The CBT-I Question: What 40 Years of Insomnia Research Reveal About the First Line Treatment Doctors Still Underprescribe

In 1972, a psychologist at the University of Arizona named Richard Bootzin published a short paper describing a behavioral protocol for people who could not sleep. The instructions were almost embarrassingly simple. Go to bed only when sleepy. Use the bed only for sleep. If you cannot fall asleep within about 20 minutes, get out of bed and do something quiet in another room. Return to bed only when sleepy again. Wake at the same time every morning, regardless of how the night went. He called it stimulus control therapy.

Presented By Our Partners

A few years later, a clinical psychologist at Beth Israel Hospital in New York named Arthur Spielman added a second behavioral lever. He observed that people with chronic insomnia tended to spend a great deal of time in bed, hoping that sheer opportunity would translate into sleep. The result was the opposite. Long fragmented nights eroded the association between bed and sleep and reduced the homeostatic pressure that produces deep sleep. Spielman’s response was sleep restriction therapy, a counterintuitive protocol that temporarily limits time in bed to match the actual hours of sleep the person is getting, then expands the window slowly as efficiency improves.

In 1993, a Canadian psychologist at Laval University named Charles Morin combined these two behavioral protocols with cognitive restructuring of unhelpful sleep beliefs and sleep hygiene education into a multicomponent treatment he called Cognitive Behavioral Therapy for Insomnia, abbreviated CBT-I. In the three decades since, CBT-I has accumulated what is, by the standards of behavioral medicine, an unusually robust evidence base. Dozens of randomized controlled trials. Multiple meta-analyses. Endorsement as a first line treatment by the American College of Physicians in 2016, the American Academy of Sleep Medicine in 2017, the European Sleep Research Society in 2017, and the British Association for Psychopharmacology in 2019.

And yet, in the United States in 2026, fewer than 15 percent of patients with chronic insomnia receive CBT-I before being prescribed a sleep medication. The most evidence backed treatment for the most common sleep disorder is the one most often skipped.

This evening, the question is not whether CBT-I works. The trial literature has answered that question. The question is what 40 years of research actually shows, why the gap between the evidence and clinical practice has persisted, and what an honest reading of the science suggests for a person dealing with chronic insomnia this week.

What Chronic Insomnia Actually Is

The clinical definition of chronic insomnia, codified in the International Classification of Sleep Disorders Third Edition and refined in the Fourth, has three components. First, a complaint of difficulty initiating sleep, maintaining sleep, or waking too early, despite adequate opportunity to sleep. Second, daytime consequences such as fatigue, mood disturbance, impaired performance, or distress. Third, persistence of the problem for at least three months at a frequency of three nights per week or more.

The epidemiology is striking. Roughly 10 percent of adults meet the full criteria for chronic insomnia disorder. Another 20 to 30 percent have insomnia symptoms that fall short of the disorder threshold but still impair function. The condition is more common in women than in men, in older adults than in younger ones, and in people with concurrent medical or psychiatric conditions. It is one of the most common reasons adults present to primary care, and one of the most consistently mismanaged.

The cost of chronic insomnia, in cohort data, is substantial. Higher rates of cardiovascular disease, type 2 diabetes, depression, anxiety, and accidents. Lower quality of life and work productivity. Increased healthcare utilization. The signal is consistent enough that the National Institutes of Health and the American Academy of Sleep Medicine have both treated insomnia as a public health condition with population level implications, not simply a personal complaint.

The Spielman Model of Insomnia

The conceptual foundation of CBT-I is a model of insomnia that Arthur Spielman published in 1987 and that has held up well in subsequent research. Spielman’s framework, often called the 3P Model, distinguishes three categories of factors.

Predisposing factors are the traits a person brings to insomnia risk: family history, anxiety prone temperament, hyperarousability, female sex, older age. These set the baseline vulnerability. Precipitating factors are the acute triggers that move a person from a vulnerable state into actual insomnia: a death in the family, a job loss, an illness, a divorce, a difficult shift change. Most acute insomnia resolves on its own when the precipitant resolves. Perpetuating factors are the behaviors and beliefs a person adopts to cope with the acute insomnia, and which then keep it going long after the original cause has passed.

The perpetuating factors are where CBT-I does its work. Spending more time in bed to compensate for poor sleep. Drinking caffeine to push through fatigue, which then disrupts the next night. Worrying about the consequences of bad sleep, which produces the arousal that prevents sleep. Napping during the day, which reduces homeostatic sleep pressure. Going to bed earlier on the theory that more opportunity will yield more sleep. All of these are intuitive responses to a frustrating problem. All of them tend to make the problem worse over time.

The CBT-I therapist’s central insight is that breaking the perpetuating cycle is enough, in most cases, to allow normal sleep regulation to reassert itself. The treatment does not need to address the original precipitant. It needs to dismantle the maladaptive coping that has kept the insomnia alive after the precipitant passed.

The Five Components of CBT-I

A standard course of CBT-I delivered by a trained clinician includes five components, typically across four to eight sessions.

Stimulus control therapy, the Bootzin protocol, rebuilds the association between bed and sleep. Use the bed for sleep and intimacy only. Go to bed only when sleepy. Leave the bed if you have been awake for 20 minutes or more. Return only when sleepy again. Wake at the same time every morning. Avoid daytime naps. The instructions sound trivial. They are operationally difficult to follow for several weeks and are responsible for a large fraction of the treatment effect.

Featured Partner

Invest in the Infrastructure Behind Modern Medicine

As healthcare expands beyond hospital walls, the buildings and campuses supporting that shift are generating compelling returns for investors who move early. The Healthcare Real Estate Fund offers qualified investors direct access to a curated portfolio of medical office, outpatient, and specialty care facilities.

Learn More →

Sleep restriction therapy, the Spielman protocol, temporarily compresses the sleep window. If a patient reports an average of 6 hours of actual sleep over an 8 hour time in bed, the prescribed sleep window is initially set at 6 hours. The patient picks a fixed wake time and counts backward to the bedtime. The window is expanded by 15 to 30 minutes only after sleep efficiency, the ratio of actual sleep to time in bed, exceeds about 85 percent for several consecutive nights. The early week of sleep restriction is hard, often producing daytime sleepiness. The compressed window restores deep sleep architecture and consolidates the night, after which the window can be gradually expanded.

Cognitive therapy targets the beliefs and worries that maintain insomnia. Catastrophic thinking about the consequences of a bad night. Unrealistic expectations about sleep need. The belief that one bad night requires recovery sleep on subsequent nights. The therapist works with the patient to identify these cognitions and develop more accurate, less arousal producing alternatives.

Sleep hygiene education addresses the environmental and behavioral factors that support or undermine sleep: caffeine, alcohol, exercise timing, bedroom temperature, light exposure, screen use. Sleep hygiene alone is generally insufficient to treat chronic insomnia, a point the literature has been clear about for two decades. As one component of a multicomponent program, it is useful.

Relaxation training provides specific techniques to reduce the autonomic arousal that keeps anxious sleepers in a state of physiological readiness for waking. Progressive muscle relaxation, paced breathing, guided imagery, and body scan techniques are the most commonly used.

What the Evidence Says

The trial literature on CBT-I is unusually consistent for a behavioral intervention. A 2015 meta-analysis by James Trauer and colleagues, published in Annals of Internal Medicine, aggregated 20 randomized controlled trials. The pooled effect on sleep onset latency was a 19 minute improvement. Wake after sleep onset improved by 26 minutes. Sleep efficiency improved by roughly 10 percentage points. These improvements were sustained at follow up assessments ranging from three to twelve months.

A 2019 meta-analysis in JAMA Internal Medicine by Eric van Straten and colleagues, including 87 studies and over 11,000 participants, came to similar conclusions. CBT-I produced moderate to large effects on insomnia symptoms, with effect sizes comparable to or exceeding those reported for sleep medications in their respective trial literatures, and with effects that persisted long after treatment ended. Sleep medications, by contrast, lose effect once they are stopped and carry risks that accumulate with chronic use.

Head to head trials are particularly informative. A 1999 trial by Charles Morin in JAMA compared CBT-I to temazepam, a benzodiazepine commonly used for insomnia, in older adults with chronic insomnia. Both treatments produced short term improvement. Twenty four months after treatment ended, only the CBT-I group had maintained the improvements. A 2004 trial in JAMA by Gregg Jacobs compared CBT-I to zolpidem, the brand name Ambien, the most prescribed sleep medication in the United States. CBT-I outperformed zolpidem on the primary outcome of sleep onset latency, both at the end of treatment and at six month follow up.

The clinical guidelines have been unambiguous about what this evidence base supports. The American College of Physicians 2016 guideline recommended CBT-I as first line treatment for all adults with chronic insomnia, before any medication is considered. The American Academy of Sleep Medicine 2021 guideline reached the same conclusion. The European and British guidelines concurred. The case is, in evidence based medicine terms, closed.

Why CBT-I Is Underprescribed

The gap between guideline recommendations and clinical practice has several causes. There are not enough trained CBT-I therapists. A 2016 survey identified roughly 750 to 1,000 certified CBT-I providers in the entire United States, against an adult chronic insomnia population in the tens of millions. Primary care physicians are rarely trained in the protocol. Insurance coverage for CBT-I has been spotty, although it has improved meaningfully since 2018. Patients often expect a prescription and feel a behavioral treatment is not real medicine. And prescription sleep medications are easier, faster, and historically more profitable to deliver.

Several digital CBT-I programs have emerged to address the capacity gap. The most studied is Sleepio, a digital CBT-I program from Big Health, evaluated in multiple randomized trials including a 2022 JAMA Psychiatry study that showed efficacy comparable to in person CBT-I in a large primary care population. Other digital options include SHUTi, CBT-I Coach from the US Department of Veterans Affairs, and several smartphone based programs. The evidence base for digital CBT-I is now substantial enough that the American Academy of Sleep Medicine and several insurance plans have begun to treat it as a valid first line option.

The Role of Sleep Medications

The trial literature on sleep medications is, by contrast, more limited than the marketing suggests. The benzodiazepines, including temazepam and lorazepam, produce reliable short term improvement in sleep onset and maintenance but carry tolerance, dependence, and cognitive side effects that accumulate with chronic use. The newer non benzodiazepine hypnotics, including zolpidem, eszopiclone, and zaleplon, were initially positioned as safer alternatives but have produced their own set of concerns including complex sleep behaviors, next day impairment, and a black box warning from the FDA in 2019 for sleepwalking and sleep driving.

The dual orexin receptor antagonists, including suvorexant, lemborexant, and daridorexant, are a newer pharmacologic class with somewhat better safety profiles in older adults. They have a role for patients who cannot access or tolerate CBT-I. Trazodone, while widely prescribed off label for sleep, has a thinner evidence base than its prescription frequency suggests. Over the counter antihistamines such as diphenhydramine carry significant anticholinergic effects, particularly in older adults, where they have been associated with increased dementia risk in cohort data.

None of these agents address the perpetuating factors that maintain chronic insomnia. They can suppress symptoms while the patient is taking them. They do not retrain the brain to sleep.

What This Means For Your Practice

The bridge from insomnia research to a real practice is, in this case, unusually direct. Forty years of evidence point to a behavioral protocol that works, that lasts, and that has a benign side effect profile compared to medications. The protocol is operationally demanding for the first several weeks. The science says it is worth the discomfort.

If you have chronic insomnia, three months or more of difficulty sleeping at least three nights per week, ask your physician about CBT-I before accepting a sleep medication prescription. If your physician cannot refer you to a trained CBT-I clinician, ask about digital CBT-I programs that have evidence based clinical trial backing.

Set a fixed wake time. Pick a time that is feasible seven days a week and protect it. Sleeping in to recover lost sleep is one of the most common perpetuating behaviors and one of the most counterproductive. The fixed wake time anchors the circadian system.

Limit your time in bed to roughly match your actual sleep. If you sleep an average of 6 hours and spend 8 hours in bed, your sleep efficiency is 75 percent. Compressing the window to 6.5 hours, picking the bedtime by counting backward from your wake time, will feel restrictive in the first week. It is what consolidates the night.

Get out of bed if you have been lying awake for 20 minutes or more. Do something quiet and dim in another room. Read a paper book. Avoid screens. Return only when sleepy. The rule is hard. It is also one of the most effective components of CBT-I.

Use the bed for sleep only. Working, watching television, scrolling on a phone, and arguing in bed all weaken the conditioned association that produces sleep. The bedroom should be a place the brain reliably associates with sleep, not stimulation.

Skip the daytime nap, particularly in the first weeks of behavioral treatment. Daytime sleep reduces homeostatic pressure for night sleep. A 20 minute nap before 2 PM is generally tolerable for people not in active treatment. During CBT-I, naps slow progress.

Reduce caffeine after noon and alcohol within three hours of sleep. Caffeine’s half life is six to eight hours and longer in older adults. Alcohol’s specific suppression of REM and slow wave sleep is unambiguous in the literature. Both are common perpetuating factors that respond to simple change.

Identify the catastrophic thoughts about sleep. The belief that one bad night will ruin tomorrow, that adults need eight hours, that you will never sleep normally again. These thoughts produce the arousal that keeps you awake. Replace them with more accurate alternatives. Sleep is robust. One bad night is not a catastrophe. Your sleep will normalize as the behavioral protocol does its work.

Be consistent for at least four to six weeks before judging the protocol. The first two weeks are typically the hardest. Sleep efficiency, the ratio of actual sleep to time in bed, is the most useful outcome to track. Most patients see meaningful improvement by week three or four. Maintenance gains continue for months after the active program ends.

Sleep medications have a role for short term acute insomnia, for select medical situations, and for patients who cannot tolerate or access CBT-I. They are not, by the standards of the trial literature, the right first treatment for chronic insomnia in most adults. The evidence on this point has been clear for nearly two decades. The most important conversation a patient with chronic insomnia can have with their physician is the one that begins with the question, can we try the behavioral treatment first.

The work is unglamorous. The protocol is demanding for the first several weeks. The science is as strong as behavioral medicine evidence gets. Forty years of trials say that the brain can relearn how to sleep, and that the relearning, once it happens, tends to stick. That is the kind of intervention that earns the word fundamental.

Free Daily Briefing

The Latest Longevity Science.
Delivered Every Morning.

Join researchers, physicians, and health professionals getting daily breakthroughs in AI-driven medicine, epigenetics, and longevity research.

Support the research that powers this editorial

Subscribe Free

No spam. Unsubscribe anytime. We respect your inbox.