Side-sleeping advice begins at 28 weeks, and before that, back-sleeping wasn’t linked to harm

Key takeaways

• In a study at 41 UK maternity units, going to sleep on the back in late pregnancy was tied to 2.3 times the odds of stillbirth, but it accounted for an estimated 3.7% of late stillbirths¹.
• An NIH-funded U.S. study found no link between sleep position and stillbirth, small babies or blood pressure disorders through 30 weeks; the NHS advises settling to sleep on either side from 28 weeks⁴⁵.
• First-time mothers with sleep-disordered breathing in mid-pregnancy had about twice the odds of preeclampsia in a large U.S. study⁷.
• In a trial of 208 pregnant women with insomnia, 4% of those who used a digital insomnia program had probable major depression three months after birth, compared with 18% who got standard care⁹.
• New and expecting parents can call or text the National Maternal Mental Health Hotline at 1-833-852-6262, free and open 24 hours a day¹⁴.

In 2011, New Zealand researchers reported a risk factor for late stillbirth that hadn’t been considered before: the position a woman went to sleep in². A study from Sydney found a similar link in 2015, and a larger New Zealand study in 2017 put the risk of going to sleep on the back at about four times higher².

Then came MiNESS. In November 2017, a team led by Alex Heazell of Tommy’s Stillbirth Research Centre in Manchester reported results from 41 UK maternity units¹². Women who went to sleep on their backs late in pregnancy had 2.3 times the odds of stillbirth, compared with those who went to sleep on their left side¹.

That same month, the UK charity Tommy’s launched a #SleepOnSide campaign². Two years later, a large U.S. study funded by the National Institutes of Health (NIH) found no link between sleep position and stillbirth before 30 weeks⁵.

Both findings can be true, and seeing why shows how pregnancy advice gets made. This article walks through that story, then covers the rest of pregnancy and postpartum sleep: snoring, insomnia, sleep aids, and how sleep and mood connect after birth.

The side-sleeping advice grew out of case-control studies

The UK study compared 291 women whose babies were stillborn at 28 weeks or later with 733 women whose pregnancies were ongoing¹. This is a case-control design: researchers start with an outcome and look back to see what differed. It suits rare outcomes like stillbirth.

The design has limits. It shows an association, not proof of cause, and it relies on what women recall about their sleep. The MiNESS authors wrote that more work is needed to show whether helping women avoid back-sleeping prevents stillbirths¹.

The study turned up other patterns too. Women who had a stillbirth were more likely to report sleeping under 5.5 hours the night before, getting up to use the toilet once or less, and napping every day¹. These are associations only, and no one has shown that changing them prevents loss.

A second analysis looked only at the women with healthy ongoing pregnancies from four case-control studies in New Zealand, Australia and the UK³. Just 57 of 1,760, or 3.2%, usually went to sleep on their backs³. Their babies weighed 144 grams less on average after adjustment, about 5 ounces by our math³.

One possible reason is blood flow. The NHS says the added risk after 28 weeks may have to do with the flow of blood and oxygen to the baby⁴. Our article on sleep comfort covers a pooled analysis of these studies.

What the numbers mean, and why they shouldn’t bring guilt

Odds of 2.3 times sound alarming. A second number puts them in context. MiNESS estimated that back-sleeping accounted for 3.7% of late stillbirths, with a wide margin of uncertainty, from 0.5% to 9.2%¹.

By our math, 3.7% is about 1 in 27. Put another way, about 26 in 27 late stillbirths weren’t attributed to sleep position.

That matters for anyone who has lost a baby. If you went to sleep on your back, these studies can’t tell you that it caused your loss, and most late stillbirths aren’t linked to sleep position. If you’ve lost a baby, you deserve support, not blame.

The U.S. study filled in the earlier months. Researchers led by Robert Silver of the University of Utah used data from nuMoM2b, a study of women in their first pregnancy⁵. Women reported their sleep position twice, once between weeks 6 and 13 and again between weeks 22 and 29⁵.

Back and right-side sleepers were no more likely to have a stillbirth, a small baby or a high blood pressure disorder⁵. More than 400 women also wore a device that recorded position. Spending over half the night on the back wasn’t linked to problems either⁵.

The study stopped at 30 weeks, so it couldn’t speak to late pregnancy⁵. The UK and New Zealand researchers pointed out that it measured position in mid-pregnancy, while their studies asked about going-to-sleep position after 28 weeks⁶. The U.S. team’s reply conceded the last trimester. But it said its findings refuted older advice to start side-sleeping after 20 weeks or after the first trimester⁶.

Here is the practical version from the NHS, England’s health service⁴:

• From 28 weeks, settle to sleep on your side. Left or right are both fine.
• If you wake on your back, don’t worry. Just roll back onto your side.
• If your baby is moving less than usual, or the pattern changes, call your midwife or maternity unit right away.

Snoring in pregnancy can signal a breathing problem tied to preeclampsia

Sleep-disordered breathing means the airway narrows or closes during sleep, causing pauses or shallow breaths. Obstructive sleep apnea is the best-known form.

The nuMoM2b team tested 3,705 first-time mothers at home with sleep monitors, once early in pregnancy and once in mid-pregnancy⁷. Women and their doctors generally weren’t told the results, so the study could see what happened naturally⁷. Sleep-disordered breathing showed up in 3.6% early on and 8.3% by mid-pregnancy⁷.

Women with it had about twice the odds of preeclampsia, a dangerous blood pressure disorder, at both time points⁷. Their odds of gestational diabetes were roughly three times as high⁷. The more breathing pauses per hour, the higher the odds of high blood pressure disorders and gestational diabetes⁷.

What does that mean in people? Across the whole study, 6.0% developed preeclampsia⁷. Doubling the odds on a baseline like that means roughly 11 in 100 instead of 6 in 100, by our rough math.

These are associations. The study didn’t test whether treating breathing problems in pregnancy prevents preeclampsia.

If you snore loudly, wake gasping or feel heavy daytime sleepiness, mention it at a prenatal visit, given these links⁷. Our sleep apnea article explains how testing and treatment work.

Insomnia in pregnancy responds to therapy, even through an app

Insomnia affects about one in seven pregnant women, according to researchers at the University of California, San Francisco (UCSF)⁸. Their earlier work linked an insomnia diagnosis to nearly double the odds of early preterm birth, an association that can’t prove cause⁸.

Pregnant women often prefer treatments that don’t involve drugs, the UCSF team noted⁸. So in a randomized trial, it tested digital cognitive behavioral therapy for insomnia (CBT-I). CBT-I is a structured program that retrains sleep habits and the worried thinking that keeps people awake. This version was an app-based program called Sleepio⁸.

The trial randomly assigned 208 pregnant women with insomnia, up to 28 weeks along, to six weekly sessions or standard care⁹. In the 2020 report in JAMA Psychiatry, insomnia severity fell more than twice as much with the app, and depression and anxiety symptoms improved more⁸.

The follow-up, published in 2022 in the journal Sleep, tracked women after birth⁹. Insomnia scores no longer differed between groups, but more app users were in remission from insomnia at six months⁹. The bigger surprise was mood. Three months after birth, 18% of the standard-care group had probable major depression, compared with 4% of app users⁹.

By our math, that’s 14 fewer cases per 100 women. Among the 143 women with few depressive symptoms at the start, the figures were 18% and 0%⁹. The authors called the results strong preliminary evidence, and it is still a single trial⁹. The pages we opened didn’t say who funded it or whether the app’s maker played a role.

If an urge to move your legs keeps you up, our restless legs article covers why sleep doctors now check iron first, including in pregnancy.

Melatonin and sleep aids: little data in pregnancy

Melatonin is a common question. MotherToBaby, an information service on medicines and other exposures in pregnancy and breastfeeding, updated its melatonin fact sheet in May 2026¹⁰. It says melatonin supplements in pregnancy haven’t been well studied, and it isn’t known whether they raise the chance of birth defects¹⁰.

The few human studies haven’t suggested a higher risk of birth defects, and studies of melatonin used for infertility haven’t shown more miscarriages¹⁰. One older study found that very high daily doses, 75 to 300 mg, lowered reproductive hormones in women¹⁰. MotherToBaby advises talking with your provider before taking melatonin or any supplement in pregnancy or while breastfeeding¹⁰.

For breastfeeding, the National Library of Medicine’s LactMed database notes that melatonin is a normal part of breast milk, with higher levels at night¹¹. It says brief use of a typical evening dose is unlikely to harm a breastfed baby. Some experts advise against it because data are limited and preterm newborns clear it slowly¹¹.

You can call MotherToBaby at 866-626-6847 or text 855-999-3525 with questions¹⁰. Our melatonin article covers how it works as a body-clock signal.

After birth, protecting sleep and mood go together

Perinatal depression, which starts in pregnancy or the year after birth, strikes up to 1 in 7 women, the U.S. Preventive Services Task Force (USPSTF) says¹². In 2019 it gave a “B” grade to counseling for pregnant and postpartum people at higher risk¹². Higher risk includes a history of depression, current mild symptoms, low income, being a teen or single parent, recent partner violence, high anxiety, or hard life events¹².

Counseling such as CBT or interpersonal therapy lowered the chance of perinatal depression by 39% across 20 trials with 4,107 people¹². In absolute terms, if 19 in 100 would otherwise develop depression, about 1 case is prevented for every 13 or 14 people counseled¹². An April 2025 draft update, still not final in October 2026, keeps the “B” grade and spells out that it covers teens as well as adults¹².

Programs to help babies sleep longer showed mixed results for parents’ mood. One of three trials found a 39% lower likelihood of high depression scores at six months, and the 2025 draft says more research is needed¹².

Our suggestion, not something these trials tested: plan the nights before the baby arrives. One option is to trade shifts so each adult gets one longer stretch of sleep, with one feed covered by pumped milk or formula. Every feeding choice is valid, and the right plan is the one your household can sustain.

Medicines for postpartum depression. On August 4, 2023, the Food and Drug Administration (FDA) approved zuranolone (Zurzuvae), the first pill for postpartum depression¹³. It’s taken in the evening for 14 days¹³. In two placebo-controlled trials, symptoms improved more by day 15, and the benefit held through day 42¹³.

The drug’s developers, Sage Therapeutics and Biogen, announced results from one of those trials themselves and filed jointly for approval, which the FDA granted to Sage¹³¹⁵. The pages we checked didn’t name who paid for the trials.

The drug carries a boxed warning: no driving or running heavy machinery until at least 12 hours after a dose, since people may not realize how impaired they are¹³. It may harm a fetus, so the label calls for effective birth control during treatment and for one week after¹³. The label also warns that it may cause suicidal thoughts and behavior¹³.

Where to get help. Call or text the National Maternal Mental Health Hotline at 1-833-852-6262 (1-833-TLC-MAMA). It’s free, confidential and available 24/7 in English and Spanish¹⁴. If you’re thinking about harming yourself, call or text 988. Our article on major life transitions covers the warning signs of postpartum depression.

Where your baby sleeps is its own safety question. Our article on sharing a bed covers the American Academy of Pediatrics’ safe-sleep rules for babies, and our article on how much sleep children need covers the years ahead.

The pregnancy and postpartum sleep guide

Rest is something natural-health readers already prize. Most of the steps below are drug-free and cost little, and the insomnia therapy with trial support in pregnancy involves no pills⁸. None of it should come with guilt.

Stage	Common problem	What helps	When to call	Sources
Before 28 weeks	Worry about sleep position	Sleep in whatever position is comfortable; no link to harm was found through 30 weeks	Once you feel movements, baby moving less than usual: contact your midwife or unit	45
From 28 weeks	Going to sleep on your back	Settle on either side; turn over if you wake on your back	Baby moving less, or a changed pattern: contact your midwife or unit right away	4
Any trimester	Loud snoring or gasping	Mention it at a prenatal visit	At your next visit, given the links to preeclampsia	7
Any trimester	Insomnia most nights	Ask about CBT-I, including digital versions	Sleeplessness with low mood or worry: call the hotline	8914
Pregnancy or breastfeeding	Thinking about melatonin or a sleep aid	Check with your provider or MotherToBaby first	Before starting any product	1011
After birth	Broken sleep and low mood	Plan night shifts; counseling if you’re at higher risk	Low mood or worry that lingers: call the hotline	1214
After birth, if prescribed	Zuranolone	Taken in the evening for 14 days, per the label; no driving for at least 12 hours after a dose	Any thoughts of suicide, a labeled risk: call or text 988	13

How strong is the evidence?

Claim	Rating	Why
Going to sleep on your side from 28 weeks lowers stillbirth risk	Promising but early	Consistent case-control studies; no trial123
You must avoid back-sleeping before 28 weeks	Evidence says no, or unsafe	No link to stillbirth, small babies or blood pressure disorders through 30 weeks in a large U.S. study5
Treating sleep-disordered breathing prevents preeclampsia	Not enough evidence	A link exists, but treatment wasn’t tested7
Digital CBT-I eases insomnia in pregnancy	Promising but early	One randomized trial89
Insomnia therapy in pregnancy prevents postpartum depression	Promising but early	18% vs. 4% in one trial9
Counseling prevents perinatal depression in people at higher risk	Strong evidence	20 trials; USPSTF “B” grade12
Melatonin is safe in pregnancy	Not enough evidence	Few human studies10
Zuranolone relieves postpartum depression in the short term	Strong evidence	Two placebo-controlled trials from the drug’s developers; follow-up only to day 421315

Questions readers ask

The bottom line

The side-sleeping advice is a good example of careful science under uncertainty. Several case-control studies point the same way for late pregnancy, a large U.S. study found no added risk in earlier months, and the practical answer is simple: from 28 weeks, settle on your side, and don’t panic if you wake on your back.

Beyond position, the pregnancy sleep problems worth raising are loud snoring and stubborn insomnia, which responds to therapy without pills. After birth, sleep and mood travel together. Counseling has strong evidence for people at higher risk, and help is a phone call or text away.

Keep exploring: exercise during pregnancy.

Related reading

• A weighted blanket eased insomnia in one trial, but babies should never sleep under one
• Couples move more when they share a bed, yet most say they sleep better
• Sleep apnea care now goes well beyond CPAP, and mouth tape isn’t part of it
• Doctors’ first choice for insomnia is retraining your sleep with CBT-I
• Exercise in pregnancy is safe for most women, and most don’t get enough

Sources

1. Heazell AEP, Li M, Budd J, et al. “Association Between Maternal Sleep Practices and Late Stillbirth: Findings From a Stillbirth Case-Control Study.” BJOG, 2018;125(2):254–262 (published online November 2017; White Rose Research Online record). eprints.whiterose.ac.uk
2. Tommy’s. “MiNESS: Is There a Link Between Stillbirth and a Woman’s Position Going to Sleep?” Research page, accessed October 2026. tommys.org
3. Anderson NH, Gordon A, Li M, et al. “Association of Supine Going-to-Sleep Position in Late Pregnancy With Reduced Birth Weight: A Secondary Analysis of an Individual Participant Data Meta-analysis.” JAMA Network Open, 2019;2(10):e1912614 (University of Auckland record). researchspace.auckland.ac.nz
4. NHS. “What You Can Do to Reduce the Risk of Stillbirth.” Last reviewed April 24, 2024. nhs.uk
5. Eunice Kennedy Shriver National Institute of Child Health and Human Development. “Sleeping Position During Early and Mid Pregnancy Does Not Affect Risk of Complications, NIH-Funded Study Suggests” (on Silver RM, et al., Obstetrics & Gynecology, 2019). Science Update, September 20, 2019. nichd.nih.gov
6. McCowan LME, Cronin RS, Gordon A, O’Brien L, Heazell AEP. Letter on “Sleep Position Through 30 Weeks,” with reply by the nuMoM2b investigators. Obstetrics & Gynecology, 2020;135(1):218 (copy posted by Star Legacy Foundation). starlegacyfoundation.org
7. Facco FL, Parker CB, Reddy UM, et al. “Association Between Sleep-Disordered Breathing and Hypertensive Disorders of Pregnancy and Gestational Diabetes Mellitus.” Obstetrics & Gynecology, 2017;129(1):31–41 (RTI International record). rti.org
8. UCSF Osher Center for Integrative Health. “Digital Treatment for Prenatal Insomnia Found to Be Twice as Effective as Standard Care” (on Felder JN, Epel ES, Neuhaus J, Krystal AD, Prather AA, JAMA Psychiatry, 2020). January 22, 2020. osher.ucsf.edu
9. Felder JN, Epel ES, Neuhaus J, Krystal AD, Prather AA. “Randomized Controlled Trial of Digital Cognitive Behavior Therapy for Prenatal Insomnia Symptoms: Effects on Postpartum Insomnia and Mental Health.” Sleep, 2022;45(2):zsab280 (Duke Scholars record). scholars.duke.edu
10. MotherToBaby. “Melatonin.” Fact sheet, May 1, 2026. mothertobaby.org
11. Drugs and Lactation Database (LactMed), National Library of Medicine. “Melatonin.” Revised April 15, 2026. ncbi.nlm.nih.gov
12. U.S. Preventive Services Task Force. “Perinatal Depression: Preventive Interventions.” Final recommendation, February 12, 2019; and draft recommendation, April 22, 2025. uspreventiveservicestaskforce.org and uspreventiveservicestaskforce.org
13. U.S. Food and Drug Administration. “FDA Approves First Oral Treatment for Postpartum Depression.” News release, August 4, 2023. fda.gov
14. Health Resources and Services Administration, Maternal and Child Health Bureau. “National Maternal Mental Health Hotline.” Accessed October 2026. mchb.hrsa.gov
15. Sage Therapeutics and Biogen. “Sage Therapeutics and Biogen Announce that the Phase 3 SKYLARK Study of Zuranolone in Postpartum Depression Met its Primary and All Key Secondary Endpoints.” Company press release, Business Wire, 2022. businesswire.com