Does CBT-I work for menopause insomnia? What the trials show

Yes, in the trials we found. CBT-I (cognitive behavioural therapy for insomnia, a structured treatment that changes the habits and thinking that keep insomnia going) improved sleep in women going through menopause, including women who were still having hot flashes. In a pooled analysis of four trials, with each treatment measured against its own control, CBT-I produced the largest improvement in insomnia scores of the seven treatments compared. ^[1][2] This page covers the trials; the wider picture is in perimenopause insomnia.

What the research says

In one trial, 106 women around and after menopause with moderate insomnia and at least two hot flashes a day had six telephone sessions of CBT-I or menopause education. Insomnia scores fell 9.9 points with CBT-I against 4.7 with education, a difference of 5.2 points. Hot flashes were no fewer, but interfered less with life after CBT-I, still at 24 weeks. ^[1] Pooling four trials in 546 women, CBT-I's 5.2-point drop was the largest of seven treatments; exercise, venlafaxine, yoga, estradiol and escitalopram did far less and omega-3 nothing, and the authors backed CBT-I as the first treatment for healthy midlife women with insomnia and moderately bothersome hot flashes. ^[2] In 150 postmenopausal women, CBT-I and sleep restriction alone both improved fatigue, energy, sleepiness and work function for six months; only CBT-I improved emotional well-being. ^[3] Across 11 trials and 973 women, CBT-I improved sleep quality and cut insomnia severity by about 4.5 points, in person or remotely. ^[4] In plain terms, CBT-I helped these women sleep while hot flashes carried on.

The studies behind this

• MsFLASH trial (McCurry, 2016): 106 perimenopausal and postmenopausal women aged 40 to 65, all with moderate insomnia symptoms and at least two hot flashes a day. They were randomly assigned to six telephone sessions over eight weeks of either CBT-I or a menopause-education control. At 8 weeks, insomnia scores had fallen 9.9 points with CBT-I and 4.7 points with education, a difference of 5.2 points (95% confidence interval 3.3 to 6.1; p < .001). Diary-reported time to fall asleep, time awake and sleep efficiency also improved more with CBT-I. ^[1]
• Hot flashes in that trial: the number of hot flashes per day did not differ between groups. How much hot flashes interfered with life dropped more with CBT-I, and that difference was still there at 24 weeks. ^[1]
• Pooled analysis of four MsFLASH trials (Guthrie, 2018): 546 women with insomnia symptoms and bothersome hot flashes, seven treatments compared with control. CBT-I lowered insomnia scores by 5.2 points relative to control, the largest effect. Exercise and venlafaxine (an SNRI, a type of antidepressant) each lowered scores by about two points. Yoga, oral estradiol (a form of hormone therapy) and escitalopram (an SSRI, another type of antidepressant) produced similarly small decreases. Omega-3 supplements did not improve insomnia symptoms. The authors concluded the results support CBT-I as the first treatment for healthy midlife women with insomnia and moderately bothersome hot flashes. ^[2]
• Detroit trial (Kalmbach, 2019): 150 postmenopausal women with chronic insomnia were assigned to CBT-I, sleep restriction alone (one part of CBT-I), or sleep education. CBT-I and sleep restriction both improved fatigue, energy, sleepiness and work function at the end of treatment and six months later. CBT-I also improved emotional well-being, which sleep restriction alone did not. ^[3]
• Meta-analysis (Moon, 2025): 11 randomised trials with 973 women. CBT-I improved sleep quality (standardised difference -1.01) and cut insomnia severity scores by about 4.5 points. It worked whether delivered in person or remotely (telephone and internet programs were included). ^[4]

Why was the education group included?

The control group in the MsFLASH trial still received sessions and attention, and its insomnia scores improved by 4.7 points. ^[1] This matters: some of any gain comes from being supported and informed. The trial's finding is that CBT-I improved scores by roughly twice as much. A study that compared CBT-I with doing nothing would flatter it.

What do guidelines say?

For chronic insomnia in general, the American College of Physicians and the European Insomnia Guideline both name CBT-I as the first treatment to try. ^[5][6] The North American Menopause Society's 2023 review also lists cognitive behavioural therapy among its evidence-supported non-hormonal options for hot flashes and night sweats. ^[7] Learn the basics in what is CBT-I.

What about hormone therapy and medicines?

The same menopause society says hormone therapy remains the most effective treatment for hot flashes and night sweats, with benefits and risks that depend on your age, time since menopause, dose and health. ^[8] Its 2023 statement recommends several non-hormonal options for hot flashes: cognitive behavioural therapy, clinical hypnosis, certain antidepressants (SSRIs and SNRIs), gabapentin, oxybutynin and fezolinetant. ^[7] In the pooled analysis, hormone therapy and the two antidepressants tested produced smaller changes in insomnia scores than CBT-I did. ^[2] This comparison covers insomnia scores only. It is not a statement about whether any treatment is right for you, and the trials measured insomnia, not hot flashes. Nothing here is advice to start, stop or change a medicine. Hormone therapy and prescription medicines are decisions for you and your doctor or pharmacist.

Where evidence is thinner

• Small trials. The anchor trial had 106 women at a single site. ^[1]
• Self-report. Most results rely on questionnaires and diaries. In a secondary analysis of the Detroit trial (113 women), neither CBT-I nor sleep restriction produced clinically meaningful change in sleep measured on a laboratory recording. ^[9]
• Some people respond less. Women with lower sleep efficiency on a laboratory recording before treatment were less likely to have their insomnia remit (37.8% versus 61.8%). ^[9]
• Hot flashes did not go away. Hot flash counts did not change with CBT-I in the anchor trial, though interference fell. ^[1]
• Short follow-up. The anchor trial followed women for 24 weeks. ^[1]
• Healthy midlife women. The pooled analysis's conclusion is framed for healthy midlife women, so it says less about people with other conditions. ^[2]

Does it make sense to do CBT-I during menopause?

If you are having trouble sleeping and are perimenopausal or postmenopausal, these trials say that CBT-I is a well-supported treatment to ask about, and that you do not have to wait for hot flashes to stop first.

Two cautions. First, the early weeks can be hard: the sleep-restriction part of CBT-I often brings more daytime tiredness before sleep improves, which we cover in sleep restriction making you more tired and CBT-I feels worse before it gets better. Second, the American Academy of Sleep Medicine's 2021 guideline says sleep restriction may be contraindicated for people in high-risk occupations such as driving or operating heavy machinery, people predisposed to mania or hypomania (as in bipolar disorder), and people with poorly controlled seizure disorders, because it involves short-term sleep loss. ^[10] If any of these applies to you, or you have another medical condition, talk with your doctor before starting. See what is CBT-I for more on safety.

Where does this leave you?

CBT-I works best when it fits your own numbers. Small Hours turns a nightly sleep diary into a plan built around your sleep. You can sign up at trysmallhours.com.

General education, not medical advice.