HRV biofeedback: what the evidence actually says.

What HRV biofeedback is

Heart-rate variability is the beat-to-beat variation in heart rate. A healthy heart at rest doesn't beat at perfectly regular intervals — it speeds up slightly on inhale and slows on exhale, an effect called respiratory sinus arrhythmia. The amplitude of this variation reflects vagal tone, autonomic flexibility, and overall regulatory capacity. Lower HRV correlates with poorer health across many domains. [1]

HRV biofeedback is the practice of breathing at a specific rate — typically around six breaths per minute, the so-called "resonant frequency" — while watching the HRV signal on a screen. At resonant frequency, the breath-driven variability and the underlying baroreflex-driven variability fall into phase, producing notably amplified HRV. Over training sessions, the autonomic system learns to spend more time in this regulated, vagally-mediated state. [2]

The strongest signals in the meta-analytic literature

Lehrer et al.'s 2020 systematic review and meta-analysis pooled 58 randomised controlled trials and reported significant effects across emotional regulation (g = 0.83), physical health markers (g = 0.39), and performance under pressure (g = 0.55). [3] The effect sizes here are substantial — particularly on emotional outcomes, where they sit in the range expected of established psychological interventions.

Goessl, Curtiss, and Hofmann's 2017 meta-analysis focused specifically on stress and anxiety, pooling 24 RCTs. They reported a large effect on self-reported stress and anxiety symptoms (Hedges' g = 0.81), with effects maintained at follow-up. [4]

Pizzoli et al.'s 2021 meta-analysis examined depressive symptoms, finding moderate-to-large effects (g = 0.68) across 14 RCTs, with the strongest signal when HRV biofeedback was used as adjunct to other treatment rather than as a stand-alone. [5]

The picture across these meta-analyses is unusually consistent for the field: HRV biofeedback at resonant frequency, dosed adequately, produces meaningful improvements on anxiety, stress, depressive symptoms, and performance variables, with effect sizes that compare favourably to many established treatments.

What "dosed adequately" means

The trials that show the largest effects share a structure. Typically:

• An initial in-clinic calibration session to identify the patient's individual resonant frequency (between 4.5 and 7 breaths per minute, varies by person).
• 4–10 weekly clinic sessions of 30–45 minutes each, with the patient breathing at resonant frequency while the HRV display provides feedback.
• Daily home practice — typically 20 minutes, often split into two 10-minute sessions — using a phone app or simple HRV-capable device.
• Ongoing monitoring of HRV metrics across the programme, with re-calibration if needed.

Studies that allow patients to do less than this — fewer clinic sessions, no daily home practice, no individualised resonant-frequency calibration — show smaller effects, sometimes indistinguishable from sham conditions. The dose matters.

Where the strongest clinical fits sit

In our clinic, the highest-yield uses of HRV biofeedback are:

Anxiety presentations. Particularly chronic over-arousal, generalised anxiety, and panic where the autonomic ground is the engine. Often paired with CBT delivered by a referring psychologist; HRV biofeedback adjuncts well to psychological work.

Stress and burnout recovery. Patients whose autonomic system has been running on emergency settings for sustained periods tend to be highly trainable, and the home-practice carries low risk.

Performance under pressure. Athletes, performers, surgeons, pilots — anyone whose cognitive output depends on autonomic recovery between high-load events. Effect sizes on performance variables are robust in the literature.

Trauma adjunct. Strictly as autonomic-regulation skill-building alongside trauma-focused psychotherapy. Never as a substitute for the psychological work itself.

The limits

HRV biofeedback is not a cure for any condition. It does not replace medication, psychotherapy, or primary medical care. Patients with significant cardiac arrhythmia or autonomic conditions (POTS, dysautonomia) need medical clearance before starting — we screen for this in intake.

The literature, while consistent, is not without weakness. Many trials are small. Sham-controlled blinding is genuinely hard in a biofeedback context. Effect sizes attenuate at long-term follow-up if patients stop the home practice. The practice has to continue, at least for a maintenance period, for the gains to stick.

The honest summary

HRV biofeedback is unusual in our field for having a meta-analytic literature that genuinely supports clinical use, with effect sizes comparable to established psychological interventions. It is also unusual for being something patients can largely self-administer once trained — the cost-effectiveness profile is favourable, the risk profile is mild, the integration with other care is straightforward.

It is not magic. The dose matters; the daily practice matters; the clinical fit matters. Done properly, it is one of the better-supported interventions in the modalities we use.

---

About the author. Dr Ash Connell is a chiropractor (AHPRA CHI0001772308), board-certified in quantitative EEG (QEEG-D), and the founding clinician of The Healthy Brain Clinic. He practises from Geelong and Camperdown, Victoria, and online Australia-wide. Read his bio →

References

1. Thayer JF, Yamamoto SS, Brosschot JF. (2010). The relationship of autonomic imbalance, heart rate variability and cardiovascular disease risk factors. International Journal of Cardiology, 141(2), 122–131.
2. Lehrer P, Vaschillo E. (2008). The future of heart rate variability biofeedback. Biofeedback, 36(1), 11–14.
3. Lehrer P et al. (2020). Heart rate variability biofeedback improves emotional and physical health and performance: a systematic review and meta analysis. Applied Psychophysiology and Biofeedback, 45, 109–129.
4. Goessl VC, Curtiss JE, Hofmann SG. (2017). The effect of heart rate variability biofeedback training on stress and anxiety: a meta-analysis. Psychological Medicine, 47(15), 2578–2586.
5. Pizzoli SFM et al. (2021). A meta-analysis on heart rate variability biofeedback and depressive symptoms. Scientific Reports, 11, 6650.

Editorial note. This article was drafted by Dr Ash Connell with structural support from a large language model, then reviewed for clinical accuracy and AHPRA compliance before publication. Citations are real, peer-reviewed sources. The clinical interpretations are Dr Ash's own.