Alcohol and sleep: why a 'nightcap' is the wrong word

Of all the substances commonly mixed with sleep, alcohol is the one with the most consistent literature, and it's not flattering. The "nightcap" intuition gets the first hour right and everything after that wrong.

What happens in the first hour#

Alcohol is a CNS depressant. It reduces sleep latency (time to fall asleep) and increases slow-wave sleep in the first half of the night. This is the part most drinkers notice; you fall asleep faster, sometimes feeling pleasantly drowsy. The acute sedative effect is real.

What happens in the second half#

As alcohol is metabolised, the picture inverts:

• REM sleep is suppressed for the duration of the alcohol exposure, then rebounds in the second half of the night, causing vivid dreams and fragmented sleep.
• Arousals increase: measured awakenings and brief micro-arousals are markedly more frequent.
• Heart rate stays elevated through the night, with reduced HRV.
• Body temperature stays higher than baseline.
• Sleep architecture shifts toward lighter stages.

The net effect: shorter total sleep, more fragmentation, less REM, lower restorative quality. Multiple controlled studies (Ebrahim et al. 2013 meta-analysis; Roehrs and Roth review) replicate this pattern across populations.

The dose-response#

The effect size scales roughly with dose:

• Low (one drink, ≤0.2 g/kg): minimal effect on REM but small reductions in HRV and slight increases in arousals.
• Moderate (two drinks, ~0.5 g/kg): measurable REM suppression in first half, REM rebound in second half, elevated nocturnal heart rate.
• High (3+ drinks) — dramatic REM suppression, marked sleep fragmentation, the well-known 3 am wake-up.

There's also a sex difference: women metabolise alcohol more slowly than men on average, so the same number of drinks lands harder on sleep.

What this looks like on a wearable#

A consistent finding across consumer-wearable HRV reports is the "alcohol signature":

• Resting heart rate elevated by roughly 4 bpm after moderate drinking, up to around 9 bpm after heavy drinking
• HRV-based recovery score depressed by roughly 20–40 points at moderate-to-heavy intake
• Sleep score (whatever your device calls it) lower
• More wake events
• Less deep / REM in stage estimates (with all the usual caveats about wearable stage classification)

For people newly using HRV-tracking wearables, the alcohol signature is often the first reliably reproducible signal they spot in their own data. It's a useful demonstration that the device picks up something real.

The recovery timeline#

Alcohol's effects on sleep aren't gone the next morning. Studies measuring sleep on consecutive nights show:

• Night of drinking: marked disruption
• Night after: partial recovery; HRV often still below baseline
• 2+ nights after: usually fully recovered, depending on dose

This is why an honest read of HRV trends in people who drink twice a week or more shows depressed weekly average HRV even though no single night looks catastrophic.

What this isn't#

Not a verdict on whether to drink. Many people drink in moderation and prioritise the social or culinary value over sleep optimisation. That's a legitimate trade-off. What's not legitimate is believing that alcohol aids sleep; it sedates, which is different.

If sleep quality matters and drinking is part of life, two adjustments make a measurable difference:

• Stop drinking 3+ hours before bed so metabolism is mostly complete before sleep onset
• Cap the dose: the effects are dose-dependent, and the second half of the night is what suffers most

NHS guidance reference#

UK Chief Medical Officers' guidance is no more than 14 units per week, spread across 3+ days with several alcohol-free days. The guidance isn't sleep-specific; the underlying physiology is consistent with the sleep effects described here. Daily drinking, even at low doses, produces persistently lower sleep quality compared to abstinent control nights in trial data.

The honest summary: alcohol is, biologically, one of the more sleep-disruptive substances available over the counter. The first hour is misleading. If sleep is a priority, the cleanest intervention is fewer drinks, earlier in the evening, with more alcohol-free days per week.