The gut-brain axis is tested in a well-designed trial, almost positive
STUDY: Seamans KM et al, Experimental and Clinical Psychopharmacology 2026
STUDY TYPE: Randomized, double-blind, placebo-controlled trial
FUNDING: PrecisionBiotics Group Ltd., part of Novonesis, which manufactures the study product; several authors are Novonesis employees
Background
Probiotics treat depression in many small trials. Though they ring positive in meta-analyses, we need more large trials (n > 100) to confirm them. This one does that with Bifidobacterium longum-1714, a strain with benefits in stress and sleep.
The Study
- 168 adults with mild to moderate major depressive disorder
- Double-blind trial
- Randomized to B. longum 1714 or placebo daily for 8 weeks
- Primary outcome = change in depression severity at week 8.
Results
The probiotic didn’t beat placebo on the primary outcome. Depression scores fell substantially in both groups, but the difference between groups at week 8 was not significant (mean difference −1.4; p = .13).
At week 4, however, several secondary measures favored the probiotic:
- Depression scale (PHQ-9) scores fell more in the probiotic group (difference −1.0; p = .045)
- Sleep quality improved more (Sleep scale (PSQI) difference −0.9; p = .009), with poor sleepers dropping from 87% to 58% versus 79% in the placebo group
- Quality-of-life subscores (SF-36) improved for vitality, mental health, and social functioning
None of those promising differences survived to week 8, except vitality, which remained modestly better in the probiotic group (p = .038). Adverse events were similar in both groups.
Limitations: single-center trial in a predominantly White, female, Irish cohort. The mild-to-moderate severity means strong placebo response was expected, and that’s exactly what happened. Multiple secondary endpoints also increase the risk of false-positive findings.
Practice Implications
- An inconclusive trial. Not positive on the primary outcome.
- This was a single strain study, and probiotics work better when multiple strains are used together. This page has some that passed.
— Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
