Can't sleep? What to do in the middle of the night

Awake in the dark and unable to drift off again? The research points away from three common reactions: checking the clock, trying harder to sleep, and adding up the hours you have left. What it supports instead is lowering effort and arousal. For repeated nights, a full CBT-I course has far more evidence than any single trick. ^[1][2][3]

What the research says

The research on those 3am reactions is small but consistent. In one experiment, 30 good and 30 poor sleepers were told either to watch a clock or not while trying to fall asleep; the clock watchers worried more beforehand and took longer to drop off. A second experiment in 38 people with diagnosed insomnia compared a clock with a neutral digital display: the clock was rated more worrying and more sleep-interfering, and clock watchers overestimated how long they took to fall asleep. ^[1] These clock results fit a wider model of insomnia in which worry about sleep raises arousal, pulls attention towards signs of trouble, and leads people to overestimate the deficit. ^[2] For treatment, the American Academy of Sleep Medicine suggests stimulus control (keeping the bed linked with sleep) and relaxation as stand-alone options, and advises against relying on sleep hygiene alone. ^[3] In plain terms, the clock and the effort are part of the problem, and a full CBT-I course is the answer with evidence behind it.

The studies behind this

• Clock monitoring (Tang, 2007): in one experiment, 30 good and 30 poor sleepers were told either to watch a clock or not while trying to fall asleep. Clock watchers reported more worry before sleep and took longer to fall asleep. ^[1]
• A second experiment with 38 people diagnosed with primary insomnia compared watching a clock with watching a neutral digital display. Clock watching was rated more worry-provoking and sleep-interfering, and clock watchers overestimated how long they took to fall asleep. ^[1]
• Cognitive model (Harvey, 2002): worry about sleep and its consequences raises arousal, directs attention to sleep-related threats, and can lead people to overestimate the deficit. ^[2]
• Treatment guidelines (AASM, 2021): suggest stimulus control (a CBT-I component) and relaxation therapy as single-component treatments, and advise against relying on sleep hygiene alone. ^[3]

Why does waking at night feel worse than it is?

The cognitive model proposes that worries about sleep and its effects are themselves part of what keeps you awake, and that attention to the clock and to bodily signs feeds the loop. ^[2] The clock experiments are consistent with that, though they are short and tested one night. ^[1]

What does the research point to?

These are the directions the studies above point in. They are general, and a CBT-I course tailors them to your own sleep records.

• Not watching the clock. In the experiments, people told not to monitor a clock worried less and fell asleep sooner than people told to watch one. The evidence is from two small experiments, but it is direct. ^[1] A sleep tracker score can play the same role as the clock.
• Less effort, not more. Trying hard to sleep, and monitoring for signs of it, are exactly what the cognitive model identifies as fuel for insomnia. ^[2]
• Keeping the bed linked with sleep. Stimulus control, which the AASM suggests as a stand-alone treatment (a conditional recommendation), includes going to bed only when sleepy and getting out of bed when unable to sleep, so that the bed stays associated with sleeping rather than with lying awake. ^[3]
• Lowering arousal. Relaxation therapy is also suggested by the AASM as a single-component option. ^[3]

How long to stay in bed, when to get up, and how to set your own schedule are matters for a full CBT-I plan. We do not publish those rules here, because the right settings depend on your sleep records.

Is it normal?

Brief awakenings occur in healthy sleep, and time awake after falling asleep tends to rise with age in healthy adults. ^[5] There is more on this in Is it normal to wake up at night? An occasional bad night is not a disorder. Repeated trouble, with distress or daytime impact, is worth discussing with a clinician, and CBT-I is the recommended first treatment for chronic insomnia. ^[3][6]

When should you see a doctor?

• Nights like this happen repeatedly over months, or are hurting your mood, work or safety.
• You snore loudly, gasp or stop breathing in your sleep, or have other symptoms suggesting another sleep disorder. The European guideline names breathing-related and limb-movement disorders as ones to evaluate if suspected. ^[6]
• You are leaning on alcohol or medication to get through nights. Speak to a doctor before changing any medicine.

Where the evidence is thin

• The clock experiments tested single nights in small samples. ^[1]
• No trial we found tests a specific 3am routine against another. The supported approach is the whole CBT-I package, reviewed across 37 studies in an American Academy of Sleep Medicine task force update. ^[4]

If this is a regular pattern, Small Hours walks you through CBT-I based on your own sleep diary. You can sign up at trysmallhours.com.

General education, not medical advice.