Elmsword
← Sleep

Sleep · Habits · 4 min read

A nightcap and your sleep: what alcohol does to the night

A 2025 meta-analysis of 27 studies found that even a low dose of alcohol, about 0.35 to 0.50 g per kg of body weight, can delay and shorten REM sleep.

A drink before bed to help you drift off is a common habit. A 2025 systematic review and meta-analysis in Sleep Medicine Reviews looked at what that drink actually does to the night. The authors included 27 studies in healthy adults and looked at how the amount and timing of alcohol changed sleep.

What the studies found

The clearest change was in REM sleep, the stage often called dreaming sleep. After alcohol, REM started later in the night and there was less of it overall.

It didn't take much. REM started later at about 0.35 g of alcohol per kg of body weight, roughly two standard drinks, and there was less REM at about 0.50 g/kg. The paper calls 0.50 g/kg about two drinks in its summary and about three in its dose figure. The more people drank, the worse it got.

The "helps me fall asleep" part only showed up at a high dose. At about 0.85 g/kg, roughly five standard drinks, people fell asleep a bit faster. Reaching deep sleep sooner took about 0.95 g/kg, roughly six drinks. The authors write that drinking this much likely makes the later REM disruption worse, and in the full paper they say using alcohol as a sleep aid is not an appropriate strategy.

For total sleep time, sleep efficiency and time awake during the night, they couldn't say either way. The results were too uncertain.

What you can try

  • If you drink in the evening, try a few nights with none and a few with your usual amount, and compare how you feel in the morning. A simple sleep diary helps.
  • Don't count on a nightcap to get you to sleep. In this review, faster sleep onset only appeared at around five drinks, and that came with more REM disruption.
  • If you're having a drink anyway, keep it small. Effects on REM grew with the amount.
  • If you lie awake, the "get up for a while" method is an option that doesn't involve alcohol.

What this review can't tell you

Everyone in these studies was a healthy adult aged 18 to 70. The authors say the results may not apply to teenagers, adults over 70, or people with alcohol use disorder or sleep disorders such as insomnia.

Timing is still an open question. In most studies people drank within about 3 hours of bedtime, so the authors couldn't work out whether drinking earlier in the evening makes a difference.

They also couldn't tell whether men and women respond differently, because few studies reported results by sex. The drink counts assume a 62 kg adult and a 10 g standard drink. A standard drink in the US holds about 14 g of alcohol and a UK unit is 8 g, so treat the drink numbers as rough guides. The authors also note that 21 of the 27 studies had some risk of bias.

References

  1. SMR 2025Gardiner C, Weakley J, Burke LM, et al. The effect of alcohol on subsequent sleep in healthy adults: a systematic review and meta-analysis. Sleep Med Rev. 2025;80:102030. doi:10.1016/j.smrv.2024.102030
    doi.org/10.1016/j.smrv.2024.102030

This guide is for education only and isn't medical advice. Results vary. If sleep problems last, or come with other symptoms, talk to a doctor or a trained CBT-I provider.