Which Sleep Position, For Which Problem

By Dovy Paukstys, Founder, Komori Care

This article is educational and is not medical advice. It does not diagnose, treat, prevent, or cure any condition. Komori is not a medical device, is not FDA-cleared, does not detect seizures, and is not intended for people with epilepsy. Talk to a clinician about anything on this page that applies to you.

Position advice arrives one problem at a time. A gastroenterologist says left side. A physical therapist says knees supported. Someone on the internet says never your stomach. Nobody puts it in one place, and the advice quietly conflicts — the side that helps your reflux is the side your shoulder hates.

So here is the whole thing in one table, with the evidence behind each row and an honest label on how strong that evidence is. Some of these are well established. One of them is actively contested and gets sold to you as settled.

The table

Now the evidence, row by row, because "best" and "worst" are doing a lot of work up there.

Acid reflux: the strongest row on the table

This is the one where the effect is large and the mechanism is obvious.

A study in the American Journal of Gastroenterology measured esophageal acid exposure in GERD patients sleeping on their right side versus their left. Left-side sleeping came out at roughly a 76% reduction in acid exposure compared with the right.

That is not a nudge. That is a bigger swing than most interventions you can make without a prescription, from changing which way you roll.

The anatomy is the reason: the stomach sits to the left, and on your left side the acid pool sits away from the valve at the top of the stomach. On your right side, it sits against it. Full detail in acid reflux and sleep position.

Strength: strong. Measured physiological outcome, clear mechanism, consistent direction.

Snoring and airway: strong, and more specific than "sleep on your side"

On your back, gravity pulls the tongue and soft palate toward the airway. Narrower passage, faster airflow, more tissue vibration. On your side, the same tissues fall sideways instead. It is unglamorous physics and it holds up.

The number worth knowing is that an estimated 50–60% of people with obstructive sleep apnea have positional OSA — apnea events at least twice as frequent on their back as on their side. That is a majority, and it is why positional therapy is a real clinical category rather than folk advice.

One finding is worth more than it usually gets credit for: a study in the Journal of Clinical Sleep Medicine compared positional therapy against CPAP in positional OSA patients. CPAP reduced AHI slightly more — but positional therapy patients had significantly better compliance. The treatment people actually use beats the one they abandon in a drawer. More in snoring and sleep position and sleep apnea and position data.

Strength: strong for snoring, strong for the positional-OSA subset. Note the subset part. Apnea is a medical condition and needs a clinician, not a table.

Low-back pain, shoulders and hips: real, but individual

Low back pain is the most common pain complaint in sleep — NINDS estimates around 80% of adults experience it at some point, and position is one of the modifiable factors.

The useful insight here is not which position wins. It is that duration matters as much as position. Thirty minutes on your back is unremarkable. Five hours on your back with spinal stenosis is a different morning entirely. A table can tell you a position is risky for you; it cannot tell you that you spent a third of the night in it.

And the shoulder/hip row runs directly against the reflux row. Side sleeping parks hours of body weight on one shoulder and one hip. If the shoulder you need to lie on is the shoulder that hurts, the "best" position for your reflux is the one that wakes you at 4 a.m. There is no resolution to that in a chart — it is a trade you make. Chronic pain and sleep position goes further.

Strength: moderate. The mechanism is sound, the personalization is the whole game, and most of the evidence is about populations rather than about you.

Neck: it is a pillow problem, not a position problem

The neck row is really a geometry row. The right pillow height changes with how you sleep, and getting it wrong is measurable:

• Going from a 110 mm pillow to a 170 mm one raised average cranial pressure by about 30% and increased cervical angle by 66% (PeerJ, 2016).
• In 84 people, adjusting pillow height dropped neck pain from 6.8 to 4.1 out of 10 over three months (Journal of Physical Therapy Science, 2023).
• A shoulder-scaled pillow-height metric predicted cervical curvature well — in only 9 subjects (Medical & Biological Engineering & Computing, 2025).

I include that last sample size on purpose. Nine people is a pilot, not a finding, and you should discount it accordingly. The full breakdown is in pillow height and neck pain.

Stomach sleeping is the exception that is not about pillows at all: to breathe, your head is rotated roughly ninety degrees for hours. No pillow fixes a rotation. How to stop sleeping on your stomach is the practical page.

Strength: moderate, with one weak study I have flagged rather than hidden.

Late pregnancy: the row that gets oversold

Every other page will tell you flatly to sleep on your left in the third trimester, and never on your back. The evidence is messier than that, and pregnancy is the worst possible topic to be falsely confident about.

What supports the advice: the Midlands and North of England Stillbirth Study, a case-control study of over 1,000 women in the British Journal of Obstetrics and Gynaecology, and a study in The Lancet which found going to sleep supine was associated with a 3.7-fold increased risk of late stillbirth. The mechanism is real too — in late pregnancy the uterus can compress the inferior vena cava, and the left side keeps it open.

What complicates it: a prospective cohort of over 8,700 women found no significant association between self-reported sleep position and stillbirth, small-for-gestational-age babies, or gestational hypertension.

Both of those are in the literature. The case-control studies asked women to recall their position after an outcome had already occurred; the prospective study followed women forward. Those designs have different failure modes, and reasonable people read the balance differently.

Note also what all of them measured: going-to-sleep position, self-reported. Not the position anyone actually held through the night.

The honest version is that the advice is low-cost and plausible, the risk if real is small in absolute terms, and the certainty in most articles is manufactured. Take it to your OB, not to a table on the internet. Sleep position in pregnancy walks through all of it.

Strength: contested. Anyone presenting this as settled has not read the second half of the literature.

Setting up for each position

Practical, and none of it is a treatment:

Left or right side. A pillow lofted enough to fill the gap between your ear and your shoulder, so your neck stays level with your spine. A pillow between the knees so the top hip does not drag your lower back into a twist. Skip the side your painful shoulder or hip is on, whatever the reflux row says.

Back. Thinner pillow than you think — enough to support the neck curve, not to push your chin toward your chest. A pillow under the knees takes the arch out of your lower back. Skip if you snore, if you have positional apnea, or in late pregnancy if your clinician has advised against it.

Stomach. The thinnest pillow you can tolerate, or none. This is damage limitation rather than setup, because nothing addresses the rotated neck.

The part the table cannot do

Every row above assumes you can choose. You choose for about a minute.

The rest of the night is a thing that happens to you. You start on your left because your reflux is bad, and you wake up on your back with no idea how long you were there. The advice was correct. Whether you followed it is unknown.

That is the actual gap, and it is why the strongest row on this table — reflux, with a 76% acid-exposure difference — is also the one where compliance is impossible to verify. A gastroenterologist can tell you which side. Nobody can tell you which side you were on at 3 a.m.

I build a contactless position monitor, so I am not a neutral party on that point and you should weigh it accordingly. Komori is designed to log position — supine, prone, left, right — and how long each lasted, through blankets, without a camera or a wearable. It is a wellness product. It does not diagnose anything, it does not treat anything, and it will not tell you which row of this table applies to you. That is a conversation with a clinician.

What it is for is the narrower question underneath all of this: the advice was good, but did the night match it?

Sources

• Esophageal acid exposure by sleeping side — American Journal of Gastroenterology, discussed in acid reflux and sleep position
• Positional OSA prevalence and positional therapy vs CPAP compliance — Journal of Clinical Sleep Medicine, discussed in sleep apnea and position data
• Low back pain prevalence — NINDS, discussed in chronic pain and sleep position
• Pillow height and cranial pressure — PeerJ, 2016
• Pillow height adjustment and neck pain, n=84 — Journal of Physical Therapy Science, 2023
• Shoulder-scaled pillow height, n=9 — Medical & Biological Engineering & Computing, 2025
• Going-to-sleep position and late stillbirth — The Lancet; and the Midlands and North of England Stillbirth Study, BJOG
• Prospective cohort, n=8,700+, null result — discussed in sleep position in pregnancy

Keep reading

• Acid reflux and sleep position — the strongest evidence on this page, in full.
• Sleep position in pregnancy — both halves of the contested literature.
• Every couples sleeping position study is a survey — the same measurement gap, for two people.
• How to stop sleeping on your stomach — if the table says your habit is the problem.