CBT-I: An Insomnia Treatment Doctors Now Recommend Over Pills
Millions of people lie awake each night, watching the clock and willing themselves to fall asleep. Sleep medicine has a treatment for chronic insomnia that outperforms pills in head-to-head trials and skips the dependency risk that comes with nightly sedatives, yet most people with insomnia have never heard of it. Cognitive behavioral therapy for insomnia, known as CBT-I, retrains the thoughts and habits that keep the brain wired for wakefulness at bedtime, using techniques tested in randomized trials spanning thousands of patients. Major medical bodies, including the American College of Physicians and the American Academy of Sleep Medicine, now call it the preferred starting point for treating chronic insomnia, ahead of sleeping pills.
Randomized Trials Show Large, Consistent Gains
A 2015 meta-analysis by sleep researcher Trauer and colleagues, published in Annals of Internal Medicine and later summarized by the sleep-research team at RISE Science, pooled 20 randomized controlled trials covering 1,162 adults with chronic insomnia and no other major sleep disorder. Patients receiving CBT-I fell asleep about 19 minutes faster than before treatment and spent about 26 fewer minutes awake after waking in the night. Sleep efficiency, the percentage of time in bed spent asleep, rose by nearly 10 percentage points on top of those gains.
Total sleep time rose by only about 7.6 minutes, too small a gain for the researchers to rule out, which tells a specific story: CBT-I mostly fixes broken, fragmented sleep rather than simply adding more hours in bed. No trial in the meta-analysis reported adverse effects from the therapy, and the gains held up, though with less precision, at later follow-up points.
Head-to-Head Trials Favor Therapy Over Pills
A randomized trial led by behavioral sleep researcher Gregg Jacobs, covered by the Harvard Gazette in 2004, compared CBT, the sleep medication zolpidem (Ambien), a combination of the two, and placebo in adults with chronic sleep-onset insomnia, treating each group for eight weeks. Midway through treatment, the CBT and combination groups showed the strongest improvement in how quickly patients fell asleep, while the zolpidem-only group improved more modestly.
Once patients stopped taking zolpidem, those modest gains faded, but the CBT and combination groups kept their improvement through the end of the trial and produced the highest share of patients who became "normal sleepers" by standard sleep-lab criteria. Adding zolpidem on top of CBT gave patients no extra benefit over CBT by itself, a result Jacobs summarized plainly: he said behavioral techniques worked better than sleeping pills, the most common insomnia treatment.
Major Guidelines Now Recommend CBT-I First
In 2016, the American College of Physicians reviewed the evidence on chronic insomnia treatment and published formal guidance in Annals of Internal Medicine recommending CBT-I as the first-line, initial treatment for adults, ahead of medication. ACP President at the time, Wayne J. Riley, explained the reasoning in a statement: "Cognitive behavioral therapy for insomnia is an effective treatment and can be initiated in a primary care setting," he said, adding that it "is likely to have fewer harms" since "sleep medications can be associated with serious adverse effects."
ACP's guideline recommends starting patients on CBT-I, then adding medication through shared decision-making only if CBT-I alone fails to help, and it flags that FDA-approved sleep drugs are meant for short courses of four to five weeks rather than ongoing nightly use.
A 2026 Update Confirms CBT-I's Edge Even in Combination
In April 2026, the American Academy of Sleep Medicine issued fresh guidance, published in the Journal of Clinical Sleep Medicine, addressing a question its earlier guidelines hadn't tested directly: does combining CBT-I with medication outperform either one alone? Lead author Daniel Buysse of the University of Pittsburgh concluded that it mostly doesn't change the underlying advantage of therapy: "Our analysis suggests that CBT-I by itself is the most efficacious first-line treatment for insomnia," he said, noting that adding medication "may provide modest benefit for some specific outcomes, such as total sleep time."
AASM's 2026 guideline favors combining CBT-I with medication over medication alone, but it recommends against adding medication to CBT-I for most patients, since CBT-I's benefits tend to last without the ongoing risks that come with sedative-hypnotics. Both recommendations carry a "conditional" strength and rest on evidence the panel rated as low-certainty, a limitation the guideline states directly.
Medication Works Fast, Therapy Keeps Working
Sedative-hypnotics like zolpidem and eszopiclone can improve sleep within days, which makes them attractive for short-term crises, but their benefits tend to fade once a patient stops taking them, as the Jacobs trial above demonstrated. A 2012 review in BMC Family Practice led by researcher Matthew Mitchell rated the evidence favoring CBT-I over the long term as low to moderate quality, while rating the evidence favoring benzodiazepines in the short term as very low quality, a gap that reflects how few trials directly compare the two treatments over extended periods.
Follow-up data from a later trial found that 48 to 74 percent of patients who completed a six-week course of CBT-I remained in remission from insomnia a year later, with 44 to 63 percent still in remission at two years. Medication rarely produces numbers like that once the prescription runs out, since most hypnotics are approved only for a few weeks of continuous use in the first place.
Bottom Line
Randomized trials, meta-analyses, and now two generations of national guidelines point toward the same conclusion: CBT-I produces durable improvements in falling asleep and staying asleep, carries no recorded adverse effects in the trials reviewed, and keeps working well after treatment ends, something sedative medication rarely manages once a patient stops taking it. Access remains the real obstacle, since trained CBT-I providers are scarce in many regions and sessions take more time and effort than swallowing a pill. Insurance coverage for CBT-I also varies widely across plans, which is part of why so many doctors still reach for a prescription pad first despite guidelines that say otherwise. Patients who can find a CBT-I provider or a structured app-based program covering the same core techniques stand to gain more from the investment than most would expect from a medication-only approach.
CBT-I works because it targets the same habits and daily rhythms that shape sleep more broadly, and our piece on why sleep anchors overall wellness covers the wider case for treating sleep as a foundation rather than an afterthought.
Sources
American College of Physicians — ACP Recommends Cognitive Behavioral Therapy as Initial Treatment for Chronic Insomnia
American Academy of Sleep Medicine — Combination Treatment Chronic Insomnia Guideline
Harvard Gazette — Study Says Therapy Better Than Pills in Treating Sleep-Onset Insomnia
RISE Science — Is CBT-I Effective?



Comments