry funding to pay for a high quality study. If we used that bar for pharmaceuticals, we’d find a very short list.
Looking beyond that conservative lens, here is the full list of supplements that surpassed placebo in the basic analysis:
- Lavender (combined with Dracocephalum ruyschiana, but also is effective in trials of its own)
- Rhodiola Rosea
- Biokult
- Tensilen
- Omeag-3 EPA
- Zishen pingchan granules
- Vitamin B
- L-theanine
- Nepeta menthoides
- Vitamin D3
- Ellagic acid
- Perilla frutescens
- Saffron
- St. John’s wort extract ZE117
- S-adenosylmethionine (SAMe)
Probiotics also beat placebo when the analysis lumped them together.
For anxiety in depression, the most conservative view supported:
- Saffron (effect size 1.2)
- Shugan granules (effect size 0.6)
- Curcumin (effect size 0.6)
An earlier network meta-analysis found efficacy for a different set of supplements:
- Omega-3 EPA + DHA
- S-adenosylmethionine (SAMe)
- Curcumin
- Zinc
- Tryptophan
- l-Methylfolate
- Saffron
Other rigorous analyses support probiotics, which fell out of this review, perhaps because the various strains were analyzed independently.
Side Effects
Tolerability didn’t differ from placebo overall, and no nutraceutical performed worse than placebo on dropouts, except citicoline, which was linked to significantly more dropouts than placebo.
Limitations
Trials were small and used different designs and outcome measures. Effect sizes consistently shrank as quality rose, suggesting the biggest claimed benefits came from the weakest studies, and confidence in the evidence was low or very low for nearly all comparisons.
Practice Implications
- This is a good starting place, but there the trials vary in design, making it hard to lump them together. There are many ways to look at the evidence, which is why meta-analyses and systemic reviews.
- When choosing supplements, I’ve selected low-cost brands that have the right ingredients and past quality tests by independent labs.
—Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
