Sleep restriction therapy — the most effective insomnia treatment most people have never heard of

Cognitive-behavioural therapy for insomnia (CBT-I) is the first-line treatment for chronic insomnia in every major international guideline — the American Academy of Sleep Medicine, the European Insomnia Network, NICE, and the American College of Physicians all recommend it before prescription medication for adults with chronic insomnia. Its most powerful single component is sleep restriction therapy. The mechanism is counter-intuitive: you spend less time in bed, on purpose, for a defined period, in order to sleep more reliably afterwards. The catch is that the first 7–10 days are demanding, and the protocol requires daily sleep-diary compliance to work. When delivered properly it produces durable improvements in sleep efficiency and total sleep time that hold long after the protocol ends, with no dependence, no tolerance, and no medication interaction. It should be the first thing tried in chronic insomnia, before — or alongside — any prescription.

What CBT-I actually is

Cognitive-behavioural therapy for insomnia is a structured 5- to 8-session protocol, traditionally delivered by a trained sleep psychologist, that addresses the thoughts and behaviours maintaining chronic insomnia. The original protocol — developed in the 1980s and refined since — has five components:

1. Sleep restriction therapy — the engine of the whole protocol.
2. Stimulus control — re-pairing the bed with sleep.
3. Cognitive therapy — addressing the catastrophic thoughts that maintain hyperarousal.
4. Sleep hygiene education — the small environmental fixes.
5. Relaxation training — usually progressive muscle relaxation or guided imagery.

Of these, sleep restriction tends to do most of the work in head-to-head dismantling trials. Hygiene and relaxation, on their own, are not usually enough for chronic insomnia. They help. They do not fix it.

Why sleep restriction works

The mechanism is straightforward once you see it. Chronic insomnia is sustained by two physiological errors:

1. The bed has stopped being a sleep cue. Months of lying in bed not sleeping have re-trained the brain to associate bed with wakefulness, anxiety, and rumination.
2. The sleep drive is being diluted. People with chronic insomnia spend long hours in bed trying to sleep, often for 9 or 10 hours, ending up with maybe 5 hours of broken sleep. The body learns to spread its sleep need across the whole window, so no part of the night accumulates enough pressure to sleep deeply.

Sleep restriction fixes both. The patient is given a strict, narrow time-in-bed window — initially set to roughly the patient's actual sleep duration (e.g. 5 hours and 30 minutes). They go to bed at the agreed time and get up at the agreed wake time, no matter how poorly they sleep. The first 7 to 10 days are awful — sleep-deprived, foggy, hard to drive safely. Then two things happen:

• The sleep drive becomes overwhelming. The patient falls asleep within minutes of getting in bed, every night.
• Sleep efficiency climbs from typically 50–60% to over 90%. The bed is now a sleep cue again.

Once sleep efficiency is sustained above ~85% for a week, the time-in-bed window is extended by 15 minutes per week, slowly, until natural sleep need is met (usually around 7 hours for most adults). The result is a stable, efficient sleep pattern that does not require medication and tends to hold.

Why it is not used more often

Three reasons, in plain language:

1. It is hard. Patients are asked to deliberately sleep-deprive themselves for 1–2 weeks. Most cannot do it without support and a clear explanation of the mechanism.
2. It requires skilled delivery. A clinician needs to set the time-in-bed window correctly (too short = unsafe; too long = no effect), monitor sleep diaries, and adjust weekly. Done poorly, it does not work.
3. It is not commercially aligned. A patient who completes CBT-I and is sleeping reliably has no further need for prescriptions, supplements, or recurring services. There is no recurring-revenue model for delivering it well.

Despite all of this, the evidence base is overwhelming. Head-to-head trials of CBT-I versus zolpidem, trazodone, benzodiazepines, and cognitive therapy alone all show CBT-I winning on long-term outcome.

Who is a good candidate

A short list:

• Adults with chronic insomnia who want a non-pharmacological first-line.
• Patients who tried a medication and want to come off it — CBT-I done concurrently with a slow taper is the single most successful exit strategy.
• Patients with insomnia and a history of substance use — medication carries higher risk; CBT-I has none.
• Older adults — fall risk and confusion risk from any sleep medication go up sharply over 65; CBT-I has neither.

Not the right candidate:

• Patients with active depression, mania, or psychosis — those need treating first.
• Patients with untreated sleep apnea — the apnea is the cause of the disrupted sleep, not the conditioning. CBT-I will not fix it.
• Patients in an acute crisis — when the immediate problem is no sleep at all, the right answer is usually a short medication course first, then CBT-I once the patient is functional enough to engage.

How Slumbr fits this

Slumbr does not deliver CBT-I as a stand-alone product — there are several excellent specialist services for that in South Africa and several validated app-based programmes (Sleepio, Somryst, CBT-i Coach are international names; local equivalents exist). What Slumbr does:

• Identifies when CBT-I is the right next step — via the Sleep Pattern Assessment™ and the specialist consultation.
• Refers to a vetted CBT-I provider where the patient is a good candidate.
• Supports the patient through the first 1–2 weeks of sleep restriction (the hardest part) with low-dose, dependence-free pharmacological support if clinically appropriate — typically prolonged-release melatonin for over-55s, or briefly a Schedule 0 sleep-support formulation.
• Manages any prescription medication a patient is already on and helps with a structured taper alongside CBT-I, where that is the right approach.

The clinic that earns recurring revenue from medication has every reason to avoid recommending CBT-I. We recommend it where it is right, and our consultation fee is the same whether you leave with a prescription or with a CBT-I referral.

What this is not

This article is general clinical information, not a diagnosis. If you are in crisis or experiencing thoughts of self-harm, please contact SADAG on 0800 567 567 (24/7) or your nearest emergency department. Slumbr Sleep Clinic does not provide emergency care.

If your insomnia has been chronic for more than 3 months and basic hygiene fixes have not helped, the right first step is the free Sleep Pattern Assessment™ — which will tell you whether you are a good CBT-I candidate, a medication candidate, or both.

Reviewed by an HPCSA-registered specialist physician with sleep-medicine training. References on file. Last updated May 2026.