Skip to content
Blog

Eye Mask and Earplugs in Pregnancy: A Trial

A 2026 trial gave 240 pregnant women who slept 7 hours or less an eye mask and earplugs, or nothing, for a week. Wrist-tracked sleep and glucose test results showed no statistically significant difference.

PulseOxTracker Team
Infographic summarising a 2026 trial. Night sleep on a wrist tracker after one week: 347 minutes with eye mask and earplugs against 349 minutes with nothing. No statistically significant difference in sleep or glucose. Who was studied: 240 pregnant women sleeping 7 hours or less, at one hospital in Kuala Lumpur, Malaysia, randomly given a mask and earplugs or nothing for the week before a routine glucose tolerance test. What they found, mask and earplugs against nothing. Fasting glucose: 4.4 vs 4.3 mmol/L, no clear difference. Glucose 2 hours after the sugar drink: 6.3 vs 6.4 mmol/L, no clear difference. Awake after falling asleep, wrist tracker: 46 vs 40 minutes, and sleep efficiency 88.2% vs 89.6%. Met gestational diabetes criteria: 19% vs 14% by Malaysian criteria, 10% vs 10% by WHO criteria, no clear difference. Self-rated sleep quality: rose in both groups, no clear difference. Self-reported night sleep: up 0.35 vs 0.16 hours, an extra analysis. What it cannot show: whether longer use would make a difference. Only one week, and how often the mask was worn is not reported. Wrist sleep data from 203 of 240 women, glucose from 233. No wrist tracking before the week, and naps not counted. Not blinded: everyone knew which group they were in.
A summary of the 2026 trial: a week of eye mask and earplugs in 240 pregnant women who slept 7 hours or less, against no aid.

Short sleep in pregnancy has been linked with gestational diabetes in earlier studies. Earlier trials from the same hospital found night sleep went up by about 23 to 25 minutes from where it started when women in late pregnancy wore an eye mask and earplugs at home.

This randomised trial asked whether a week of them would also lower glucose test results. It found no statistically significant difference in glucose or in wrist-tracked sleep.

Who was in the trial?

  • 240 pregnant women at a university hospital in Kuala Lumpur, Malaysia, recruited in 2021 and 2022.
  • All said they slept 7 hours or less a night. The median was about 6 hours.
  • They were aged 18 to 40, on average 24 to 25 weeks pregnant, and due a routine glucose tolerance test for gestational diabetes.
  • Women with a BMI over 35, night shift workers, smokers and those with sleep or other health conditions were left out.

Half, chosen at random, were given an eye mask and earplugs to wear in bed for the 7 nights before the test. The other half got nothing. Both groups wore a wrist activity tracker (an actigraph) at night and kept a sleep diary. The trial was funded by the hospital's own obstetrics department.

What did it find?

No statistically significant difference on any of the three main measures:

  • Night sleep on the tracker: 347 minutes with the mask and earplugs, 349 without.
  • Fasting glucose: 4.4 against 4.3 mmol/L.
  • Glucose 2 hours after the sugar drink: 6.3 against 6.4 mmol/L.

The other tracker measures showed no statistically significant difference either:

  • Awake after first falling asleep: 46 against 40 minutes.
  • Sleep efficiency: 88.2% against 89.6%.

How many women met the criteria for gestational diabetes showed no statistically significant difference, whichever of three sets of criteria was used. By the WHO criteria it was 10% in both groups.

Self-rated sleep quality rose in both groups between joining the trial and the test, by a similar amount.

Was there anything positive?

One thing, in an extra analysis the authors added afterwards. Self-reported night sleep rose by 0.35 hours with the mask and earplugs and by 0.16 hours without. That's a difference of about 12 minutes, and it was statistically significant.

The wrist tracker didn't show it, and extra analyses like this are weaker evidence than the main results.

No significant harms from the mask and earplugs were reported.

Why might it not have worked?

The authors suggest a few reasons:

  • Less sleep-short to begin with. The women reported a median of 5.8 hours at the start, against 4.6 and 5.0 in the earlier trials.
  • Earlier in pregnancy. Under 34 weeks here, 34 to 36 weeks before.
  • Adherence. More than 10% dropped out, mostly because they didn't wear the tracker. The authors presume some didn't wear the mask and earplugs either.
  • One week. Other sleep and lifestyle programmes that changed glucose ran for 1 to 8 weeks. Longer use might give a different result.

What can't it show?

  • A starting point on the tracker. Nobody wore it before the week, so the authors couldn't compare before and after.
  • Naps. Only night sleep was tracked. Some women may have made up for short nights in the day.
  • Blinding. Everyone knew which group they were in. The authors note the tracker and diary might have felt like an intervention of sorts to the comparison group too.
  • Other places. It was one hospital.

What can you do with this?

This trial reported no significant harms from an eye mask and earplugs. But a week of them showed no statistically significant difference in tracked sleep or glucose results in this group.

If you want to see whether they make a difference to you, track some nights with and without them. How many nights to record explains why one or two nights aren't enough.

For another pregnancy sleep trial, see insomnia therapy in pregnancy. For how close a wrist actigraph gets to a sleep study, see EEG headband vs actigraphy.

Sleeping badly or worried about your glucose in pregnancy? Talk to your midwife or doctor.

Not a medical device. Does not diagnose, treat or prevent any condition. Discuss results and changes with a clinician.

References

  1. Nurulhuda AS, Mohammad Som N, Hong J, Teo IH, Vimaladevi A, Gan F, Hamdan M, Tan PC. Eye-mask and earplugs as sleep aid on glucose metabolism in sleep-deprived pregnancy: a randomized controlled trial. BMC Pregnancy and Childbirth. 2026;26(1):1004. doi:10.1186/s12884-026-09283-y