Noon, Rest from Work by Vincent van Gogh (wait, the top therapy for bipolar sleep warns us not to sleep in the day)
A modified form of cognitive behavioral therapy rises to the top for insomnia in bipolar disorder
STUDY: Baldini V et al, Bipolar Disorders 2026
STUDY TYPE: Systematic review (10 randomized controlled trials, 1 observational study)
FUNDING: Independent
Background
In bipolar disorder, sleep problems lead to relapse, worse symptoms, and even raise the suicide risk. This review weighed the evidence on what helps, from melatonin and sedative-hypnotics to behavioral and light therapies.
The Study
- 11 studies (10 randomized trials, 1 observational) from 2015-2025, across the U.S., Europe, and Asia.
- Participants had bipolar I or II, mostly euthymic; sample sizes ranged from 19 to 124.
- Compared pharmacologic options (melatonin, suvorexant, antidepressants) against non-pharmacologic ones (CBT for insomnia adapted for bipolar disorder, bright light therapy, blue-blocking glasses, the Benson relaxation technique).
Results
CBT for insomnia (CBTI-BD) was the winner here. This approach starts with standard CBT-insomnia (available through apps or in-person therapy), and adds a few modifications for bipolar, such as
- Don’t restrict time in bed to less than 6 hours per night
- Focus on daytime activity, encouraging key activities at regular times (eg work, chores, socializing, exercising, meals)
- Use dim lights at night (or dark therapy), particularly if manic symptoms arise
- Address the tendency to be “activity junkies” at night, and many people with bipolar are night owls.
CBTI-BD was tested in three trials, where it improved sleep efficiency, latency, and total sleep time, along with fewer depressive symptoms. In one trial of euthymic (stable) bipolar with insomnia, CBTI-BD improved sleep at first and had a delayed effect on mood, leading to 8-fold reduction in depressed days 4-6 months later (compared to another effective therapy, group psychoeducation). CBTI-BD did not, however, improve manic symptoms.
The Benson relaxation technique improved sleep quality and irritability. Bright light therapy was a mixed bag, raising the depression remission rate in one trial (68% versus 22% with placebo) but showing only modest sleep gains in others. In studies of non-bipolar patients, early morning lightbox improved insomnia.
Melatonin improved circadian alignment in patients with a delayed sleep-wake phase. Suvorexant modestly increased total sleep time without moving subjective sleep quality or mood. Antidepressant use tracked with poorer sleep and more impulsivity in one study. Though not covered in this review, dark therapy improved sleep regularity in a trial of mania, and it separately improved insomnia in three trials of non-bipolar insomnia.
Surprisingly, there are no controlled trials of benzos or z-hypnotics for sleep (or mania) in bipolar disorder, and these could have disinhibiting effects in actively manic patients.
Side Effects
Non-pharmacologic approaches were the safest, with only mild, self-limited effects such as brief early sleepiness with CBTI-BD. Melatonin and suvorexant were well tolerated. No trial reported a treatment-related switch into mania, though sedative-hypnotics carry separate, well-known misuse concerns.
Limitations
Small trials, from 19 to 124 patients; they used different comparators, measures, and follow-up lengths, so pooling effect sizes wasn’t possible. Each intervention was tested in only one mood state, so it’s still unclear whether clinical phase changes the response. Several trials had blinding or reporting concerns, and the single observational study carried a serious risk of bias.
Practice Implications
- I start with CBTI-BD, with blue-light blockers 1-2 hours before bed, and a dawn simulator to encourage regular wake times.
- Learn more about CBT-insomnia and lifestyle therapies for mood in The Depression and Bipolar Workbook.
—Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
