Anuloma Viloma Timer
Slow yogic breathing lowered resting heart rate across five randomised trials and 439 people with high blood pressure.
That is the firmest number on the desk this week. It comes from a systematic review and meta-analysis in the Indian Heart Journal (January 2026) that pooled seven randomised controlled trials and 683 participants in all. For heart rate, the standardised mean difference was -0.43, with a 95% confidence interval of -0.52 to -0.34: a moderate effect, and nowhere near zero. Three of the seven trials tested alternate nostril breathing. Two tested Sheetali, the cooling breath, one tested Bhramari, and one tested general yogic breathing.
Blood pressure was less tidy. Systolic and diastolic readings fell significantly under one statistical model and were only marginal under the more conservative one, which is the model you would usually trust when the pooled trials differ from each other. The authors’ own summary is the right one to carry around: pranayama lowers heart rate, may modestly lower blood pressure, and is a safe addition to standard care.
What a meta-analysis cannot easily tell you is the thing a reader most wants to know on the way to the cushion: how much, and how often. Pooling trials smears the dose. But reading this week’s studies side by side, two kinds of frequency keep coming up. One is the pace of the breath. The other is how many times a day people actually sat down to practise. The first is reasonably well settled. The second is a pattern that nobody has tested head to head, and it may matter more than how long each session lasts.
Frequency one: the pace
The cleanest evidence on pace is a four-week randomised comparison in Scientific Reports (2026). Eighty-eight people were split between breathing at their individually calibrated resonance frequency, breathing at a fixed 0.1 Hz (six breaths a minute for everyone), and a control group. Both breathing groups improved on stress, anxiety and depressive symptoms against control, and they improved by about the same amount. A Bayesian analysis favoured there being no meaningful difference between the tuned pace and the fixed one.
In plain terms: you do not need a lab session to find your number. Six a minute is close enough for most people.
A very different kind of study points the same way from the far end of biology. A Fudan University team, publishing in PNAS this summer, traced a pathway in mice running from sensory neurons in the nose, through the olfactory bulb and perirhinal cortex, to the amygdala. Slow stimulation of the nasal nerves (2 Hz) reduced anxiety-like behaviour. Fast stimulation (7 Hz) increased it. This is a mouse study, and a mouse breathes far faster than you do, so none of those numbers transfer to a human practice. What transfers is the principle: the speed of air through the nose is not neutral. The same channel carries a calming signal when it is slow and an alarming one when it is fast.
And from the human behaviour side, the Stanford trial published in Cell Reports Medicine in 2023 (Balban and colleagues) compared four five-minute daily practices over a month: cyclic sighing, box breathing, cyclic hyperventilation and mindfulness meditation. All four reduced anxiety to a broadly similar degree. Cyclic sighing, the one built around a long, slow exhale, produced the largest daily lift in positive mood and the biggest drop in resting breathing rate.
Put those together and the pace question has a workable answer. Slow, around six breaths a minute, through the nose, with the exhale never rushed. Alternate nostril breathing lands there almost by accident: with a count of five on each half, one full round takes twenty seconds, which is three rounds and six breaths a minute.
Frequency two: the schedule
Now look at how this week’s clinical trials actually delivered the practice, rather than what they found.
Sleep apnea. A randomised trial in BioPsychoSocial Medicine (2026) gave adults with mild to moderate obstructive sleep apnea eight weeks of home pranayama: 15 minutes, three times a day, every day, with some sessions supervised online by a physiotherapist. Among the 34 people analysed, the pranayama group’s Pittsburgh Sleep Quality Index fell from 9.12 to 6.88 (lower is better), while the control group did not improve. Daytime sleepiness, depression and anxiety scores also favoured the pranayama group.
Dialysis. A quasi-experimental study in the Caspian Journal of Internal Medicine (2026) gave 30 hemodialysis patients Sama Vritti, plain equal breathing, twice a day for two months, and compared them with 30 on routine care. Insomnia severity and fatigue were both significantly better in the breathing group afterwards.
Hypertension. A randomised trial by Mittal and colleagues in Annals of Neurosciences (2025) added ten minutes of nadi shodhana a day to standard non-drug care for six weeks. Systolic and diastolic pressure both fell.
Mood. The Stanford trial: five minutes a day.
Set against those, the week’s other trial looks different. In Applied Psychophysiology and Biofeedback (September 2026), 88 teenage athletes were randomised to a wait-list or to 60 minutes a day, six days a week, of postures, pranayama and meditation combined. Depression, anxiety and stress scores all improved by a moderate margin (Hedges’ g of 0.62, 0.56 and 0.55). It worked, but it was a large, multi-part dose, and nobody can say how much of it was the breath.
So here is the pattern. The trials that moved sleep and fatigue in patient populations did not ask for one long sitting. They asked for short sessions, repeated through the day. The sleep apnea trial’s 45 daily minutes came in three pieces, not one block.
It is worth being plain about how far that pattern goes, which is not very far. No study this week compared three short sessions with one long one. The samples were small, the sleep apnea trial was open-label, the dialysis study was not randomised, and in both the control groups received nothing extra, so attention alone may explain part of the effect. This is a hypothesis drawn from how the trials happened to be designed, not a finding any of them produced.
It is still a plausible one. Every session is a repetition of the same skill: noticing the body is braced, and shifting it down. A skill practised three times a day gets rehearsed three times as often, and it gets rehearsed in different states (groggy in the morning, scattered in the afternoon, wired at night) rather than in the one calm slot where it is easiest. The athletes’ trial makes the same point from another direction: a breathing technique you can call on under pressure is one you have practised often when nothing was at stake.
The wellness industry has drifted toward the same idea without the trials. The Global Wellness Institute’s Breathe Initiative trends for 2026 includes “micro-dosing breath”: resets lasting from thirty seconds to a few minutes, built into the transitions of an ordinary day. The label is new. Three short sittings a day is very close to how pranayama has traditionally been taught.
The trend reaches for tools
The broader conversation this week is about regulation. The youth-research firm Ypulse put it bluntly: “Gen Z’s new self-care is regulating their nervous system.” On TikTok, #nervoussystemhealing sits at around 230,000 videos, and Dazed reported that a single video about icing the side of the neck to calm down passed 3 million views.
Most of what circulates under that banner is a one-off: an ice pack, a tapping sequence, an ear massage. Some of it may help in the moment. Very little of it is something you would do three times a day for eight weeks, which is roughly the dose the sleep apnea patients kept up. The breath is the one regulation tool that sits comfortably at both frequencies at once: slow enough within each minute, and portable enough to repeat across the day without carrying anything.
This week’s podcast episode, Dual-Zone Control: The New Science of Alternate Nostril Breathing, spends its time on the left-and-right question, what each side of the practice is doing. This piece is the timing half of the same story.
What this means for your next session
Tonight: get the pace right. Sit upright, shoulders soft. Close the right nostril and breathe in through the left for five. Switch and breathe out through the right for five. In through the right for five, switch, out through the left for five. That is one twenty-second round. Do fifteen of them, which is five minutes at six breaths a minute. If a count of five feels like strain, use four; a slightly faster pace that you can hold calmly beats a perfect one you are fighting.
Once that feels easy, shift the count to four in and six out. The breath still takes ten seconds, so you stay at six a minute, but the exhale now carries more of it, which is the direction both the Stanford result and the old teaching point.
Tomorrow: split the dose. Instead of one longer sitting, try three short ones tied to moments that already exist in your day: after you wake, mid-afternoon, and before bed. Five to ten minutes each is plenty. Keep it up for two weeks and track something you can actually notice, such as how long it takes to fall asleep or how heavy the mid-afternoon feels, rather than a number on a wearable.
If you are managing sleep apnea, kidney disease, high blood pressure or any other diagnosed condition, every study above treated breathing as an addition to care, never a replacement. Keep your CPAP, your dialysis schedule and your medication, and tell your clinician what you are adding.
The hardest part of the three-a-day version is not the breathing. It is counting while switching nostrils, and remembering to start. Anuloma Viloma Timer is built for exactly that: set the counts once and it paces each round for you, so your attention can stay on the breath. It is free on the App Store, no subscription necessary.
The pace has been found more than once, by labs and by traditions. The schedule is the part still being worked out. Until a trial settles it, the reasonable bet is the one this week’s patients made: slow breaths, short sittings, and more than one of them a day.