The short version
- Slow breathing reliably changes heart rate variability. Whether it reduces anxiety is a separate and less settled question.
- The "resonance" frequency is individual, usually between 4.5 and 6.5 breaths per minute, not a universal 5.5.
- Trials in anxiety and depression are small, heterogeneous, and rarely blinded; effects are modest and often not sustained at follow-up.
- Slow breathing is a reasonable adjunct. It is not a treatment for a panic disorder on its own.
Two claims travel together so often that they are usually treated as one. The first is physiological: slow breathing at around six breaths per minute increases the amplitude of heart rate variability, because the cardiovascular and respiratory rhythms become coupled. The second is clinical: this change reduces anxiety. The first is well established. The second is where the evidence gets thin, and where most popular writing stops looking.
Separating two claims that get merged
Heart rate variability is the variation in the interval between heartbeats. It is influenced by the balance of sympathetic and parasympathetic activity, and by respiration itself: heart rate rises slightly during inhalation and falls during exhalation, a phenomenon called respiratory sinus arrhythmia.
Breathe slowly and this oscillation becomes larger and more regular. At a certain rate — the resonance frequency — the respiratory and baroreflex rhythms align, and the measured amplitude peaks. This is a robust, repeatable finding. It is also a mechanical consequence of paced breathing, not evidence that the nervous system has been "reset".
The clinical claim requires different evidence: randomised trials with credible comparators, blinded outcome assessment where possible, and follow-up beyond the end of the intervention. That literature exists, but it is smaller and messier than the physiological one, and the two are frequently conflated in marketing.
Why heart rate variability is not anxiety
HRV is a useful signal and a poor proxy. Higher resting HRV is associated with better cardiovascular health and greater autonomic flexibility at the population level, but an individual's HRV varies enormously with age, posture, sleep, caffeine, alcohol, illness and the measurement device. A rise in HRV during a breathing exercise tells you that the exercise changed the measurement. It does not tell you that the person feels calmer, and the correlation between the two across individuals is modest.
A breathing practice can change the signal without changing the suffering. Both matter, but they are not the same outcome.
This is not an argument against measuring. It is an argument against using the measurement as the goal. When people chase HRV numbers, they tend to breathe harder and longer, which produces a larger oscillation and more light-headedness, and they interpret the light-headedness as depth.
What the trials measured
The trials that exist fall into three groups: slow breathing as an adjunct in anxiety and panic, slow breathing for stress in healthy or occupational samples, and slow breathing for conditions where autonomic tone is plausibly involved, such as hypertension and chronic pain.
| Population | Typical protocol | Reported effect | Confidence |
|---|---|---|---|
| Healthy adults, acute stress | 5–6 breaths/min, 5–10 min | Lower self-reported tension, higher HRV during practice | Moderate for immediate state; low for lasting trait change |
| Panic disorder | Breath retraining plus CBT | Reduced panic frequency when added to therapy | Moderate; the breathing component is not isolated |
| Generalised anxiety | Daily practice, 4–8 weeks | Small reductions on rating scales | Low to moderate; small samples, weak blinding |
| Hypertension | Device-guided slow breathing | Modest systolic reduction, inconsistent across trials | Low; effect shrinks in better-controlled studies |
| Chronic pain | Slow breathing with attention training | Improved pain tolerance in some trials | Low; heterogeneous methods |
The pattern is consistent with a modest, real, non-specific effect. Slow breathing changes arousal in the moment, and repeated practice may shift the baseline slightly. It is not a treatment for a disorder, and the trials that claim large effects tend to be the ones with the weakest designs.
Dose, pace and the 5.5-second myth
The most persistent number in breathwork is 5.5: a five-and-a-half-second inhale and a five-and-a-half-second exhale, or 5.5 breaths per minute. It comes from work on resonance frequency, and it is a reasonable starting point because it sits in the middle of the range where most adults show peak HRV amplitude.
It is not universal. Individual resonance frequencies in the literature span roughly 4.5 to 6.5 breaths per minute, and they vary with age, body size and posture. A pace that feels effortful is the wrong pace for you, regardless of what the number says.
Dose matters more than precision. Five minutes, twice a day, practised consistently, is a reasonable protocol for a healthy adult. Longer sessions do not produce proportionally larger effects, and they increase the risk of hyperventilation-related dizziness in people who try to breathe "deeply" rather than slowly.
Slow is not deep
The most common error is combining a slow rate with a large volume. Slow breathing should reduce tidal volume slightly, not increase it. If you feel light-headed, tingly, or spaced out, you are over-ventilating. Return to normal breathing and try again with less air.
How to read the next study you see
Four questions will filter most of the noise. Was there a control group, and what did it do? Were the outcomes measured by someone who knew which group the participant was in? Was the effect measured during the practice or weeks later? And how many participants were there?
A trial of 30 people showing a large reduction in anxiety after two weeks of slow breathing is a hypothesis, not a finding. A trial of 300 with an active control and blinded assessment showing a small effect is a finding, even though the number looks less impressive.
None of this means slow breathing is worthless. It means the honest claim is narrow: it reliably shifts autonomic activity in the moment, it is safe for most people, and it may help some people manage arousal. That is enough to justify practising. It is not enough to justify replacing treatment.
References
- Bernardi L, et al. Slow breathing increases arterial baroreflex sensitivity in patients with chronic heart failure. Circulation. 2002;105:143–145.
- Lehrer PM, et al. Resonant frequency biofeedback training to increase cardiac variability. Appl Psychophysiol Biofeedback. 2000;25(3):177–191.
- Zaccaro A, et al. How breath-control can change your life: a systematic review on psycho-physiological correlates of slow breathing. Front Hum Neurosci. 2018;12:353.
- Fincham GW, et al. Effect of coherent breathing on anxiety: a randomised controlled trial. BMC Complement Med Ther. 2023.
- Thayer JF, et al. Heart rate variability, prefrontal neural function, and cognitive performance. Ann N Y Acad Sci. 2009;1171:349–364.
- Meuret AE, et al. Respiratory and cognitive mediators of treatment change in panic disorder. J Consult Clin Psychol. 2010.
