steady

Sleep · In the night

Awake in bed? Get up

20 minEvidence: decent

Do this

Lying awake for what feels like twenty minutes: get up, sit somewhere dim, do something dull, go back when sleepy.

How to do it

  1. Don't check the clock. Estimate. Clock-watching raised worry and time to fall asleep.
  2. Get up and go somewhere dim. Bright light at night pushes your body clock later.
  3. Do something dull until you're sleepy, then go back. Repeat if you need to.
  4. Get up at your usual time, however the night went. A fixed wake-up time is part of the same rules.

How bed becomes the problemTap a word for what it means

  1. Bedbecomes a cue

    The idea is that hours of lying awake teach your brain that bed is for being awake. Widely accepted, not proven directly.

  2. The clockfeeds the worry

    People told to watch a clock while trying to sleep worried more and took longer to fall asleep.

  3. Body clockstays put in dim light

    Bright light at night pushes your body clock later. Dim light keeps tomorrow on schedule.

  4. Youback to bed when sleepy

    Bed gets paired with falling asleep again. In trials of these rules, gains were modest and the evidence low in quality.

Why it works

This is called stimulus control, and it's a standard part of the insomnia treatment with the best evidence. Evidence: solid On its own, its evidence is thinner: helpful, with modest gains, from small trials. Evidence: decent

The idea behind it: your brain learns by association. Hours of lying awake teach it that bed is a place for being awake. Getting up breaks the pairing. That theory is widely accepted but hasn't been proven directly. Evidence: decent

The long readEvery step of the chain, each study with its numbers, and what's still unproven. 3 min read.

What happens in your brain

Bed becomes a cue. The theory is that your brain links places with what usually happens there. Lie awake in bed often enough and bed starts to cue wakefulness instead of sleep. Evidence: decent That's the reasoning behind the rules; it hasn't been shown directly. Evidence: worth trying

The clock feeds the worry. In two experiments, people told to watch a clock while trying to sleep worried more and said they took longer to fall asleep than people who didn't, or who watched a meaningless number display. Evidence: decent Don't watch the clock. Estimate. Clock-checking itself keeps you up. Evidence: decent

Dim light keeps the clock steady. Bright light before the overnight low point in body temperature pushes the body clock later. Evidence: solid So if you get up in the night, keep the lights low. Evidence: worth trying

Back to bed when sleepy. Going back only when you're sleepy means bed gets paired with falling asleep again, not with lying there. Evidence: worth trying

The proof

  • The 2021 US sleep medicine guideline found 8 trials of stimulus control on its own in adults with long-running insomnia. It rated the evidence low in quality and gave a conditional recommendation for using it. Evidence: solid
  • In the pooled trials, people fell asleep about 10 minutes sooner and spent about 19 minutes less awake in the night than the comparison groups. Neither passed the guideline's bar for a clearly meaningful difference, and both could have been chance. Evidence: decent
  • In the one trial where remission could be compared, the insomnia remission rate was 19% higher with stimulus control than in the comparison group, and that difference passed the guideline's bar for a meaningful one. One trial, so read it with care. Evidence: decent
  • The same guideline gives its only strong recommendation to the full package, CBT-I (cognitive behavioural therapy for insomnia), which includes these rules. Evidence: solid
  • In the second experiment, 38 people with insomnia watched either a clock or a meaningless number display while trying to sleep. The clock-watchers reported more worry and a longer time to fall asleep than on their normal night; the display-watchers worried less. Evidence: decent

How to do it

  • Don't watch the clock. Estimate. Evidence: decent
  • After what feels like twenty minutes awake, get up. The guideline says to get up when you can't sleep; twenty minutes is a common rule of thumb, not a tested number. Evidence: decent
  • Sit somewhere dim and do something dull, and go back to bed when you're sleepy. Repeat if you need to. Evidence: decent
  • Get up at your normal time, however the night went. A fixed wake-up time is part of the same set of rules. Evidence: decent

Watch out

On their own, these rules have modest, low-quality evidence. The strong evidence is for the full treatment they belong to. Evidence: decent If you've slept badly three or more nights a week for three months or more, ask a doctor for CBT-I. It's the recommended first treatment for long-running insomnia. Evidence: solid During the 3am moment, the steps are laid out one at a time.

Arrows to move, Enter to open, Esc to close.