Sleep restriction making you more tired? Why it happens and how long it lasts

Yes, it can. In the one laboratory study that tracked this closely, daytime sleepiness was higher in the first three weeks of sleep restriction and back at baseline by three months. A second small study found no significant change in sleepiness. ^[1][2] The guideline view is that the early effects are usually mild and fade. ^[3]

Sleep restriction is the part of CBT-I that aims to bring the time you spend in bed closer to the sleep you are actually getting. This article covers what studies found about how it feels. It does not explain how it is set up, because that depends on your own sleep records and your own clinician.

What the research says

In the one laboratory study that tracked this closely, 16 adults were followed for four weeks and again at three months: recorded sleep fell by over an hour, reaction-time lapses rose, and sleepiness was higher in weeks one to three but back at baseline by three months, while diary sleep and insomnia severity improved. ^[1] A second two-week study of 16 adults found no significant change in sleepiness, reaction time or simulated driving, though sleep duration stayed about the same and there was no control group. ^[2] The AASM guideline calls the early downsides, sleepiness and trouble concentrating, typically mild and expects them to fade as sleep improves. ^[3] The treatment itself has support: four adequately designed trials showed moderate-to-large gains in diary measures of falling asleep and night waking and a small gain in total sleep, ^[4] and the HABIT trial of 642 primary-care adults found insomnia severity 3.05 points lower at six months with nurse-delivered sleep restriction than with a leaflet alone, with no serious adverse events judged related to it. ^[6]

The studies behind this

• Kyle laboratory study (2014): 16 adults with insomnia, tracked for four weeks and again at three months. Recorded sleep time fell by an average of about 70 to 90 minutes at each of three lab checks. Reaction-time lapses rose. Sleepiness scores were higher at weeks 1, 2 and 3 and had returned to baseline by three months. Diary-reported sleep and overall insomnia severity improved over the same period. ^[1]
• Whittall study (2018): 16 adults over two weeks. Diary-reported sleep time stayed similar, and there was no significant change in sleepiness, reaction time or driving-simulator performance. The authors say this holds only when sleep duration stays about the same, and call for studies with a control group. ^[2]
• AASM guideline (2021): early downsides can include daytime sleepiness and trouble concentrating. These typically fade as treatment progresses and sleep improves. The guideline also says the early effects are typically mild and resolve quickly for most patients. ^[3]
• Evidence for the treatment itself: a review of stand-alone sleep restriction found four adequately designed trials, with moderate-to-large improvements in diary measures of falling asleep and night-time waking, and a small improvement in total sleep time. ^[4]
• HABIT trial (2024): 642 adults in English primary care were given either nurse-delivered sleep restriction plus a sleep hygiene leaflet, or the leaflet alone. At six months, insomnia severity was lower with sleep restriction (by 3.05 points on the Insomnia Severity Index). Eight people in each group had serious adverse events and none were judged related to the treatment. ^[6]

How long does it last?

The honest answer is "it varies, and the data are sparse." In the Kyle study, sleepiness was measured weekly for four weeks and then at three months, so what happened in between is unknown. ^[1]

The AASM says the initial downsides typically resolve by the end of treatment. ^[3] Our reading is that tiredness which lingers or worsens is worth raising with your clinician instead of waiting it out. Tiredness that was there long before treatment is a different question, covered in Tired all day.

Why would it happen?

Nobody has shown. A 2018 review of the mechanism evidence found 15 studies of poor overall quality. None of them tested whether the proposed explanations account for the benefit. ^[5] Be wary of confident-sounding explanations, including ours.

Who should take extra care

The AASM guideline says sleep restriction may be contraindicated for people in high-risk jobs (for example drivers or heavy-machinery operators), people prone to mania or hypomania, and people with poorly controlled seizure disorders, because of the sleep loss involved. For stand-alone sleep restriction it adds excessive daytime sleepiness to that list. ^[3] If any of these apply, talk with your doctor before starting.

Driving and risky tasks

The two small studies disagree. One found slower reactions during the early weeks, and the other found no driving-simulator change. ^[1][2]

The guideline singles out drivers and machinery operators because sleepiness at the wheel or at a machine is a safety risk for anyone, and the early weeks of this treatment can add to it. The Kyle study's authors describe their findings as relevant to the safe delivery of treatment. ^[1][3] Drowsiness while driving or operating machinery is a reason to stop and to tell your clinician, not something to push through, and your own doctor is the person to ask about work that involves safety risks.

What do people say about sticking with it?

• Group sleep restriction in Swedish primary care (2025): 14 patients were interviewed. Understanding the reasons behind the treatment strengthened their commitment, and they described it as challenging but also offering solutions. Life circumstances and depressive symptoms got in the way of following it. ^[8]
• HABIT process evaluation (2024): 16 patients were interviewed. They described difficulty keeping going, negative attitudes and low confidence. Those who found the process positive improved, and those who struggled did not. This is observational and cannot show cause and effect. ^[7]
• AASM guideline: people may find the treatment hard because of the shorter time in bed and the extra tiredness, and for some groups (older adults, people with chronic pain or depression) it can be hard to find meaningful things to do with time that used to be spent in bed. ^[3]

Where the evidence is thin

• The sleepiness findings come from two studies of 16 people each, with no comparison group, and they differ. ^[1][2]
• The AASM rates the evidence for stand-alone sleep restriction as low quality. ^[3]
• Evidence on objective sleep and on daytime functioning is limited. ^[4]
• How the treatment works is unknown. ^[5]

Where does this leave you?

Feeling more tired early on is a documented effect that usually fades, and it should be taken seriously where driving or safety is involved. Small Hours is a six-week CBT-I program that works from your own sleep diary, and the diary is free. You can sign up at trysmallhours.com. For the wider picture, see CBT-I feels worse before it gets better.

General education, not medical advice.