Depression and Sleep: Insomnia, Hypersomnia, CBT-I and the Bidirectional Evidence

Bottom line: Depression and sleep can reinforce each other, but “depression causes insomnia” is incomplete. Major depression can involve insomnia, early-morning waking, hypersomnia, irregular timing or non-restorative sleep. When chronic insomnia is present, CBT-I should be treated as a real intervention rather than waiting for mood treatment to fix sleep automatically. Randomized evidence shows CBT-I can improve both insomnia and depressive symptoms, but sleep treatment is an adjunct to—not a replacement for—depression-specific care.[1-9]

Depression does not have one sleep pattern

Sleep disturbance is built into the clinical picture of depression, but people with depression do not all sleep the same way.

Common patterns can include:

• difficulty falling asleep;
• repeated nighttime awakenings;
• early-morning waking;
• non-restorative sleep;
• sleeping much longer than usual;
• difficulty getting out of bed;
• irregular sleep-wake timing; and
• large weekday/weekend timing shifts.

That means a person who sleeps 11 hours and still feels exhausted is answering a different sleep question from someone who lies awake from 3 a.m. onward.

The intervention should follow the bottleneck.

Insomnia and depression are bidirectional

Insomnia can precede depression, occur during depression and persist after mood symptoms improve.[4,7-9]

Depression can also worsen sleep through:

• rumination;
• reduced daytime activity;
• circadian disruption;
• medication effects;
• altered reward and motivation;
• pain or other comorbidity; and
• changes in sleep opportunity.

A 2026 longitudinal study of 2,666 adults found that insomnia symptoms predicted later depressive symptoms and depressive symptoms predicted later insomnia symptoms, although the cross-lagged effects were generally small.[8]

The important conclusion is not that one direction “wins.”

It is that sleep can remain a meaningful treatment target even when depression is the more obvious diagnosis.

CBT-I has direct evidence in major depressive disorder with insomnia

A 2024 systematic review and meta-analysis identified 19 randomized trials involving 4,808 participants with major depressive disorder and insomnia.[1]

Compared with control conditions, CBT-I improved:

• depression response: OR 2.28 (95% CI 1.67–3.12); and
• insomnia remission: OR 3.57 (95% CI 2.48–5.14).[1]

The GRADE certainty for both outcomes was rated moderate.[1]

With a median control-group depression response rate of 17% at about eight weeks, the review estimated a CBT-I response rate of about 32% (95% CI 26%–39%).[1]

That is meaningful evidence that the antidepressant effect is not merely a change in one sleep item on a depression questionnaire.

But CBT-I is not a replacement for depression treatment

The same meta-analysis supports a careful conclusion:

Treating insomnia can improve depression outcomes when both conditions are present.

It does not establish:

• that CBT-I is sufficient treatment for every episode of major depression;
• that antidepressants or depression-focused psychotherapy are unnecessary;
• that all depressive symptoms are caused by insomnia; or
• that improving sleep automatically resolves suicide risk, anhedonia or severe functional impairment.

Sleep and depression should be treated as interacting clinical targets rather than competitors for the “real cause.”

The depression benefit can persist after CBT-I

A 2026 systematic review and meta-analysis examined the long-term effects of CBT-I on depressive symptoms in adults with insomnia.[2]

The review included 53 articles representing 13,608 individuals.

After outlier removal, CBT-I remained superior to control conditions for depressive symptoms at:

• 3 months: d = -0.35 (95% CI -0.49 to -0.21);
• 6 months: d = -0.32 (95% CI -0.59 to -0.05); and
• 12 months: d = -0.25 (95% CI -0.39 to -0.12).[2]

The effect becomes smaller over time, but it does not simply vanish immediately after treatment.

That supports a durable secondary mood benefit while preserving an important scale distinction: these are small-to-moderate depressive-symptom effects, not evidence that insomnia treatment substitutes for full depression care.

Digital CBT-I also has supportive evidence

A 2024 meta-analysis of seven articles involving 1,864 participants with insomnia and depression found digital CBT-I improved insomnia severity and depressive symptoms relative to control conditions.[6]

A newer 2025 meta-analysis of 16 randomized trials likewise concluded that unguided digital CBT-I improves depressive symptoms in adults with comorbid insomnia and depression.[5]

Digital delivery therefore has a meaningful evidence base, but “digital” is not one standardized intervention.

Programs vary in:

• therapist guidance;
• number of modules;
• adherence support;
• sleep-restriction implementation;
• depression severity; and
• whether participants have diagnosed MDD or elevated symptom scores.

Do not treat every sleep app as equivalent to validated digital CBT-I.

Can treating insomnia prevent depression?

This is where the evidence gets especially interesting—and especially easy to overstate.

A randomized clinical trial enrolled 291 adults age 60 or older with insomnia disorder who did not have major depression at enrollment.[3]

Participants received CBT-I or sleep education therapy.

Over follow-up, incident or recurrent major depression occurred in:

• 12.2% of the CBT-I group; and
• 25.9% of the sleep-education group.[3]

The hazard ratio was 0.51 (95% CI 0.29–0.88).[3]

That is a substantial randomized prevention signal.

But the population matters enormously.

The defensible statement is:

In older adults with insomnia, one randomized trial found CBT-I reduced incident or recurrent major depression compared with sleep education.

The indefensible version is:

“CBT-I prevents depression in everyone.”

The prevention literature is stronger for depression than anxiety—but still developing

A 2024 systematic review of CBT-I for prevention and early intervention retained 11 studies.[4]

Seven randomized trials examined depressive outcomes, compared with only one anxiety-prevention trial.[4]

The pooled preventive/early-intervention effect on depressive symptoms was approximately -0.52, while the pooled insomnia effect was approximately -0.81.[4]

The authors concluded that the prevention/early-intervention evidence is most supportive for mood disorders, while also noting heterogeneity.[4]

A separate 2023 systematic review found only six studies meeting criteria for insomnia-treatment prevention of depression, reinforcing that the evidence base is promising rather than limitless.[9]

Insomnia remission may matter more than simply receiving CBT-I

In the older-adult prevention trial, sustained remission of insomnia was associated with a particularly low depression risk.[3]

Participants in the CBT-I group who achieved sustained insomnia remission had an estimated 82.6% lower likelihood of depression compared with sleep-education participants without sustained remission (HR 0.17, 95% CI 0.04–0.73).[3]

That finding is clinically provocative, but it should not be interpreted as a simple causal mediation proof.

People who achieve sustained remission can differ from non-remitters in many ways.

The strongest randomized causal comparison remains CBT-I versus sleep education.

What if depression causes hypersomnia instead of insomnia?

CBT-I evidence is primarily about insomnia.

It should not be automatically repurposed for someone whose main symptom is excessive sleep duration or overwhelming sleepiness.

Long sleep can reflect:

• depression-associated hypersomnia;
• insufficient sleep recovery;
• medication effects;
• circadian delay;
• sleep apnea;
• central disorders of hypersomnolence; or
• spending extended time in bed without actually sleeping continuously.

This distinction matters because “I sleep 12 hours” and “I am in bed for 12 hours” are not necessarily the same physiology.

Early-morning waking is not uniquely diagnostic of depression

Waking hours before the intended time can occur in depression, but it can also occur with:

• insomnia disorder;
• circadian phase advance;
• menopause-related symptoms;
• pain;
• alcohol effects;
• environmental disturbance; and
• sleep apnea or other sleep disorders.

A sleep pattern can support a clinical formulation without functioning as a standalone diagnosis.

Antidepressants can change sleep—but not all in the same direction

Medication effects complicate the depression-sleep relationship.

Some antidepressant regimens may feel sedating; others can feel activating. Individual drugs can also alter REM sleep, sleep continuity or daytime alertness in different ways.

That means a person who develops insomnia or excessive sleepiness after a medication change should not assume the symptom reveals what the underlying depression is doing biologically.

Medication timing, dose, interactions and adverse effects belong in clinical review rather than DIY sleep-stack escalation.

A sedating feeling is not the same thing as treating insomnia

This principle matters throughout the sleep cluster.

A medication or substance can make someone drowsy without producing the durable behavioral and cognitive changes targeted by CBT-I.

Likewise, a sleep medication may improve sleep without directly treating core depression symptoms.

The question is always:

• what outcome changed;
• how large was the change;
• compared with what;
• for how long; and
• in what population?

Subjective sleep and objective sleep remain different outcomes

Depression can amplify fatigue, negative appraisal and perceptions of non-restorative sleep.

That does not make subjective sleep unimportant.

It means that:

• sleep diaries;
• insomnia questionnaires;
• actigraphy;
• wearables; and
• polysomnography

answer different questions.

See Subjective vs Objective Sleep for why those measurements should not be treated as interchangeable.

Depression can coexist with another sleep disorder

Do not stop the differential diagnosis at “depression.”

Sleep apnea

Snoring, witnessed pauses, choking, morning headaches or marked daytime sleepiness can justify evaluation for obstructive sleep apnea even when depression is already diagnosed.

Restless legs

An evening urge to move the legs belongs in a different pathway from depressive rumination.

Circadian delay

A person who sleeps normally on a later schedule may have a clock-timing problem layered on top of depression.

Insufficient sleep

Someone can be depressed and still simply not have enough sleep opportunity.

See Insomnia vs Sleep Deprivation.

Sleep improvement can help mood without explaining every mood symptom

The 2026 mechanistic review describes several plausible pathways through which CBT-I-related sleep improvement could influence depression, including emotional regulation, cognition, stress physiology, inflammation and behavior.[7]

These mechanisms are plausible and increasingly studied.

But mechanistic plausibility is not permission to say:

“Fix your sleep and your depression will go away.”

That is precisely the kind of overclaim the randomized evidence does not support.

A practical evidence-first decision tree

If chronic insomnia and depression coexist

Treat both as real clinical targets. CBT-I has direct evidence for insomnia and meaningful evidence for depressive-symptom improvement.[1,2]

If depression improves but insomnia remains

Do not assume the insomnia must disappear eventually. Persistent insomnia can become independently maintained and deserves treatment.

If insomnia improves but depression remains severe

Do not keep escalating sleep interventions as though sleep were the only remaining mechanism. Depression-specific treatment still matters.

If the main problem is excessive sleep or extreme daytime sleepiness

Reassess the phenotype rather than forcing an insomnia treatment model onto it.

If breathing symptoms, restless legs or circadian delay are present

Investigate those pathways too.

What the evidence does not show

Current evidence does not justify saying that:

• insomnia is the sole cause of depression;
• every person with depression has insomnia;
• CBT-I replaces antidepressants or depression-focused psychotherapy;
• CBT-I prevents depression in every age group and population;
• a sedating medication automatically treats insomnia well;
• sleeping longer proves depression-associated hypersomnia; or
• improved sleep eliminates the need to assess severe depression or suicide risk.

Evidence verdict

Evidence grade: Moderate.

The evidence that CBT-I improves insomnia when depression is present is strong, and the evidence for a secondary improvement in depressive symptoms is now supported by large meta-analyses.[1,2]

The prevention evidence is promising and includes an important randomized trial in older adults, but population-specific limits remain.[3,4,9]

The cleanest conclusion is:

Treating chronic insomnia can improve depression outcomes and may reduce future depression risk in selected populations, but insomnia treatment should be integrated with—not substituted for—depression-specific care.