Somnera
CBT-I science

CBT-I vs sleeping pills: what the evidence actually compares

Short-term they look similar. The difference shows up later, and in what happens when you stop. What the guidelines say, and why this is not an either/or.

5 min read

Last reviewed September 2, 2026

Two paths rising together to a marked point, after which one holds its level and the other falls away.

Share

Over six weeks they can look similar. The interesting comparison is what your sleep does in the weeks after each one stops.

That is the whole argument, and it is the reason every major guideline puts them in the order it does.

What the guidelines actually say

The American College of Physicians, 2016: all adult patients should receive CBT-I as the initial treatment for chronic insomnia. Strong recommendation. Their second recommendation - about short-term medication - is graded weak, on low-quality evidence, and is framed as a shared decision for people in whom CBT-I alone was not enough.

The AASM's behavioral guideline, 2021: multicomponent CBT-I is its only strong recommendation.

The AASM's medication guideline, 2017: fourteen drug recommendations, every one of them weak, on low or very low quality evidence. Its own text says all patients with chronic insomnia should receive CBT-I as the primary intervention, with medication considered mainly for people who cannot take part in CBT-I, who still have symptoms after it, or as a temporary adjunct.

The European guideline, 2023: CBT-I first-line for adults of any age, in person or digital. Benzodiazepines and the newer z-drugs "can be used for the short-term treatment of insomnia" - defined there as four weeks or less.

None of that says medication is bad. It says the evidence supporting it is weaker, and it is meant for shorter.

The head-to-head trial worth knowing

Most comparisons are indirect. One is not.

A double-blind, placebo-controlled trial in older adults with chronic insomnia compared CBT against zopiclone against placebo, measuring sleep with polysomnography rather than asking people how they slept.

At six weeks, measured sleep efficiency went from 81.4% to 88.9% in the CBT group. In the zopiclone group it went from 82.3% to 81.5% - essentially nowhere. Time awake during the night fell 52% with CBT and 4% with the drug. Slow-wave sleep rose 27% with CBT and fell 20% on the medication.

At six months: 90.1% against 81.9%.

Total sleep time did not differ significantly between the groups - which is consistent with the pooled data on CBT-I, where total sleep time is the one measure that does not reliably move.

Where the real difference is

Not the size of the effect at week six. The mechanism.

A hypnotic changes your sleep while it is in your system. It does not change the schedule, the associations, or the beliefs that are maintaining the problem - so when it stops, the thing that was producing the insomnia is still there, and often the first few nights are worse than baseline before settling. That rebound is expected and temporary, and it is frequently misread as proof that the drug was necessary.

A behavioral program changes the inputs. When it ends, nothing is withdrawn. Pooling trials with long follow-up, the effect on insomnia severity is moderate at three months and smaller but still present at twelve.

Take the free assessment to see:

  • ✓Your Insomnia Severity Index (ISI) score
  • ✓A personalized 6-week CBT-I plan, built by a sleep physician
  • ✓How your sleep improves, night by night
Start free assessment

That is a durability argument, not a potency argument. It is worth being precise about the difference.

This is not an either/or

The framing that makes people put off starting is the belief they have to choose.

In practice a very common starting point is: still taking something, begin the behavioral work anyway, and have the medication conversation with your prescriber later, from a better position.

The 2026 AASM guideline on combination treatment lands somewhere careful. It suggests combining CBT-I with medication over medication alone. It also suggests against combining them over CBT-I alone. Both are conditional recommendations on low-certainty evidence. Read plainly: adding CBT-I to medication is better than medication by itself, and there is no demonstrated advantage to adding medication to CBT-I.

If you want to come off something

This is a conversation with the person who prescribed it, and there is a real reason for that: stopping some sleep medications abruptly, particularly benzodiazepines, can be genuinely harmful. Do not adjust or stop anything on your own, and do not use anything on this page as a schedule.

What the evidence supports is that the behavioral work makes the attempt more likely to succeed. In a randomized trial of older adults who had been taking a benzodiazepine long-term, being medication-free at the end of treatment was achieved by 85% of those doing a supervised taper combined with CBT, against 48% for taper alone and 54% for CBT alone. About 60% across groups were still off it at twelve months.

The practical order that follows from that: start the behavioral program, get some weeks of diary data, then take both to your prescriber and plan the taper with them. Not the reverse.

How to have that appointment covers what to bring.

What this article is not

It is not advice about your prescription. There are no doses here, no drug comparisons, no suggestion to switch or stop anything, and none of that is a decision that should be made from a website - including this one.

What it is: the reason the guideline hierarchy looks the way it does, so that when your clinician says "let's try the behavioral approach first," you know that is not a brush-off. It is what the guidelines say.

More in CBT-I science, including how long the behavioral route takes.

Somnera is a self-guided education program built on CBT-I, founded and written by Dr. Camilo Ruiz, DO, FACOI, FAASM. It is not a diagnosis, does not replace care from your own clinician, and gives no guidance about medication. The assessment is free.

Citations

  • Qaseem A, et al. Management of chronic insomnia disorder in adults: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2016;165(2):125–133. doi:10.7326/M15-2175
  • Edinger JD, et al. Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2021;17(2):255–262. doi:10.5664/jcsm.8986
  • Sateia MJ, et al. Clinical practice guideline for the pharmacologic treatment of chronic insomnia in adults. J Clin Sleep Med. 2017;13(2):307–349. doi:10.5664/jcsm.6470
  • Buysse DJ, et al. Combination treatment for chronic insomnia disorder: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2026;22(1):56.
  • Riemann D, et al. The European Insomnia Guideline: an update 2023. J Sleep Res. 2023;32(6):e14035. doi:10.1111/jsr.14035
  • Sivertsen B, et al. Cognitive behavioral therapy vs zopiclone for treatment of chronic primary insomnia in older adults. JAMA. 2006;295(24):2851–2858. doi:10.1001/jama.295.24.2851
  • Morin CM, et al. Randomized clinical trial of supervised tapering and cognitive behavior therapy to facilitate benzodiazepine discontinuation in older adults with chronic insomnia. Am J Psychiatry. 2004;161(2):332–342. doi:10.1176/appi.ajp.161.2.332
  • van der Zweerde T, et al. Cognitive behavioral therapy for insomnia: a meta-analysis of long-term effects. Sleep Med Rev. 2019;48:101208. doi:10.1016/j.smrv.2019.08.002

Frequently asked questions

It is graded higher. The AASM gives multicomponent CBT-I its only strong recommendation for chronic insomnia; in the medication guideline, all fourteen drug recommendations are weak on low or very low quality evidence. The clearest difference is what happens after treatment stops.

No. Many people start a behavioral program while still taking something. Any change to a prescription is a decision for you and your prescriber, and this is not the place to make it.

Shorter than most people use them. The European guideline puts short-term use at four weeks or less, and the ACP notes that FDA approval for these medications is for four to five weeks of use.

The 2026 AASM guideline suggests combination treatment over medication alone, but suggests against combination over CBT-I alone. Both are conditional recommendations on low certainty evidence. Read as: CBT-I alone is at least as good as doing both.

There is trial evidence that it improves the odds when it is paired with a taper supervised by a prescriber. In one randomized trial, 85% of people doing a supervised taper plus CBT were medication-free at the end, against 48% for taper alone. Never adjust or stop a sleep medication on your own.

More in CBT-I science

Related posts