Does Alcohol Affect Sleep? What Two Reviews Found, and How to Check Your Own Nights

5 min read · Updated 2026-09-26

Yes, though not the way most people expect. A 2025 meta-analysis of 27 studies found that even about two standard drinks delays and cuts REM sleep, and more makes it worse. Falling asleep faster only showed up at around five drinks. What it couldn't settle is whether you sleep fewer hours.

The nightcap has a reputation for a reason. A 2013 review of every study its authors could find on alcohol and sleep in healthy people concluded that at all doses, alcohol shortens the time it takes to fall asleep and makes the first half of the night more solid. So the part you notice, dropping off quickly, is real enough.

The part you don't notice is the second half of the night. The same review found more disrupted sleep after the halfway point, and a delay in the first REM period at every dose, which it called the most recognizable effect alcohol has on REM sleep.

What a 2025 meta-analysis added

Gardiner and colleagues pooled 27 studies in 2025 and looked at dose. REM sleep was delayed and shortened, and the damage started at their low dose, up to about two standard drinks, then got worse as the dose went up.

Their numbers disagree with the older review on one point. Falling asleep faster only showed up at the high dose, around five standard drinks or more. So at the amount most people mean by a nightcap, the benefit is the part with the weaker evidence, and the REM cost is the part with the stronger evidence.

What neither one can tell you

The meta-analysis couldn't pin down what alcohol does to total sleep time, sleep efficiency, or time awake during the night. The uncertainty there was large. That matters if you're checking this on yourself, because hours slept is the number most people track, and it's the one the research is least sure moves.

It also can't tell you about you. Both reviews are about healthy adults on average. Your body, your dose, and how late you drink are the variables they point to and can't answer for you.

How to check it on your own nights

Treat it like the studies do, just smaller. Log a habit for drinking, log your sleep every morning, and compare the nights after drinking against the nights after not. The method for telling a pattern from a coincidence applies here the same way it does to anything else.

In HabitSync, a drinking habit gets tested against your sleep hours the same day and the next, and against your mood, and nothing surfaces until there are at least five days on each side and the gap survives a significance test. It also records how you woke up each morning, but it doesn't test that one yet, so given what the research says about hours, keep an eye on that trend yourself.

One thing to watch for: if you mostly drink on Fridays and Saturdays, a gap could be the weekend and not the drinks. Later bedtimes, a different morning, a different week. A few weeknight drinks, or a few dry weekends, is what separates the two.

And if you're drinking to get to sleep most nights, that's worth raising with a doctor rather than working out alone.

Does one drink affect sleep?

The 2025 meta-analysis found REM disruption starting in its low-dose group, which went up to about two standard drinks. It didn't split out a single drink, so the honest answer for one is that the research doesn't say.

Does alcohol help you fall asleep?

Sometimes. The 2013 review found faster sleep onset at every dose. The 2025 meta-analysis only found it at around five drinks or more, which is also where the REM cost is largest.

Keep reading

If you want to see this in the app, here is how HabitSync tracks sleep.