How to sleep better without pills: the CBT-I method

The treatment sleep doctors recommend first for chronic insomnia is not a pill or a sleep-hygiene list. It is a set of rules about when you are in bed, and it starts with spending less time there.

A tree lit up at night in Santiniketan, West Bengal
Photo: Rajeeb Dutta, CC BY-SA 4.0, via Wikimedia Commons

To sleep better without pills, use cognitive behavioural therapy for insomnia (CBT-I): get up at the same time every day, go to bed only when sleepy, leave the bed if you cannot sleep, and temporarily cut your time in bed to roughly the hours you actually sleep, then extend it as your sleep consolidates. The American Academy of Sleep Medicine strongly recommends CBT-I for chronic insomnia, and a meta-analysis of 20 trials found it cut the time to fall asleep by 19 minutes and night waking by 26 minutes.

Key takeaways
  • CBT-I is the only "strong" recommendation for chronic insomnia in the AASM's 2021 guideline.
  • 19 minutes faster to fall asleep and 26 minutes less awake at night, across 20 trials.
  • Fixed wake time, seven days a week, is the anchor; only bedtime moves.
  • 85% and 90%: the sleep-efficiency thresholds for cutting or widening your window.
  • Sleep hygiene alone is not enough for chronic insomnia, and the guideline advises against relying on it.

Insomnia that lasts for months is usually kept going by habits that made sense at the time:

  • going to bed early to catch up;
  • lying in bed awake, trying harder;
  • napping in the afternoon or evening;
  • sleeping in on Sunday.

CBT-I targets those habits directly. Indians are short of sleep in general: in LocalCircles' 2025 "How India Sleeps" survey of more than 43,000 people in 348 districts, up to 59% said they got less than six hours of uninterrupted sleep. Short sleep is not the same as insomnia disorder, but the methods below help both.

19 minfaster sleep onset after CBT-I (Trauer et al. 2015, 20 trials)
26 minless time awake at night after CBT-I
+9.9 ptsgain in sleep efficiency
59%of Indians surveyed report under 6 h of uninterrupted sleep (LocalCircles 2025)

What is CBT-I, and why do guidelines put it before pills?

The story

Between August 2018 and March 2020, 642 adults with insomnia, aged 19 to 88, were signed up through 35 ordinary general practices across England. Half got a sleep-hygiene booklet. The other half got the same booklet plus four sessions of sleep restriction therapy, delivered not by sleep specialists but by practice nurses.

Six months later, the nurse-treated group scored 10.9 on the Insomnia Severity Index against 13.9 for the booklet group, a gap the researchers rated as a moderate-to-large effect. Eight people in each group had serious adverse events, none judged related to the treatment, and the cost worked out at about £2,076 per quality-adjusted life year. The method sleep clinics use had worked in a GP surgery.

Kyle et al.: the HABIT trial, The Lancet 2023 (Europe PMC abstract)

CBT-I is a short structured programme, often six to eight sessions, that combines behavioural rules about sleep timing with work on unhelpful beliefs about sleep. The American Academy of Sleep Medicine's 2021 guideline makes multicomponent CBT-I its only "strong" recommendation for chronic insomnia in adults.

The guideline was built on a systematic review that screened 1,244 studies, included 124 and analysed 89. It also gave conditional support to three single components on their own (stimulus control, sleep restriction and relaxation) and advised against sleep hygiene as a stand-alone treatment. That last point surprises people: the familiar list of "avoid caffeine, keep the room dark" is not enough by itself for chronic insomnia.

ComponentWhat you doAASM 2021 rating as a single treatment
Multicomponent CBT-IAll of the below, plus cognitive work, over several weeksStrong: recommended
Stimulus controlBed for sleep only; leave if awake; fixed wake time; no napsConditional: suggested
Sleep restrictionTime in bed cut to about your actual sleep time, then extendedConditional: suggested
Relaxation trainingProgressive muscle relaxation, slow breathingConditional: suggested
Sleep hygiene aloneCaffeine, light, noise, routine adviceConditional: suggested against

The effect sizes are clinically meaningful. James Trauer and colleagues pooled 20 randomised trials with 1,162 adults with chronic insomnia. Face-to-face CBT-I:

  • shortened the time to fall asleep by 19 minutes;
  • cut time awake after sleep onset by 26 minutes;
  • raised sleep efficiency by about 10 percentage points;
  • raised total sleep time by only about 8 minutes at first, because the treatment deliberately limits time in bed.

Gains appeared to last at later follow-ups, and no adverse outcomes were reported.

Average improvement after CBT-I compared with control (minutes)
See the data as a table
Improvement (min)
Time to fall asleep (shorter)19.03
Time awake after sleep onset (shorter)26.0
Total sleep time (longer)7.61

Source: Trauer et al., Annals of Internal Medicine 2015 (20 RCTs, 1,162 adults) · Post-treatment diary measures. The total sleep time confidence interval crossed zero (-0.51 to 15.74 min). Sleep efficiency rose 9.91 percentage points.

How do I do CBT-I at home, step by step?

Here is the core of the behavioural part, in the order clinicians usually introduce it. Give it four to six weeks; the first one or two will feel worse before they feel better.

  1. Keep a sleep diary for one to two weeks. Each morning, note when you got into bed, roughly when you fell asleep, how long you were awake in the night, when you woke and when you got up. Estimates are fine; do not watch the clock at night.
  2. Fix your wake-up time, seven days a week. Choose the time you need on workdays and keep it on Sundays too. This anchors your body clock more than any bedtime does.
  3. Calculate your sleep window. Average your nightly sleep from the diary. Your time in bed becomes that average, but never less than 5 hours. Count back from the wake time to get your earliest bedtime.
  4. Go to bed only when sleepy, and not before the window opens. Sleepy means eyes closing, head nodding, not just tired.
  5. If you are awake and not drifting off, get up. Go to another dim room, do something quiet and boring, and come back when sleepy. Repeat as often as needed. Use your judgment rather than a clock; the rule is about not lying there frustrated.
  6. Use the bed only for sleep and sex. No phone, laptop, TV or work in bed, so that bed becomes a cue for sleep again.
  7. No naps during the first weeks, especially not in the evening on the sofa.
  8. Adjust weekly by sleep efficiency. Sleep efficiency is time asleep divided by time in bed. Above 90% for a week: add 15 minutes to the window. Below 85% (80% if you are older): cut the window to your average sleep time again. In between: keep it the same.

These rules come from the sleep restriction protocol of Arthur Spielman and colleagues, as later refined, including the 5-hour floor that limits sleep loss in the first week. They work by building up sleep pressure and by breaking the link between bed and wakefulness.

The weekly sleep-window ruleOne week of diarySE = time asleep / time in bed x 100Above 90%add 15 minutes:go to bed alittle earlier85 to 90%keep the samewindow anotherweekBelow 85%cut window toaverage sleep,never under 5 hRepeat next weekOlder adults: use 80% instead of 85%. Wake time stays fixed throughout.Rules from Spielman's sleep restriction protocol.
The whole of sleep restriction on one card. Each week the diary decides one of three moves, and the wake time never changes.
Example: building a sleep window9 pm11 pm1 am3 am6:30Before:in bed 7 h 30 min, asleep ~5 h 45 min (SE 77%)Week 1:12:30 am to 6:15 am (5 h 45 min)Week 2:SE above 90%: bed at 12:15 amWeek 3:SE above 90% again: bed at 12 midnightWake time never moves. Only bedtime does.
Adapted from a worked example in Spielman's sleep restriction chapter: 7.5 hours in bed but under 6 hours of sleep. The window starts at the sleep you are actually getting and grows only as efficiency improves.

What is my sleep efficiency? Use the calculator

Enter your diary averages to get your sleep efficiency and a starting window under the standard rules.

Sleep efficiency and a starting sleep window

A wooden bedside table beside a bed in a historic house
Stimulus control means the bed and the bedside are for sleep. Keep the phone, the laptop and the to-do list somewhere elsePhoto: Thomas Quine, CC BY 2.0, via Wikimedia Commons

Does sleep restriction work outside a sleep clinic?

Yes. The strongest recent evidence comes from ordinary general practice: the HABIT trial told in the story above. In the trial, published in The Lancet in 2023, Simon Kyle and colleagues randomised 642 adults with insomnia at 35 English GP practices to four sessions of nurse-delivered sleep restriction plus a hygiene booklet, or the booklet alone.

At six months, insomnia severity scores averaged 10.9 in the sleep restriction group against 13.9 with the booklet, an effect size of d = 0.74, a moderate-to-large effect.

It was likely cost-effective at about £2,076 per quality-adjusted life year, and serious adverse events were equal in both groups (eight each), none judged related to treatment. The nurses had brief training, which suggests the method does not require a specialist.

Cognitive Behavioral Therapy for Chronic Insomnia: A Systematic Review and Meta-analysisJames M. Trauer, Mary Y. Qian, Joseph S. Doyle, Shantha M. W. Rajaratnam, David Cunnington · Annals of Internal Medicine · 2015

Pooled 20 randomised trials (1,162 adults, 64% women, mean age 56) of face-to-face CBT-I with at least three components against inactive controls. Sleep onset improved by 19.03 minutes, wake after sleep onset by 26.00 minutes, sleep efficiency by 9.91 percentage points and total sleep time by 7.61 minutes (not statistically certain). Benefits seemed sustained. Limit: trials excluded insomnia with other medical, sleep or psychiatric conditions, and later time points were less precise.

Read the abstract (free)
Clinical and cost-effectiveness of nurse-delivered sleep restriction therapy for insomnia in primary care (HABIT)Simon D. Kyle et al. · The Lancet · 2023

A pragmatic randomised trial at 35 English general practices: 642 adults (mean age 55, 76% women) got four nurse-led sleep restriction sessions plus a hygiene booklet, or the booklet only. Six-month insomnia severity was 10.9 versus 13.9 (adjusted difference 3.05 points, d = 0.74), and cost per QALY gained was about £2,076. Limit: open-label with self-reported outcomes, since patients knew which treatment they had.

Read the paper
Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guidelineJack D. Edinger et al. · Journal of Clinical Sleep Medicine · 2021

A task force graded the evidence with the GRADE method after a review that screened 1,244 studies and analysed 89. It strongly recommends multicomponent CBT-I, conditionally supports brief therapies, stimulus control, sleep restriction and relaxation on their own, and conditionally advises against sleep hygiene as the only treatment. Limit: guideline strength depends on trials mostly from high-income countries.

Read the guideline (free)

How do I adapt CBT-I to an Indian home?

The rules are universal, but Indian households add specific obstacles: late dinners, shared bedrooms, noisy streets, mosquitoes, and a family culture of afternoon naps. Here is how to handle each without breaking the method.

A white mosquito net hung over a bed in a tropical room
A mosquito net over a bed. In the LocalCircles survey, 22% of people with broken sleep blamed noise or mosquitoes; a net or repellent is a simple fix.Photo: Can Pac Swire from Toronto, Canada, CC BY-SA 2.0, via Wikimedia Commons
ProblemWhat to do
Dinner at 10 pm or laterYour sleep window will often start after midnight in week one anyway. Keep the wake time fixed and let bedtime float later; do not go to bed early to compensate.
Shared room, partner on the phone in bedAgree on lights and screens off when your window starts. If you must leave the bed at night, have a chair ready in another room.
Bathroom trips at nightThe most common disruption in the LocalCircles survey (72% of those with broken sleep). Limit fluids in the last two hours; if trips are frequent, see a doctor.
Street noise and mosquitoesEarplugs and a net or repellent. 22% of those with broken sleep blamed noise or mosquitoes.
Sunday afternoon napSkip it during the first weeks of restriction. 36% of LocalCircles respondents said they nap on Sunday afternoons to catch up, which weakens night sleep pressure.
Phone in bedCharge it outside the bedroom; use a cheap alarm clock. That one change also serves stimulus control.
A classic mechanical alarm clock
A plain alarm clock lets the phone charge outside the bedroom, and it sets the one time in CBT-I that never moves: your wake-up time, seven days a week.Photo: Santeri Viinamäki, CC BY-SA 4.0, via Wikimedia Commons

What about the thoughts that keep me awake?

The "C" in CBT-I is cognitive restructuring: finding and testing the beliefs that make insomnia worse. Common ones are:

  • "I must get eight hours or tomorrow is ruined";
  • "I'll never sleep normally again";
  • "if I'm awake at 3 am I should stay in bed and rest".
Most people with insomnia find they function better on short sleep than they predicted, and that discovery lowers the anxiety that kept them awake.

Each can be tested against your own diary. Most people with insomnia find they function better on short sleep than they predicted, which lowers the anxiety that kept them awake. Relaxation methods help the body side; slow breathing is the simplest, and slow breathing and the vagus nerve explains why it calms arousal. If you want a daily practice to pair with it, the beginner's meditation guide keeps it short.

Sleep also matters for more than mood: the sleep and memory consolidation post covers what the brain does with the day's learning at night. And a regular wake time is itself a habit; the habit guide shows how to anchor it.

When does insomnia need a doctor?

See a doctor if insomnia has lasted three months or more and affects your days, if it came with low mood or anxiety, or if anything in the warning box at the top applies. Loud snoring with gasping or daytime sleepiness can mean obstructive sleep apnoea, which needs a sleep study, and in which sleep restriction can be harmful.

Spielman's protocol excludes people who need constant alertness to avoid serious accidents, such as long-haul drivers and machine operators, and people whose conditions worsen with sleepiness or deep sleep, including epilepsy, parasomnias and sleep-disordered breathing. A psychiatrist, a sleep clinic or a trained clinical psychologist can deliver CBT-I with those safeguards. If cost is a barrier, the affordable therapy guide lists free and low-cost routes.

Questions people ask

Won't sleeping less make my insomnia worse?

For the first week or two you will feel sleepier, which is the mechanism working: sleep pressure builds and your sleep becomes deeper and less broken. The 5-hour floor limits the deficit, and the window grows by 15 minutes a week once efficiency passes 90%.

Is sleep hygiene useless then?

Not useless, but not enough alone for chronic insomnia. The AASM guideline suggests against using it as the only treatment. Keep the good habits and add stimulus control and a sleep window.

How long does CBT-I take to work?

Programmes typically run six to eight sessions over several weeks. The HABIT trial used four nurse sessions and measured clear benefit at six months, and in Trauer's meta-analysis gains appeared to hold at later follow-ups.

Can I do CBT-I with an app?

Digital CBT-I programmes exist and have been tested in trials. A 2026 review of economic studies found therapist-guided digital CBT-I more consistently cost-effective than self-help versions, and the health results of self-help apps ranged from slightly worse than to equal to face-to-face CBT-I. An app can be a reasonable start if the warnings above do not apply to you.

Can I keep taking my sleeping pill while I start?

Do not stop or change a prescribed medicine on your own; some need to be tapered. Discuss CBT-I with the prescribing doctor, who can plan both together.

Sources

  1. Trauer JM et al.: Cognitive behavioral therapy for chronic insomnia, a systematic review and meta-analysis, Annals of Internal Medicine (2015)
  2. Europe PMC: Kyle SD et al., HABIT trial abstract (recruitment dates, ages, adverse events; source for the story) (2023)
  3. Ministry of Health and Family Welfare: Tele MANAS, National Tele Mental Health Programme
  4. PIB: Update on National Tele Mental Health Programme (free, 24x7, 20 languages) (2026)
  5. Edinger JD et al.: Behavioral and psychological treatments for chronic insomnia disorder in adults, AASM clinical practice guideline, JCSM (2021)
  6. Edinger JD et al.: AASM systematic review, meta-analysis and GRADE assessment, JCSM (2021)
  7. Kyle SD et al.: Nurse-delivered sleep restriction therapy in primary care (HABIT), The Lancet (2023)
  8. Spielman AJ et al.: Sleep restriction therapy, in Behavioral Treatments for Sleep Disorders (Elsevier) (2011)
  9. Sleep Foundation: Cognitive behavioral therapy for insomnia
  10. ThePrint: 59% of Indians get less than 6 hours of sleep, LocalCircles survey (2025)
  11. AJMC: Economic evidence supports CBT-I for insomnia (Le et al., Applied Health Economics and Health Policy) (2026)
  12. EMJ: CBT-I improves insomnia during chemotherapy (Do et al., JAMA Network Open) (2026)
CBT-Iinsomniasleep restrictionstimulus controlsleep efficiencyAASM guidelinesleep hygiene
Sapta

Sapta (Saptarshi Sarkar) is a mechanical engineer turned marketer in Bengaluru and a solo touring rider. He writes the seven Saptaverse blogs. Work with him or read more of Mindblog.

Keep reading in Mindblog

From the other blogs