Insomnia Treatment: Why the Guidelines Put CBT-I Before Sleeping Pills

What counts as insomnia

Insomnia is defined by dissatisfaction with the quantity or quality of sleep. The classic patterns are trouble falling asleep, trouble staying asleep, and waking early without getting back to sleep. It is distinct from the occasional restless night, and roughly 6 to 10 percent of adults meet the diagnostic criteria.

A problem that common attracts remedies at the same rate. What the remedies rarely mention is the order the guidelines put them in.

What the evidence supports

The reference document is the American College of Physicians clinical practice guideline. It recommends that all adult patients with chronic insomnia disorder receive CBT-I as the initial treatment, graded a strong recommendation on moderate-quality evidence. Medication gets different handling: for patients in whom CBT-I alone fails, the guideline recommends a shared decision that weighs benefits, harms and costs, graded a weak recommendation on low-quality evidence. Pills are not forbidden. They are second.

The American Academy of Sleep Medicine's strong recommendation for multicomponent CBT-I was endorsed by the World Sleep Society, and the same endorsement names the treatment's biggest constraint: a shortage of trained therapists, with wide regional variation in training.

In May 2026 the AASM added a guideline on combination treatment. It advises against adding medication to CBT-I compared with CBT-I alone, on the grounds that behavioral and psychological therapy by itself often produces meaningful, durable improvement, and that adding drugs carries additional risk. The direction is consistent across a decade of guidance: the program first, the pill after.

If you already take sleeping pills

Nothing in this section means stop your medication. Rather the opposite.

The Korean drug regulator's prescribing information for zolpidem, one of the more explicit public documents on the subject, says the drug must not be taken above the recommended dose, and that if insomnia has not eased after 7 to 14 days of treatment, a doctor should check for other mental or physical conditions. Coming off the drug is also a procedure: tapering in steps according to how long you have taken it, alongside education about withdrawal symptoms, CBT-I and sleep hygiene education. Quitting overnight sits outside that procedure, and so does escalating on your own.

The same document carries a heavier passage. Worsening depression, including suicidal impulses, has been reported with sedative-hypnotics, mainly in patients with depression, and a 2016 epidemiological study reported increased suicidal behavior regardless of psychiatric history. A causal link has not been established, the document notes. If your mood has been sinking while you take these drugs, do not sort it out alone; talk to the prescriber or a professional support service.

How two countries compare

The comparison in this topic is not between treatments. It is between the guideline and what actually happens in consulting rooms.

Country	First-line recommendation	Actual practice	Access constraint	Level of evidence
United States	CBT-I, strong recommendation from ACP and AASM	Medication second, shared decision	Shortage of trained therapists	Medical, guideline
South Korea	Regulator's own document pairs tapering with CBT-I	About 1.26 million zolpidem patients in five months of 2025	No clear public route to CBT-I	Medical, regulatory document, plus practice

Table: insomnia treatment guidelines versus practice. Japan and the UK are absent because we could not verify sources for them.

United States: where the guideline came from, and where it is going

The evidence structure that made CBT-I first-line was assembled in the United States. The ACP and AASM guidelines are both American society documents, and the 2026 combination guideline leans hard on shared decision-making, the patient's values and treatment goals included.

The country that wrote the guideline hits the same wall as everyone else: not enough trained therapists. That constraint is why digital CBT-I, the treatment translated into software, has become an active research field. The recommendation is settled. The delivery is being rebuilt as an app.

South Korea: the guideline and the prescription pad

Start with the number. In the first five months of 2025, about 1.26 million people in South Korea were prescribed zolpidem. Around 750,000 of them, 60 percent, were women, and more than 130,000 were in their forties. Zolpidem remains the de facto first-line sleeping pill in practice, with dependence from repeat prescriptions a noted concern. The underlying demand keeps growing: national claims data counted about 670,000 people treated for sleep disorders in 2021, on an upward trend.

The interesting document here is the regulator's. The Korean food and drug ministry's prescribing information lays out step-by-step tapering and pairs it with CBT-I and sleep hygiene education. On paper, even the regulator points the same way as the international guidelines.

Then comes the gap. Ask where a patient would actually go to receive CBT-I in Korea and the answer is hard to find. We could not verify its insurance status or a public list of providers, and we would rather record that than guess. A guideline existing and a treatment being reachable are different things, and a large share of those 1.26 million prescriptions likely lives in the space between.

Put the two countries together and the state of this topic is clear. What the first-line treatment is has been settled. Who delivers it, where, and how is the part still being worked out, on both sides of the Pacific.

What you can do tonight

CBT-I is less exotic than its name. The components that recur in the guideline literature are sleep restriction, stimulus control and relaxation training: matching time in bed to time actually asleep, uncoupling the bed from everything except sleep, and training the body out of its tension. The guidelines are equally clear about what is not enough. Sleep hygiene education alone does not treat chronic insomnia; the condition needs targeted treatment.

And one more thing. If you already take a sleeping pill, tonight is not the night to quit by willpower. Tapering is a procedure, and it is one you walk through with a clinician.

When to see a doctor

Medication has not eased the insomnia after 7 to 14 days. That is a signal to check for other mental or physical conditions.
Depression or anxiety runs alongside the sleeplessness. The risk of long-term, escalating hypnotic use makes this worth a consultation.
The insomnia has become chronic. Chronic insomnia disorder calls for targeted treatment, and CBT-I is what the guidelines put first.

Frequently asked questions

Frequently asked

What does CBT-I actually involve?

The recurring components are sleep restriction, stimulus control and relaxation training, delivered as a structured program. It is a targeted treatment, which is the guidelines' point: general sleep hygiene advice alone is not considered sufficient for chronic insomnia.

Why do the guidelines advise against combining CBT-I with medication?

The 2026 AASM guideline found that behavioral and psychological treatment alone often produces meaningful and durable improvement, while adding medication brings additional risk. Combination is therefore not preferred over CBT-I alone.

Is insomnia just bad sleep hygiene?

No. The guideline position is that chronic insomnia disorder is a condition needing targeted treatment, and that sleep hygiene education by itself falls short. Hygiene advice is a floor, not a therapy.

I already take sleeping pills. Should I stop?

Not on your own. Regulatory guidance sets out gradual tapering matched to how long you have taken the drug, together with withdrawal education, CBT-I and sleep hygiene education. If the pills have not helped after 7 to 14 days, that is a reason to go back to the prescriber, not to self-adjust.