What this page is, and what it is not
Slow breathing is not a treatment for high blood pressure. It is a quiet practice that has been studied as a possible extra alongside real treatment, and the studies disagree with one another. This page walks through what they found, including the parts that are unflattering.
If you have been told your blood pressure is high, the things that matter most are the unglamorous ones: taking the medication you were prescribed, salt, alcohol, weight, movement, sleep, and having your readings taken properly by someone who knows what they are looking at. Breathing exercises sit at the edge of that list. They are free, they are pleasant, and on the current evidence they are not a replacement for any of it.
The timer above runs coherent breathing: five seconds in, five seconds out, six breaths a minute, no holds anywhere. That pace is what most of the blood-pressure literature means when it says "slow breathing" — the 2019 review below counted anything at ten breaths a minute or fewer. Ten minutes is the default here because it is a length most people can keep up daily without dreading it, not because a trial established it as the right dose. If five seconds feels like a stretch, breathe more shallowly rather than forcing the count.
What the evidence shows
Here is the honest version. Several reviews have pooled overlapping sets of trials and reached different conclusions, and the difference tracks how tightly each set of trials was controlled. Each of the four reviews below was read on its PubMed record before being described here, and every sentence in quotation marks is taken word for word from that record rather than from someone else's summary of it.
The optimistic reading: about 5 mmHg, in low-risk people
A 2019 systematic review and meta-analysis in Complementary Therapies in Medicine pooled 17 randomised controlled trials of slow breathing — device-guided and pranayama-style both — in people with hypertension or prehypertension. It reported that "slow breathing decreased SBP by -5.62 mmHg [-7.86, -3.38] and DBP by -2.97 mmHg". Its authors concluded that slow breathing "may be a reasonable first treatment for low-risk hypertensive and prehypertensive patients who are reluctant to start medication." Read that framing closely: low cardiac risk, and reluctant to start medication. It is not a claim about coming off medication someone is already taking. It is also the review that says most plainly what it counted as a practice: trials of "slow breathing at ≤10 breaths/minute for ≥5 min on ≥3 days/week" with a "total intervention duration of ≥4 weeks". Those were thresholds for including a study, not a prescription for a person.
Chaddha A, Modaff D, Hooper-Lane C, Feldstein DA. Device and non-device-guided slow breathing to reduce blood pressure: a systematic review and meta-analysis. Complement Ther Med. 2019;45:179–184. PMID 31331557. doi:10.1016/j.ctim.2019.03.005
The sponsorship problem
A 2012 review in the Journal of Hypertension looked only at device-guided breathing, pooling eight trials of one commercial device across 494 adults. It found a reduction: systolic down 3.67 mmHg (95% confidence interval −5.99 to −1.39) and diastolic down 2.51 mmHg. Then the authors did something worth respecting. They re-ran the analysis without the trials the device manufacturer had funded or taken part in. In their words, "sensitivity analysis was carried out excluding the five trials sponsored by or involving the manufacturers of the device, which revealed no overall effect on BP using the device." Their conclusion asked for longer, independent trials.
Mahtani KR, Nunan D, Heneghan CJ. Device-guided breathing exercises in the control of human blood pressure: systematic review and meta-analysis. J Hypertens. 2012;30(5):852–860. PMID 22495126. doi:10.1097/HJH.0b013e3283520077
The blinded reading: no difference from a sham
In 2014, JAMA Internal Medicine published a meta-analysis of individual patient data restricted to blinded, randomised trials with an active comparator — a sham device or music therapy, rather than nothing at all. Its conclusion was flat: "Treatment with DGB did not significantly lower office blood pressure compared with a sham procedure or music therapy." The point estimate for office systolic pressure was 2.2 mmHg, and it pointed the other way, in favour of the control group. The same paper notes that device-guided breathing was, at that time, recommended by the American Heart Association for its blood-pressure-lowering effects — which is exactly why a blinded re-analysis mattered.
Landman GW, van Hateren KJ, van Dijk PR, et al. Efficacy of device-guided breathing for hypertension in blinded, randomized, active-controlled trials: a meta-analysis of individual patient data. JAMA Intern Med. 2014;174(11):1815–1821. PMID 25222103. doi:10.1001/jamainternmed.2014.4336
The most recent reading: still no
A 2022 systematic review in Health Science Reports included 22 studies covering 17,214 participants with diagnosed hypertension, and reported that device-guided slow breathing "did not significantly reduce SBP and DBP compared to usual care." It is worth seeing how thin that pooled comparison actually is underneath the large headline number. The device-versus-usual-care estimate for systolic pressure rested on 288 people and came out at −2.13 mmHg with a confidence interval running from −12.71 to 8.44 — an interval wide enough to contain a useful benefit, no effect at all, and a rise. The authors describe the heterogeneity between those studies as high. Their conclusion was that "DGSB did not reduce BP in hypertensive patients and NDGSB is a new path for the future" — meaning the non-device forms of slow breathing, the kind a free timer can offer, are where they think the open question now sits.
de Freitas Gonçalves KS, Queiroz Godoy Daniel AC, Tatagiba Lamas JL, et al. Device and nondevice-guided slow breathing to reduce blood pressure in hypertensive patients: a systematic review and meta-analysis. Health Sci Rep. 2022;5(3):e636. PMID 35601033. doi:10.1002/hsr2.636
Put the four together and a pattern appears. When trials are unblinded, or compare slow breathing against doing nothing, slow breathing looks helpful by a few millimetres of mercury. When trials are blinded and compare it against a believable placebo, most of that effect goes away. That does not prove slow breathing does nothing. It means nobody has yet separated what the breathing pattern does from what sitting quietly for ten minutes with something to focus on does. Sitting quietly for ten minutes is still worth doing. It is just not a reason to change your treatment.
You can read the same evidence written up alongside the rest of the research we rely on, with its limitations spelled out, on what the research says.
How to do it
1. Sit down, do not lie down
A chair with a back, feet flat on the floor, hands resting in your lap. If you plan to take a blood pressure reading afterwards, sit the way you would for the reading: back supported, legs uncrossed, arm supported at heart height.
2. Give it ten minutes
Press Begin. Ten minutes is the default because it is long enough to settle into and short enough to repeat daily. Both of the reviews above that set a minimum study length — the 2019 meta-analysis and the 2014 blinded one — counted only programmes running four weeks or longer, so think in weeks rather than in sessions. One session is a pleasant ten minutes. It is not an intervention.
3. Breathe in for five, through the nose
Follow the circle as it grows. Let the air go low first — the belly moves before the chest. Keep the in-breath unhurried and quiet. You are not trying to fill your lungs to capacity.
4. Breathe out for five, unforced
Let the out-breath fall rather than pushing it. Nose or lightly parted lips, whichever is more comfortable. There is no hold at either end of this pattern, which is why it suits long sits.
5. Keep it comfortable, not exact
Five seconds should feel roomy. If you find yourself gasping at the turn, take less air in and let the count be a suggestion rather than a target. Breathing more deeply is not the goal. Breathing more slowly and more evenly is.
6. Stop if you feel light-headed
Dizziness, tingling in the hands or around the mouth, or a fluttery feeling in the chest means you are taking in more air than you need. Stop, breathe normally, and rest until it passes. If it keeps happening, leave this alone and mention it to your doctor.
When to use it, and when not to
Reasonable uses. As a daily ten-minute quiet practice, because slowing down is good for you regardless of what it does to a number. To settle yourself in a waiting room. As part of a wind-down at night. As something to do with your hands and attention that is not scrolling.
Not reasonable uses. As a reason to skip a dose. As a substitute for getting a high reading looked at. As a way to bring a number down just before a clinic appointment — that hides information your clinician needs to treat you properly, and a flattering reading is worse than an honest one.
Not a use at all. If you have symptoms, this page is the wrong place to be. Chest pain or pressure, sudden severe headache, breathlessness at rest, vision changes, weakness or numbness on one side, trouble speaking, or confusion all mean you call your local emergency number. A very high reading with any of those symptoms is an emergency, not a breathing exercise.
Who should be careful
Coherent breathing at six breaths a minute involves no breath-holding and no fast breathing, which makes it one of the gentler patterns on this site. It is still worth checking with your doctor or pharmacist before you start anything new if any of the following apply to you.
- Your blood pressure is not yet controlled, or your clinician has told you it is being actively adjusted.
- You have a heart condition, an arrhythmia, or a history of stroke or heart attack.
- You have a respiratory condition such as asthma or COPD, or you get breathless easily.
- You are pregnant, or recently gave birth.
- You have had recent surgery, or you have glaucoma or raised intracranial pressure.
- You have panic disorder, or paying attention to your breathing tends to make you more anxious rather than less.
- You take medication that lowers your blood pressure enough that you sometimes feel faint on standing.
None of that is a reason to be frightened of breathing slowly. It is a reason to ask someone who knows your history before you make it a habit. Whatever you decide, do not use any breathing pattern that involves holding your breath or breathing quickly if your blood pressure is uncontrolled — that includes the energizing pattern on this site.
Nothing here changes your medication
This deserves its own heading because it is the one thing on this page that can genuinely harm someone. Antihypertensive medication works by being taken consistently. Blood pressure that falls while you are on treatment is usually evidence that the treatment is working, not evidence that you no longer need it. Stopping or halving a dose on your own can send pressure back up, sometimes sharply.
If you want to take less medication, that is a fair goal and a normal conversation to have. Bring it to the person who prescribed it, with a record of readings taken properly at home over a couple of weeks. They can tell you whether a change is safe and how to make it. A breathing timer cannot.
If you want to track it honestly
Should you decide to see whether a daily practice makes any difference to your own numbers, measure in a way that will not fool you. The US Centers for Disease Control and Prevention sets out the basics: "Sit in a comfortable chair with your back supported for at least 5 minutes before your reading", "Put both feet flat on the ground and keep your legs uncrossed", have the cuff "against your bare skin, not over clothing", and "Take at least two readings, 1 or 2 minutes apart."
Two things beyond that are worth doing. Take your monitor to an appointment once and have someone check it against theirs, so you know whether to trust it. And write down every reading rather than the nicest one, at the same times each day, then compare a full week against a full week instead of a single reading against a single reading.
Centers for Disease Control and Prevention. Measuring Your Blood Pressure.
Expect the honest answer to be "hard to tell". Blood pressure moves a great deal from hour to hour and day to day, and the effect sizes under discussion here are small enough to disappear inside that noise. Take your log to your next appointment either way. It is useful to your clinician regardless of what the breathing did.
Common questions
Can breathing exercises lower high blood pressure?
The evidence is genuinely mixed. Reviews that pool unblinded trials report small reductions of roughly 3 to 6 mmHg systolic. The reviews that restricted themselves to blinded trials with a sham or music comparator found no significant difference. Slow breathing is not established as a treatment for hypertension, and no serious source recommends using it in place of medical care.
Should I stop or reduce my blood pressure medication if breathing helps?
No. Never change, reduce, skip or stop a prescribed medication because of anything on this site or because a home reading looked better. Dose decisions belong to the clinician who prescribed it, working from readings taken properly over time. Stopping antihypertensive medication without medical supervision can be dangerous.
How long and how often would I have to practise?
There is no established dose, and this is one of the softer parts of the evidence. The closest thing to one is a set of inclusion criteria rather than a recommendation. The 2019 meta-analysis counted only trials in which people breathed at ten breaths a minute or fewer, for at least five minutes a session, on at least three days a week, over at least four weeks. The 2014 blinded meta-analysis likewise counted only trials running four weeks or longer. Those are the thresholds the researchers used to decide what to include, not a dose anyone has shown to work. Ten minutes is the default here because it is a length most people can repeat daily. Nobody has shown that a single session changes anything lasting.
Is slow breathing safe if I have high blood pressure?
For most people, breathing at about six breaths a minute with no breath-holding is gentle and well tolerated. Stop if you feel dizzy, tingly, breathless or unwell, and speak to your doctor before starting any new practice if you have a heart or lung condition, are pregnant, or have been told your blood pressure is not yet controlled.
Do I need a device like the ones used in the studies?
The device trials used a commercial machine that set the breathing pace with tones. The timer on this page sets a fixed pace instead of adapting to you, so it is not the same intervention and no claim here transfers to it. Given that the blinded trials of the devices themselves came out null, buying one is hard to justify on the current evidence.
What should send me to a doctor instead of a breathing timer?
Any new or unusually high reading, any reading your clinician told you to report, and any symptoms at all. Chest pain, sudden severe headache, breathlessness, vision changes, weakness or numbness on one side, difficulty speaking or confusion are emergencies. Call your local emergency number rather than opening a breathing app.
A note on what this is
Help Me Breathe is a breathing timer (three sessions free, then one plan) written by Georges Rayess, who is not a clinician. Nothing here is medical care, nothing here diagnoses or treats anything, and none of it replaces the advice of a qualified professional who knows your history. Read the medical disclaimer before you rely on any of it.