Atlas›Domain X · Body, movement, breath & physical›Guide 152
Guide X-152

Cyclic sighing

2023
A 13 · L 78 · M 99 interventions in the Trinity explorer1 described for recognition only
Client cardCompare
[source: …] where a claim comes from[uncertain] not confirmed in a primary source[synthesis] the writer's reconstruction

1. Introduction

Version followed: the cyclic sighing protocol as given to participants in Balban, Neri, Kogon, Weed, Nouriani, Jo et al. (2023), "Brief structured respiration practices enhance mood and reduce physiological arousal", Cell Reports Medicine 4(1):100895 (STAR Methods), supplemented by David Spiegel's descriptions in Stanford Medicine's own news coverage. Where these differ in wording, the trial wording is followed.

What it is. Cyclic sighing is a short, low-intensity, exhale-emphasised breathing exercise built on the shape of a natural sigh: two inhalations in a row (the second one topping up the lungs), followed by a slow, complete exhalation, repeated continuously for a few minutes. In the Stanford trial it was one of three 5-minute daily breathwork practices compared with 5 minutes of mindfulness meditation over 28 days. The Stanford team was co-led by David Spiegel (psychiatry) and Andrew Huberman (neurobiology), with Melis Yilmaz Balban as first author; Huberman recorded the instruction videos participants used. [source: Balban et al. 2023; Stanford Medicine News Center 2023; Stanford Medicine Magazine 2023]

The natural sigh. Spontaneous sighs are a brainstem-generated breath of about twice normal volume that people take roughly every five minutes; physiologists studying the sigh circuit describe their function as re-opening collapsed alveoli. The deliberate exercise borrows this breath shape; the trial paper itself does not claim the exercise works through alveolar mechanics. [source: University of Alberta / UCLA news on Li et al., Nature 2016; Balban et al. 2023]

Core idea of how change happens (as the developers state it, hypotheses only). The authors propose that slow, exhale-focused breathing shifts the body away from sympathetic arousal, may influence interoceptive processing (anterior insula) and brainstem arousal centres, and that voluntarily controlling a semi-automatic function increases a person's sense of control, which is linked to lower anxiety. Spiegel's public explanation is that anxiety builds on itself like a snowball and a controlled breath interrupts the momentum. In the trial, cyclic sighing produced the largest daily rise in positive affect and the largest drop in resting respiratory rate; heart rate and heart rate variability did not change significantly, so the vagal explanation was not confirmed by the trial's own measures. [source: Balban et al. 2023; Stanford Medicine News Center 2023] [uncertain: mechanisms are proposed, not demonstrated]

Scope for psychotherapists. This is a self-administered breathing exercise that a trained psychotherapist may teach within their competence as a coping and self-regulation skill. It is not a treatment for any physical disease, it is not a substitute for respiratory physiotherapy where a breathing pattern disorder is present, and clients must never change asthma or other medication because of it. [synthesis]

2. Framework

Cyclic sighing has no phased therapy model. The published structure is a single practice protocol; the phases below are the clinical wrapper a psychotherapist puts around it. [source for protocol: Balban et al. 2023] [synthesis for the clinical phases]

Published protocol (attributed fact, Balban et al. 2023):

  • Format: self-administered at home, remote study, instruction delivered by pre-recorded video sent a few days before starting, plus daily text reminders.
  • Dose: 5 minutes per day, timed with a timer, for 28 days.
  • Posture: seated in a chair, or lying down if preferred.
  • Pattern: inhale slowly; once the lungs are expanded, inhale once more (the second inhale is normally briefer); then slowly and fully exhale all the breath. Repeat this pattern for the 5 minutes. Inhales ideally through the nose and the exhale through the mouth; all-nasal breathing was acceptable.
  • No count, ratio or pace was specified beyond "slowly", and there is no breath hold in this practice.
  • Mood and state anxiety were rated before and after each daily session.

Clinical wrapper [synthesis]:

  • Phase A, screening and consent (part of a session): physical and psychiatric screening against the trial's exclusions and breathwork cautions (section 4.1); move on only when no exclusion applies or clearance has been obtained.
  • Phase B, rationale and in-session teaching (one session, 10–20 minutes of it): psychoeducation, a demonstration, a guided practice, and a debrief. Move on when the client can perform the pattern without dizziness or strain and reports it as tolerable.
  • Phase C, home practice (about four weeks, mirroring the trial): the published 5 minutes a day, with a brief before/after rating; reviewed at each session. The practice is not lengthened, intensified or progressed beyond the published pattern.
  • Phase D, review and ending: after about four weeks, review ratings and experience together; decide whether to continue, use it only as needed, or stop. The trial gives no data beyond four weeks.

Criteria for moving on or stopping: unpleasant hyperventilation-type symptoms, panic, or worsening of a breathing complaint mean stopping and reassessing, not pushing through (section 4.1). [synthesis]

3. Guiding principles

  • Follow the published pattern, attributed, and do not embellish it. Practical meaning: teach the double inhale and slow full exhale for 5 minutes; do not add holds, counts, faster rounds or longer sessions, and say "this is the version used in the Stanford study". [source: Balban et al. 2023]
  • Gentle, not forced. Practical meaning: the exhale is slow and complete but never strained; the second inhale is small; any dizziness means slowing down or pausing. [synthesis]
  • Screen before teaching. Practical meaning: check for the conditions the trial excluded and the general breathwork cautions before the first practice; obtain medical clearance where in doubt. [source: Balban et al. 2023; Healthify] [synthesis]
  • The client stays in control. Practical meaning: eyes open or closed is their choice, they can stop at any moment, and they are told so before starting. [synthesis]
  • A coping skill, not a cure. Practical meaning: present it as one tool with promising early evidence from a single, small, mostly student trial; make no promises about anxiety disorders or physical health. [source: Balban et al. 2023 limitations; Stanford Medicine Magazine 2023]
  • Watch the function of the skill in anxiety work. Practical meaning: in panic and anxiety treatment, notice whether breathing is being used to escape feared sensations; if so, plan it so it does not undermine exposure work. [source: Meuret et al. 2003; Pompoli et al. 2018] [synthesis]
  • Respect breathing-pattern problems. Practical meaning: frequent sighing is itself a recognised feature of dysfunctional breathing; for those clients, refer for respiratory physiotherapy rather than adding more deliberate sighs. [source: Healthify] [synthesis]
  • Measure lightly. Practical meaning: a quick before/after rating keeps practice honest and shows the client their own data, echoing the trial's daily ratings. [source: Balban et al. 2023] [synthesis]
  • Safe settings only. Practical meaning: never practise while driving, operating machinery, or in or near water, even though this is a low-intensity practice. [synthesis based on standard breathwork safety guidance]

4. Intervention inventory

The trial protocol contains one technique (cyclic sighing practised daily). Items 4.1–4.2 and 4.4–4.8 are the clinical procedures needed to deliver it safely; they are marked [synthesis] where no cyclic-sighing source describes them. [source: Balban et al. 2023]

4.1 Screening, contraindications and consent

  • Purpose / when: before the first practice, and again if health status changes.
  • Steps:
    1. Ask about the conditions the Stanford trial excluded: moderate to severe psychiatric or medical conditions, including heart disease, glaucoma, seizures, pregnancy, psychosis, suicidality, bipolar disorder and substance use disorders, plus significant vision or hearing impairment. The trial was not designed to test safety in these groups, so evidence for them is absent. [source: Balban et al. 2023]
    2. Ask about the wider cautions that apply to breathwork: cardiovascular disease or uncontrolled blood pressure, epilepsy, pregnancy, psychosis or bipolar disorder, panic disorder, glaucoma or retinal problems, recent surgery, and asthma or COPD. For any of these, get the client's physician or respiratory clinician to agree before teaching; for asthma or COPD, make clear the exercise does not replace any inhaler or medication. [synthesis based on standard breathwork cautions] [uncertain: no cyclic-sighing-specific contraindication list is published]
    3. Screen for a breathing pattern disorder. Ask about breathlessness or air hunger at rest, frequent sighing or yawning, upper-chest breathing, tingling in fingers or around the mouth, dizziness, chest tightness and palpitations. Consumer-health guidance for clinicians lists "periodic deep sighing" itself as a disordered pattern. An abnormal Nijmegen Questionnaire result "may be an indication" but is not diagnostic; use the licensed instrument as published (items are not reproduced here), and refer for medical review and respiratory physiotherapy where it suggests a breathing pattern disorder. [source: Healthify, Breathing pattern disorders]
    4. Rule out new or unexplained breathlessness. Sudden shortness of breath needs urgent medical help, not a breathing exercise; persistent breathlessness needs medical assessment for asthma, cardiac, lung, blood or other causes first. [source: Healthify]
    5. Explain hyperventilation effects. Over-breathing lowers carbon dioxide and can cause dizziness, light-headedness, tingling, cold hands, chest tightness, palpitations, visual blurring, muscle cramping of hands and feet (carpopedal spasm) and panic-like feelings. Cyclic sighing is slow, but repeated deep inhalations taken too fast can still produce these. [source: Healthify] [synthesis for the application to cyclic sighing]
    6. Agree stop rules and settings: stop at dizziness, tingling, chest pain, breathlessness, panic, or any distress; return to ordinary breathing; never practise while driving, operating machinery, or in or near water (bath, pool, open water). [synthesis]
    7. Obtain consent: what the exercise involves, that evidence is limited to one small trial, that the client may stop at any moment.
  • Example language (illustrative): "Before we try a breathing exercise I need to check a few health questions, because some conditions make breathing practices unsuitable without your doctor's agreement." / "If you feel dizzy, tingly, short of breath or panicky at any point, just stop and breathe normally. Stopping is always fine."
  • Watch for: clients who minimise symptoms to please you; breathlessness that has not been medically assessed; a history of panic in which deep breathing is itself a trigger; frequent spontaneous sighing. Working if: the client can name the stop rules and has any needed clearance in place.

4.2 Rationale and psychoeducation

  • Purpose / when: at first introduction, so the client understands what the exercise is and is not.
  • Steps:
    1. Link it to the client's own stress pattern: anxious arousal speeds breathing and heart rate and tightens muscles, and noticing these sensations can make anxiety grow. [source: Stanford Medicine Magazine 2023]
    2. Describe the natural sigh: a larger breath the body produces by itself several times an hour; the exercise copies its shape deliberately. [source: University of Alberta / UCLA news 2016]
    3. Give the developers' explanation as hypothesis: slow, long exhales are thought to calm the body, and choosing how you breathe can give a sense of control. [source: Balban et al. 2023; Stanford Medicine News Center 2023]
    4. Give the evidence honestly: one randomised trial of about 108 mostly student volunteers, 5 minutes a day for a month; all groups including mindfulness improved, and cyclic sighing showed the largest gain in positive mood and the largest fall in resting breathing rate. [source: Balban et al. 2023]
    5. Invite questions and doubts.
  • Example language (illustrative): "When we're stressed, breathing tends to speed up and stay high in the chest. This exercise uses the shape of a natural sigh, two breaths in and a long, slow breath out. A Stanford study found five minutes a day lifted mood a little more than mindfulness did. It's one small study, so let's treat it as an experiment and see what it does for you."
  • Watch for: expectations of a cure; catastrophic beliefs about breathing ("if I don't breathe right I'll suffocate"), which are worth noting for later cognitive work. Working if: the client can describe the exercise in their own words and holds realistic expectations.

4.3 Teaching cyclic sighing in session (the published pattern)

  • Purpose / when: once screening is clear; teach in session before any home practice.
  • Steps (pattern as published by Balban et al. 2023; teaching sequence [synthesis]):
    1. Seat the client comfortably in a chair (or lying down if they prefer and it is appropriate in your setting). Eyes open or closed as they choose.
    2. Demonstrate one cycle yourself, slowly.
    3. Guide: inhale slowly, ideally through the nose; once the lungs feel full, take one more inhale on top (it is normal for this second inhale to be shorter); then exhale slowly and fully through the mouth until the breath is out. Breathing in and out entirely through the nose is also acceptable.
    4. Let the client repeat the pattern at their own unhurried pace for a few cycles; there is no count, ratio or hold to follow.
    5. If tolerated, practise together for up to the published 5 minutes with a timer, or for a shorter period in the first session if the client prefers; do not exceed 5 minutes. [synthesis on the shorter first session]
    6. Debrief: what they noticed in body, breath and mood; any dizziness, tingling, effort or discomfort.
  • Example language (illustrative): "Breathe in slowly through your nose… and when you're full, sip in a little bit more… now let it all go, slowly, out through your mouth… no hurry, no counting. Just repeat that shape." / "What did you notice, in your body and in your mood?"
  • Watch for: breathing faster to "do it right"; a forceful or gasping second inhale; forcing the exhale until strained; raised shoulders; light-headedness or tingling (slow down, make the inhales smaller, or stop). Spiegel notes some people feel calmer after one or two sighs, with fuller effect over the 5 minutes. Working if: the client performs the pattern smoothly and reports neutral or calmer feelings without hyperventilation symptoms. [source: Stanford Medicine News Center 2023] [synthesis for troubleshooting cues]

4.4 Daily home practice (published dose)

  • Purpose / when: after successful in-session practice; the trial format.
  • Steps:
    1. Agree a time and place: seated or lying, somewhere quiet and safe (never while driving, operating machinery, or in or near water, including the bath).
    2. Set a timer for 5 minutes and repeat the pattern from 4.3 for that time, once a day. This is the published dose; there is no target beyond it and no lengthening or progression. [source: Balban et al. 2023]
    3. Use the brief before/after rating (4.5).
    4. Support recall with a reminder of the client's choosing; the trial used daily text reminders. [source: Balban et al. 2023]
    5. Review at each session for about four weeks, the period the trial covered.
  • Example language (illustrative): "Between now and next time, would you be willing to try the five minutes once a day, the same way we did here? Missing days is fine; we'll just look at what happened."
  • Watch for: practice in unsafe settings; practice turning into a rigid ritual or a source of self-criticism; symptoms appearing at home that did not appear in session. Trial participants averaged around five to six minutes a day on roughly 18–20 of 28 days, so partial adherence is normal. In the trial, cyclic sighers benefited more the more days they practised. Working if: practice happens on most days and before/after ratings show small, consistent improvements. [source: Balban et al. 2023]

4.5 Before/after self-rating

  • Purpose / when: with every home and in-session practice, to show the client its effect on them.
  • Steps: the trial used standardised affect and state-anxiety questionnaires before and after each session; in clinic, a simpler record is enough: rate calm, mood or anxiety on a 0–10 scale before and after, and note any side effects. Do not reproduce licensed questionnaires; use them only as their publishers permit. [source: Balban et al. 2023 for the trial measures] [synthesis for the clinical version]
  • Example language (illustrative): "Just before you start, give your tension a number from 0 to 10, and again straight after. Jot down anything odd, like feeling dizzy."
  • Watch for: ratings that worsen after practice (review technique or stop); a focus on numbers that increases monitoring of anxious symptoms. Working if: the client can see a pattern in their own ratings.

4.6 Brief in-the-moment use

  • Purpose / when: to interrupt rising stress in daily life. Spiegel describes it as stopping the momentum of an anxiety spiral and says some people feel calmer after one or two sighs. The trial tested scheduled daily practice, not in-the-moment use, so evidence for this use is indirect. [source: Stanford Medicine News Center 2023; Stanford Medicine Magazine 2023] [uncertain]
  • Steps: once the daily practice is familiar, the client takes a few slow cyclic sighs in a safe moment (seated, not driving), then returns to ordinary breathing and to what they were doing. [synthesis]
  • Example language (illustrative): "When you notice the stress building, see if a couple of those double-in, long-out breaths help you step off the escalator."
  • Watch for: use as an escape from every flicker of anxiety (see 4.7); rapid repeated sighs that tip into over-breathing. Working if: a few breaths take the edge off and the client carries on with the activity rather than leaving it.

4.7 Use within CBT for anxiety and panic

  • Purpose / when: when the client is in treatment for an anxiety disorder or panic.
  • Background: a review of breathing training for panic disorder found the evidence inconclusive and argued that dropping it entirely was premature. A component network meta-analysis of CBT for panic found breathing retraining improved acceptability with only minor effects on efficacy, while interoceptive exposure improved both. Clinicians also debate whether breathing techniques used to avoid feared sensations act as safety behaviours that maintain fear. [source: Meuret et al. 2003; Pompoli et al. 2018] [uncertain: the safety-behaviour debate is summarised from general CBT practice, not from these two sources]
  • Steps [synthesis]:
    1. Note that the trial excluded moderate to severe psychiatric conditions, so cyclic sighing's effect in diagnosed anxiety disorders is untested. [source: Balban et al. 2023; Stanford Medicine News Center 2023]
    2. Frame it as a daily regulation practice, not a rescue from panic sensations.
    3. During interoceptive or in-vivo exposure, do not use it to end exposure early; if it is used, plan when, and check that the client is still learning that the sensations are safe.
    4. In panic disorder, introduce it carefully, since deep breaths can produce the very sensations the client fears; screen as in 4.1 and consider whether exposure should come first.
  • Example language (illustrative): "Let's keep the breathing for your daily five minutes. During our exposure practice, I'd like you to let the sensations come and go without using it, so you can learn they pass on their own."
  • Watch for: belief that the breathing is what prevents a catastrophe; increasing reliance; avoidance of situations unless the breathing "works". Working if: practice coexists with continuing exposure gains and reduced avoidance.

4.8 Review and ending

  • Purpose / when: after about four weeks of practice (the trial period), or earlier if problems arise.
  • Steps [synthesis]: review ratings, adherence, side effects and the client's view; decide together to continue daily, keep it for occasional use, switch to another regulation skill, or stop. Remind the client that there are no data beyond four weeks and that the practice should not be lengthened or intensified. [source: Balban et al. 2023 for the four-week limit]
  • Example language (illustrative): "Looking back over the month, what difference has it made? Is it worth keeping, and in what form?"
  • Watch for: pressure to keep going when it brings no benefit; a wish to "step up" to more intense breathwork, which needs separate screening and qualified instruction. Working if: the client has an informed, freely chosen plan.

4.9 The trial's comparison practices (for orientation only)

  • Purpose: clients who have heard of the study may ask about the other arms. Box breathing used equal-length inhale, hold, exhale and hold phases set from a self-test; cyclic hyperventilation with retention used rounds of deep, rapid inhalations with passive exhalations, each round ending in a breath hold on empty lungs; the control was a focused-attention mindfulness meditation. [source: Balban et al. 2023]
  • Steps: describe these only; this guide gives no instructions for them. For box breathing see the separate box breathing guide in this domain. Cyclic hyperventilation with retention involves over-breathing and breath holds and carries the hyperventilation risks in 4.1, including blackout risk in or near water and while driving; refer to properly qualified instruction and medical clearance rather than teaching it. In the trial it gave the smallest fall in state anxiety of the breathwork arms. Intense breathwork such as Holotropic Breathwork is covered in Domain VII. [source: Balban et al. 2023] [synthesis]
  • Example language (illustrative): "That study tested several methods; the one we're using had the clearest benefit and is the gentlest. The rapid-breathing one carries more risks and isn't something I'd teach."
  • Watch for: clients experimenting with hyperventilation or breath holds after watching online videos; ask about this directly. Working if: the client understands why the gentlest arm was chosen.

5. Handling problems

The trial reports no formal adverse events; in its debrief survey 90% of participants reported positive and 10% negative experiences with the exercises, and the nature of the negative experiences is not described in the main text. Almost everything below is therefore clinical synthesis. [source: Balban et al. 2023] [synthesis]

  • Dizziness, tingling, light-headedness during practice. Likely over-breathing. Stop and return to ordinary breathing; when symptoms have settled, resume (if at all) with smaller, slower inhales and an unforced exhale. If symptoms recur, stop the exercise and consider medical review and respiratory physiotherapy assessment. [synthesis; symptom profile from Healthify]
  • Chest pain, sudden breathlessness, fainting, palpitations that do not settle. Stop the exercise and follow medical-emergency procedures; sudden shortness of breath needs urgent medical help. Do not resume without medical assessment. [source: Healthify] [synthesis]
  • Panic triggered by the breathing. Stop, ground the client in the room (feet on floor, eyes open, naming objects), breathe normally. Afterwards explore what sensation set it off; in panic disorder, this may belong in planned interoceptive exposure rather than in breathwork. [synthesis]
  • Frequent sighing or air hunger as a complaint. Do not add deliberate sighing; refer for medical review and respiratory physiotherapy (breathing retraining for breathing pattern disorders), and address anxiety and stress psychologically alongside. [source: Healthify] [synthesis]
  • Trauma-related distress (e.g. breath focus or closed eyes bringing up memories or dissociation). Offer eyes open, a shorter practice, an external focus, and an explicit choice to stop; return to stabilisation work. Choking, suffocation or strangulation histories may make breath work aversive; ask. [synthesis]
  • "It doesn't do anything." Check technique (often the exhale is short or the pace hurried); check adherence; remember that in the trial all four groups improved and differences were modest. Accept that it may simply not suit this client and offer another regulation skill. [source: Balban et al. 2023] [synthesis]
  • Non-adherence. Explore barriers non-judgementally; link practice to an existing routine; reduce the burden to fit the published dose (no more than 5 minutes). Partial adherence was the norm in the trial. [source: Balban et al. 2023] [synthesis]
  • Over-use or rigid reliance. If the client is using sighing many times a day to manage every sensation, review its function, return to the once-daily practice, and work on the avoidance pattern. [synthesis]
  • Wanting more: longer sessions, holds, rapid breathing. Explain that the published practice is 5 minutes with no holds; more intense breathwork has different risks (hyperventilation, blackout in water or at the wheel) and needs separate screening and qualified instruction. [synthesis]
  • Scepticism or rupture ("just breathing, are you not taking me seriously?"). Acknowledge it, explain it is an optional add-on to the main work, and do not press. [synthesis]
  • Physical conditions and medication. Never advise changing asthma, cardiac or other medication; the exercise is not a treatment for physical disease. Liaise with the client's physician where relevant. [synthesis]

Evidence status (to explain to clients and colleagues)

  • One randomised controlled trial: 108 adults randomised (30 to cyclic sighing), 5 minutes a day for 28 days, delivered remotely in 2020, most participants recruited from a Stanford undergraduate psychology class. Stanford's news coverage reports 111 volunteers. [source: Balban et al. 2023; Stanford Medicine News Center 2023] [uncertain: the two figures differ]
  • Results: all groups, including mindfulness, reduced state anxiety and negative affect after sessions, with no between-group difference on anxiety; breathwork overall increased positive affect more than mindfulness, and only cyclic sighing differed significantly from mindfulness; cyclic sighing gave the largest drop in resting respiratory rate; heart rate, heart rate variability and sleep did not change significantly; trait anxiety did not change. [source: Balban et al. 2023]
  • Limits stated by the authors: an exploratory study, registered retrospectively (NCT05304000); remote practice with self-reported adherence; small arms that limit comparison between breathing types; four weeks with no further follow-up. The sample excluded moderate to severe psychiatric and medical conditions, so the findings do not establish effectiveness for clinical anxiety or mood disorders. [source: Balban et al. 2023; Stanford Medicine News Center 2023]
  • Disclosure: Andrew Huberman became an advisor to WHOOP, maker of the wearable used in the trial, in June 2022; WHOOP donated the devices and, per the paper, had no role in design or analysis. [source: Balban et al. 2023]
  • Wider context: a meta-analysis of breathwork RCTs (searches to February 2022, so before this trial appeared) found small-to-medium reductions in self-reported stress versus controls, mostly from studies at moderate risk of bias. [source: Fincham et al. 2023]

The psychotherapist's role

  • Teach cyclic sighing as published, within your training, as one self-regulation skill inside a broader treatment plan. [synthesis]
  • Own the screening: exclusions, breathwork cautions, breathing pattern disorder signs, stop rules and safe settings (4.1). [synthesis]
  • Refer onward when needed: to the client's physician for medical clearance or unexplained breathlessness, and to a respiratory physiotherapist for suspected breathing pattern disorders. [source: Healthify] [synthesis]
  • Keep the claims honest: early evidence from one small trial in mostly student, non-clinical volunteers; no promise of curing anxiety; no claims for physical illness. [synthesis]
  • Fit it to the therapy: in anxiety and panic work, plan its use so that it supports rather than replaces exposure. [synthesis]
  • Document screening, consent, any adverse reactions, and onward referrals. [synthesis]
  • No touch is needed; teach by demonstration and verbal guidance only. [synthesis]

Sources

← Previous · X-151Slow-paced, resonance or coherent breathingNext · X-153 →4-7-8 breathing