Does breathwork actually reduce stress? The trial evidence

Breathwork sits in an unusual position: it is a wellness trend with real randomised trials behind it, promoted with claims that run far ahead of what those trials measured. Box breathing, the physiological sigh and 4-7-8 are now standard advice for stressed desk workers, and two solid pieces of evidence tell you roughly how much to expect.

The short version: the effect on how you feel is real and modest, it shows up fast, and the claims about “resetting your nervous system” are doing more work than the data support.

What counts as breathwork

Breathwork means deliberately controlling the pattern of your breathing, usually by slowing it, extending the exhale, or adding holds. It is distinct from meditation, which typically observes the breath without changing it. That distinction matters because the best trial in this area used meditation as the comparison.

The three patterns tested most directly are cyclic sighing (a double inhale through the nose and a long exhale), box breathing (equal inhale, hold, exhale, hold) and cyclic hyperventilation with retention (fast breathing followed by breath holds, the Wim Hof family).

The pooled evidence

A 2023 meta-analysis in Scientific Reports gathered the randomised trials. The primary outcome self-reported/subjective stress included 12 randomised-controlled trials with a total of 785 adult participants.

The result was positive and small: breathwork was associated with lower levels of stress than control conditions, with a mean effect size of g = -0.35. Secondary outcomes moved similarly, with anxiety at g = -0.32 across 20 trials and depressive symptoms at g = -0.40 across 18.

Two honest caveats come from the paper itself. Most studies were deemed as being at moderate risk of bias. And every outcome is self-reported, which in an intervention nobody can be blinded to is exactly where expectation effects live.

The trial that compared techniques

The most-cited single study is a 2023 Stanford randomised trial published in Cell Reports Medicine. It compared three different daily 5-min breathwork exercises compared with an equivalent period of mindfulness meditation over 1 month, with participants split across cyclic sighing, box breathing, cyclic hyperventilation and a meditation control.

The winner was the exhale-focused one: breathwork, especially the exhale-focused cyclic sighing, produces greater improvement in mood and reduction in respiratory rate compared with mindfulness meditation. The authors’ practical conclusion was that daily 5-min cyclic sighing has promise as an effective stress management exercise.

Two things temper that. The study measured mood and respiratory rate; it did not find the autonomic changes people often claim, since heart rate variability and resting heart rate did not differ across conditions. And one author disclosed becoming an advisor to WHOOP, the wearable company whose device collected the data, during the study period.

What this means in practice

The evidence supports a narrow, useful claim. Slow breathing with a long exhale, done for about five minutes, reliably improves how people rate their mood and reduces breathing rate, with effects that grew over four weeks in the Stanford trial. That is a cheap, fast, side-effect-free tool for the moment before a difficult meeting or after a bad one.

It does not support claims that breathwork fixes anxiety disorders, replaces treatment, “hacks” the vagus nerve in any measurable way, or improves recovery metrics. The HRV result specifically contradicts the most common wearable-adjacent claim.

The practical protocol from the best trial is simple: two inhales through the nose, the second shorter and topping up the lungs, then a long slow exhale through the mouth, repeated for five minutes a day.

Where this sits among stress interventions

It is worth keeping the size of the effect in perspective. A standardised effect around 0.35 is the kind of difference that shows up clearly in a pooled analysis and subtly in a person. Exercise, adequate sleep and therapy all have larger and better-replicated effects on stress, and none of them is replaced by a breathing exercise.

What breathwork has that those do not is availability. It takes five minutes, needs no equipment, can be done at a desk between calls, and has no meaningful downside for most people. That combination is rare enough that a small effect is still worth having, as long as you are not relying on it to carry work that belongs to sleep, movement or professional support.

Where the evidence stops

Trials are short, mostly a few weeks. Samples are small. Nobody has compared breathwork head-to-head with exercise, therapy or sleep improvement, which are the interventions with larger effects on stress. And because blinding is impossible, the size of the genuine physiological contribution versus the ritual of stopping work for five minutes is unresolved.

This is not individual medical advice. Breath-holding practices, particularly cyclic hyperventilation, carry real risks in and around water and for people with certain cardiac or respiratory conditions; decisions about managing anxiety or a health condition belong with a clinician.

Common questions

Does breathwork really reduce stress?

Yes, modestly. A meta-analysis of 12 randomised trials with 785 adults found breathwork lowered self-reported stress with a small-to-medium effect size of about 0.35. All outcomes were self-reported and most trials had moderate risk of bias.

What is the best breathing technique for anxiety?

In the one trial comparing techniques directly, exhale-focused cyclic sighing produced greater mood improvement than box breathing, cyclic hyperventilation or mindfulness meditation over 28 days.

How long do you need to do breathwork?

The Stanford trial used five minutes a day for 28 days, and effects accumulated across the month. Nothing in the evidence requires longer sessions.

Does breathwork improve heart rate variability?

Not in the best trial. The Stanford study found no significant changes in heart rate variability or resting heart rate in any group, including cyclic sighing, despite the reduction in respiratory rate.

The bottom line

Five minutes of slow, exhale-heavy breathing is one of the cheapest interventions with real trial evidence behind it, and the effect is a small lift in mood rather than a nervous-system reset. Use it as a between-meetings tool, ignore the HRV claims, and do not let it substitute for the bigger levers on stress.