Most included studies had moderate risk of bias, and the primary outcomes were subjective.
What the paper gives us.
- 01
A modest signal for regulation
The primary analysis found a small-to-medium reduction in subjective stress. That supports a practical role in downshifting, especially when stress is obstructing work.
- 02
The evidence does not prove cognitive enhancement
The outcomes were mental-health measures, not memory, focus, or intelligence tests. A calmer state may help work indirectly, but this paper did not test that claim.
- 03
Caution is part of the finding
The authors explicitly call for better, lower-bias research. That caveat should remain visible whenever breathwork is discussed as a trend.
What was actually studied.
- Design
- Meta-analysis of randomized controlled trials
- Sample
- 12 primary stress RCTs · 785 participants
- Population
- Adults in breathwork randomized controlled trials.
- Comparison
- Breathwork interventions versus non-breathwork controls.
- Outcomes
- Self-reported stress, anxiety, and depressive symptoms.
- Effect
- Stress: Hedges’ g = −0.35 (95% CI −0.55 to −0.14).
What this may mean for work.
A brief breathing practice can be a useful boundary between a stressful event and your next task. The goal is not to force a peak state; it is to create enough room to choose what to do next.
This is VSX's practical inference, not a claim directly established by the study.
Use a simple downshift
For two to five minutes, breathe comfortably and a little more slowly than usual—without breath holds or forcing depth—before returning to work. Rate tension from 1–5 before and after. Stop if it causes dizziness or discomfort.
Where the claim stops.
- Most studies were judged at moderate risk of bias.
- The analysis pooled varied techniques and populations.
- Results are based primarily on self-reported outcomes.
Read the original research.
Guy William Fincham, Clara Strauss, Jesus Montero-Marin, and Kate Cavanagh. “Effect of breathwork on stress and mental health: A meta-analysis of randomised-controlled trials.” Scientific Reports, 2023.
Open source paperReviewed by VSX on July 18, 2026.