There is no single best position to sleep in for sleep apnea, because many but not all people with obstructive sleep apnea have a position-dependent form in which breathing events cluster on the back. In a chart review of 248 patients with sleep apnea, it was found in 49.5% of mild cases, 19.4% of moderate cases and 6.5% of severe cases, while a literature review puts the average at about 56%. In a small trial of 38 patients with position-dependent apnea, a device that kept people off their backs brought the index below 5 events per hour in 92%, against 97% with CPAP. For plain snoring the evidence is thinner: in 20 heavy-snoring men, side sleeping plus a nasal spray did not reduce snoring frequency. Whether yours is positional is something a sleep study answers; your own recordings can show whether snoring clusters on the back.
Ask ten people what the best position to sleep in for sleep apnea is and nine will say the side. They are not wrong, they are just answering a different question: which position is best for the people it works for. For some of them it works well. For others it changes almost nothing, and they spend months wedged against a pillow wondering why they still wake up gasping.
The useful question is therefore not which position, but whether your sleep apnea is the kind that cares. That has a name, a definition and a prevalence, and one part of it you can check yourself with a phone on the nightstand.
Best position to sleep in for sleep apnea: it depends on whether yours is positional
Sleep doctors call it position-dependent or positional sleep apnea. The working definition used in studies is plain: the apnea-hypopnea index, the number of breathing stops and shallow breaths per hour, drops by more than half when you are not lying on your back. In the stricter version used in one chart review it also has to fall back to a normal range off the back.
The usual explanation for why the back is worse is mechanical: lying face up lets the tongue and the soft tissue at the back of the throat sag toward the airway. I am giving that as the textbook explanation, not as something the studies below prove. What they do measure is the pattern, and the pattern is real.
How common is position-dependent sleep apnea?
The most quoted number comes from a review of sleep studies at two centers in Buffalo, New York, covering 248 patients with sleep apnea. The share with position-dependent apnea fell steeply as severity rose: 49.5% of mild cases (49 of 99), 19.4% of moderate cases (14 of 72) and 6.5% of severe cases (5 of 77). The authors concluded that it is common, particularly in mild disease.
Mador MJ, Kufel TJ, Magalang UJ, Rajesh SK, Watwe V, Grant BJ. Prevalence of positional sleep apnea in patients undergoing polysomnography. Chest 2005. PMID 16236865
Another detail in the same paper matters for anyone hoping a single sleep study will settle this. Enough sleep in both back and non-back positions to judge the question (more than 15 minutes in each) was missing in 38.7% and 33.1% of the overnight studies at the two centers, and in 82.5% of split-night studies. A review of the wider literature puts the average at about 56% of people with obstructive sleep apnea having a position-dependent form, using the same fifty-percent rule.
Ravesloot MJ, van Maanen JP, Dun L, de Vries N. The undervalued potential of positional therapy in position-dependent snoring and obstructive sleep apnea-a review of the literature. Sleep Breath 2013. PMID 22441662
Put the two together and you get a sensible reading: if your apnea is mild, a good chance exists that position matters; if it is severe, position is probably a smaller part of the story. Neither statement is about you until someone has measured you.
What staying off the back achieved compared with CPAP
The cleanest head-to-head I could verify is small. Thirty-eight patients (average age 49, average BMI 31) with position-dependent apnea tried both a positional device, worn to keep them off their backs, and CPAP. The share whose apnea-hypopnea index normalized below 5 events per hour was 92% with positional therapy and 97% with CPAP, a difference the authors did not find significant. The median index fell from 11 events per hour to 2 with the positional device and to 0 with CPAP, and that gap between 2 and 0 was statistically significant. Time spent on the back fell from a median 40% of sleep to 0%.
Permut I, Diaz-Abad M, Chatila W, Crocetti J, Gaughan JP, D'Alonzo GE, Krachman SL. Comparison of positional therapy to CPAP in patients with positional obstructive sleep apnea. J Clin Sleep Med 2010. PMID 20572416
Read the limits before the headline. These were people selected because their apnea was positional and mostly modest in size, and the device did the work of keeping them off their backs, which is not the same as deciding to sleep on your side. The 2013 literature review that covered sixteen such studies makes the same point from the other end: the effect on the apnea-hypopnea index is significant, but long-term use is the weak spot.
For the people it fits, then, position is a measurable lever and not folk wisdom. That is a long way from saying it replaces a doctor's plan.
Sleep positions for snoring: less data, same logic
The same people who read about apnea often wonder about sleep positions for snoring too, and the honest answer here is thinner. Snoring is vibration of soft tissue in the throat; it can happen with or without apnea, and what drives it differs from person to person.
Does sleep position stop snoring? The one controlled look I could verify at snoring rather than apnea studied 20 men who snored heavily. On a night when they slept on the side and used a decongestant nasal spray, the apnea-hypopnea index improved slightly, but snoring frequency did not improve. Weight loss later in the same study did move snoring, and I come back to that in the article on weight and snoring.
Braver HM, Block AJ, Perri MG. Treatment for snoring. Combined weight loss, sleeping on side, and nasal spray. Chest 1995. PMID 7750319
Twenty men is not the whole population of snorers. But it is a useful brake on the idea that the side is a guaranteed fix, and a guarantee is what most sleep position snoring advice quietly promises. The studies do not offer one; your own recordings can.
Adjustable bed for snoring, pillows and the things nobody tested
Plenty of people hope that an adjustable bed for snoring will do what position advice could not, and I looked for a randomized trial behind the idea. I did not find one I could verify. Raising the upper body is a reasonable thing to try, and plenty of people say it helps, but that is the level of evidence: reports, not measurements.
The same goes for wedge pillows, back-blocking shirts and the old tennis ball sewn into a pajama top. They exist because the logic above is sound, and a few have their own small studies, but I am not recommending or ranking any product here. If you want to try one, the fair way to judge it is the experiment in the next section, not a review site.
A fair experiment with your own nights
Here is where a recording earns its place. A microphone cannot see which way you are lying, so the position has to come from you. What it can do is put every snore, gasp and pause on a timeline with a sound you can play back.
- Pick a stretch of about a week and keep everything else as boring as possible: same bedtime, same evening, no alcohol, no new medication.
- Note how you fell asleep each night, and what you were lying on when you woke up if you remember. This is crude, because people turn over a dozen times a night.
- Record every night with the phone on the nightstand. In the morning, scroll through the timeline and listen to the loudest snoring episodes and any pauses followed by a gasp.
- Look for clusters. If the loud stretches sit early in the night, when you usually start on your back, and thin out later, that is a clue. If they are scattered across the whole night, position is probably not your main lever.
Treat it as a clue and nothing more. One bad night with a cold or a glass of wine will drown the signal, and a handful of nights is a handful of nights. On how to know if you snore I describe how to read the recordings, and the snoring app page shows how SleepTrace marks each episode with its loudness and a clip.
If the recordings show gasping or pauses, that is worth a doctor visit regardless of position. A recording can start that conversation; only a sleep study can settle it.
When position tricks are the wrong tool: loud snoring together with witnessed pauses in breathing, waking up gasping, strong daytime sleepiness, or morning headaches should be checked by a doctor before you build a routine around a pillow. SleepTrace is not a medical device and cannot diagnose sleep apnea. Position-dependent apnea is a diagnosis a sleep study can make, and a doctor decides what to do with it.
A worked example: reading one week of recordings
The numbers below are invented to show the reading, not measured. Say you record seven nights and note that you fell asleep on your back on five of them. On those five, the loudest snoring episodes and two gasp-like sounds sit in the first two hours, and the second half of the night is mostly quiet. On the other two nights, one with a cold and one after wine, the snoring runs loud from start to finish.
A fair reading is a lead, not a verdict: on ordinary nights the trouble seems to live where you start on your back, so position is worth testing properly, with a doctor if the gasps are real. The two noisy nights say that a blocked nose and alcohol can drown the position effect. What the example cannot tell you is whether you were really on your back when the sounds were made, or how many events there were per hour.
One gap remains on the research side. I found nothing that tracked a person's real home nights by position against a sleep study, which is the comparison that would settle whether a phone recording can say anything about this at all. My guess is that it can say something. It is a guess.
Frequently asked questions
For many people with obstructive sleep apnea it is the side, because lying on the back is when breathing events cluster in the position-dependent form. But that form is far from universal: it is common in mild apnea (49.5% in one chart review of 248 patients) and uncommon in severe apnea (6.5%). A sleep study that records body position is the only way to know which group you are in.
Not reliably. In a small study of 20 men who snore heavily, sleeping on the side combined with a decongestant nasal spray produced minor improvements in apnea-hypopnea index but no improvement in snoring frequency. Many people do snore less on the side, and your own recordings will show whether you are one of them, but the studies do not promise it.
I could not find a randomized trial of adjustable beds for snoring or sleep apnea that I could verify, so I do not make a claim either way. Raising the head is a plausible idea, and some people report it helps, but that is an impression, not a measured result.
A sleep study that records body position is how it is defined: the apnea-hypopnea index drops by more than half when you are not on your back. At home you can look for a pattern in recordings, such as snoring and gasping concentrated in the part of the night you spent on your back, but that is a clue for a doctor, not a diagnosis.
SleepTrace is a wellness app, not a medical device. This article is general information, not medical advice. If your symptoms are frequent, severe or worrying, please talk to a doctor.
Hear your own night. SleepTrace turns a night of audio into your sleep phases, the sounds you made, and how it all trends. No wearable, just your phone on the nightstand. Get it for iPhone or Android →