Teen and Adolescent Sleep: Circadian Delay, School Start Times, Insomnia and CBT-I

Bottom line: Teen sleep is easy to moralize and easy to oversimplify. Puberty shifts circadian timing later on average, while school and family schedules can still demand very early waking. That mismatch can create chronic sleep restriction even when a teenager is capable of sleeping normally on a later schedule. But biology does not explain every problem: adolescents can also have insomnia, delayed sleep-wake phase disorder, sleep apnea, restless legs, mood disorders, medication effects, substance-related sleep disruption, or simply too little sleep opportunity. The intervention should match the bottleneck.[1-10]

Why teenagers tend to sleep later

Adolescence is accompanied by changes in both circadian timing and sleep regulation.

A later preferred sleep-wake schedule is common during the teen years. Recent research also shows developmental changes in melatonin physiology across puberty.[6]

That does not mean every teenager has a circadian disorder.

It means that an 11 PM bedtime and a 6 AM required wake time may collide with biology more strongly in adolescence than the same schedule would in many younger children.

The useful distinction is:

• late biological preference;
• insufficient sleep opportunity;
• behaviorally delayed bedtime;
• insomnia despite adequate opportunity; and
• delayed sleep-wake phase disorder (DSWPD).

Those can look similar at 6:30 in the morning but require different solutions.

Early school schedules can turn a timing preference into chronic sleep loss

School start time is not just a convenience variable.

A 2024 systematic review and meta-analysis found that later school start times were generally associated with longer sleep duration and lower social jet lag in adolescents.[1]

Earlier literature points in the same direction. A large meta-analysis of school start times and youth outcomes found later start times were associated with longer sleep duration and better overall developmental outcomes, with less negative mood in pooled analyses.[10]

These studies do not prove that changing a bell schedule fixes every adolescent sleep problem.

A teenager can still stay up later when school starts later. Homework, employment, sports, screens, family schedules, caffeine, mental health and social demands still matter.

But the evidence does undermine a simplistic claim that early school schedules have no meaningful relationship to teen sleep.

Social jet lag is a useful concept

Social jet lag describes the mismatch between biological or preferred sleep timing and required social timing.

A student might sleep approximately:

• midnight to 6:30 AM on school nights; and
• 1:30 AM to 10:00 AM on free days.

The large shift in sleep midpoint is not identical to flying across time zones, but it reflects a recurring mismatch between the body's timing and the imposed schedule.

Later school start-time evidence suggests that reducing this mismatch can increase sleep opportunity and reduce social jet lag.[1]

See Sleep Regularity and Weekend Catch-Up Sleep for the broader timing tradeoffs.

A night owl is not automatically a sleep disorder

This is one of the most important distinctions in adolescent sleep.

A teenager who naturally prefers a late schedule but:

• falls asleep readily when allowed to go to bed late;
• sleeps continuously;
• gets adequate total sleep; and
• functions well on that schedule

may simply have a late chronotype.

Delayed sleep-wake phase disorder is a stronger claim. DSWPD involves a persistent delay that creates clinically meaningful difficulty meeting required sleep-wake times and produces impairment.[7]

The site already separates these questions in:

• Night Owl vs Delayed Sleep Phase Disorder; and
• Delayed Sleep Phase vs Insomnia.

Insomnia can coexist with circadian delay

It is also possible to have both.

A teen may prefer a late biological schedule and develop conditioned arousal, worry about sleep, long wake periods in bed, or irregular sleep behaviors that sustain insomnia.

That is why “they are just a night owl” can be as incomplete as “they need stronger melatonin.”

The treatment question starts with identifying whether the main problem is:

• clock timing;
• insufficient opportunity;
• insomnia;
• another sleep disorder; or
• several at once.

CBT-I works in adolescents

Recent randomized-trial syntheses make this one of the strongest actionable parts of the evidence base.

A 2025 systematic review and meta-analysis included four randomized trials with 527 participants in the quantitative analysis. CBT-I improved subjective sleep quality with a Hedges' g of 0.40 and insomnia symptoms with a Hedges' g of 1.04.[2]

A separate 2024 meta-analysis included eight randomized trials and 599 adolescents. At post-treatment, CBT-I significantly improved:

• insomnia severity: SMD -1.06;
• sleep-onset latency: SMD -0.99;
• total sleep time: SMD +0.50; and
• sleep efficiency: SMD +0.57.[3]

Insomnia improvement also remained significant at follow-up in that synthesis.[3]

The exact estimates differ because meta-analyses use different inclusion rules, measures and statistical models. The defensible conclusion is not that one precise effect size applies to every teen; it is that CBT-I has replicated randomized-trial support in adolescent insomnia.

Digital CBT-I is becoming especially relevant

A 2026 network meta-analysis evaluated 22 randomized trials across multiple adolescent CBT-I delivery formats.[4]

Web-based CBT-I ranked highly and, compared with usual care in the network, was associated with approximately:

• 33 minutes more total sleep time;
• 23 minutes shorter sleep-onset latency;
• 7 percentage points higher sleep efficiency; and
• 5 points lower insomnia severity.[4]

No significant difference was found for wake after sleep onset in that analysis.[4]

Network rankings should be interpreted cautiously. They combine direct and indirect comparisons and are sensitive to the studies available in the network.

Still, the result matters strategically: access barriers do not automatically mean CBT-I has to be delivered only in a specialist clinic.

School-based CBT-I is promising, but implementation matters

A 2025 systematic review and meta-analysis specifically evaluated CBT-I delivered in school settings.[5]

That research matters because school delivery can reach adolescents who would never attend a sleep clinic.

But “school-based” is a delivery setting, not one standardized intervention. Programs can differ in intensity, therapist involvement, digital content, adherence and whether participants have diagnosed insomnia versus general poor sleep.

So positive school-based evidence should not be converted into a claim that a one-off classroom sleep lesson is equivalent to full CBT-I.

Melatonin is not just a sedative question

Melatonin is commonly framed as “something that makes you sleepy.”

For circadian delay, the more important concept is phase shifting.

The 2015 AASM circadian-rhythm guideline endorsed strategically timed melatonin for selected circadian disorders and light therapy with behavioral interventions for children/adolescents with DSWPD, while also emphasizing that evidence quality and population matter.[9]

More recent adolescent DSWPD reviews continue to describe melatonin and light as components of a multimodal approach.[7]

The guardrail is crucial:

The clock time of melatonin can change what it does.

That is why this page should not give one universal “teen melatonin dose” or one bedtime cutoff.

See Melatonin Timing vs Dose for the timing framework.

Puberty changes melatonin physiology, but that does not create a supplement deficiency diagnosis

A 2025 systematic review and meta-analysis found that endogenous melatonin secretion changes substantially across puberty.[6]

That is interesting physiology.

It does not mean that a normal adolescent with later sleep timing necessarily has a melatonin deficiency that should be corrected with a supplement.

Developmental physiology and treatment efficacy are different evidence questions.

This is the same evidence-discipline rule used across the site: a plausible mechanism cannot substitute for an intervention trial.

Bright light can help timing—but timing cuts both ways

Light is one of the strongest signals to the circadian system.

Morning or appropriately timed bright light can help advance a delayed sleep schedule in selected adolescents, while strong late-evening light can push timing in the opposite direction.[7,9]

That does not justify a universal protocol such as “use a light box at exactly 7 AM.”

The phase response to light depends on the person's current circadian phase, sleep schedule and treatment goal.

See Morning Light and Sleep Timing for the deeper circadian explanation.

Screens are more than blue light

Teen screen use is often reduced to a wavelength story.

Blue-enriched light can affect circadian signaling, but bedtime screens can also delay sleep by:

• replacing sleep time;
• increasing cognitive or emotional arousal;
• creating social pressure to stay available;
• exposing the user to unpredictable content; and
• making bedtime drift later night after night.

So “turn on a blue-light filter” does not automatically solve a screen-driven sleep problem.

See Blue Light, Screens and Sleep.

Mood and circadian delay can overlap

A 2025 systematic review and meta-analysis found an association between DSWPD and depression among young people.[8]

That does not establish a simple one-way causal chain.

Depression can disrupt routines, activity, light exposure and sleep timing. Circadian misalignment and chronic sleep loss may also worsen mood vulnerability. Shared risk factors can contribute to both.

The correct conclusion is that significant circadian delay plus mood symptoms deserves more attention—not that every late bedtime causes depression.

Weekend catch-up sleep is not a complete solution

Sleeping longer on weekends can partially recover acute sleep debt.

But if a teen repeatedly shifts several hours later on Friday and Saturday, the recovery can come with more social jet lag and a harder Sunday-night/Monday-morning transition.

This creates a genuine tradeoff:

• recovery from weekday sleep loss can help;
• large schedule shifts can reinforce circadian misalignment; and
• neither solves an early weekday schedule by itself.

That is why “never sleep in” and “sleep until noon every weekend” are both overly simplistic rules.

Caffeine can hide the problem without fixing it

Caffeine can temporarily improve alertness after insufficient sleep.

But late or high-dose caffeine can also delay sleep and reduce sleep quality, creating a self-reinforcing loop:

1. insufficient sleep;
2. more caffeine;
3. later or worse sleep;
4. greater morning sleepiness;
5. more caffeine.

See Caffeine and Sleep Timing for the dose × timing evidence.

What the evidence does not support

Current evidence does not justify claims that:

• teenagers are tired only because they are lazy;
• every adolescent night owl has DSWPD;
• every late bedtime is biologically predetermined;
• later school starts cure insomnia;
• sleep hygiene alone is equivalent to CBT-I;
• melatonin is simply a stronger sleep switch;
• one universal melatonin dose or clock time is correct for every adolescent;
• a blue-light filter neutralizes all bedtime screen effects; or
• an association between circadian delay and depression proves that one directly causes the other.

A better teen-sleep decision tree

When an adolescent is chronically exhausted or cannot sleep at the desired time, ask:

1. How much actual sleep opportunity exists on school nights?
2. Does sleep become normal when the teen is allowed a later schedule?
3. Is the main problem falling asleep, waking up, or getting out of bed at the required time?
4. How large is the weekday-weekend timing shift?
5. Are caffeine, nicotine, cannabis, medications or late screens contributing?
6. Are there insomnia symptoms despite adequate opportunity?
7. Are there signs of apnea, restless legs, depression, anxiety or another condition?

The answer determines whether the next move is more sleep opportunity, circadian timing work, CBT-I, another sleep-disorder evaluation, or a combination.

Bottom line

Adolescent sleep is a timing-and-opportunity problem for many teens, but not for all of them.

Pubertal biology tends to push sleep later. Early school schedules can compress sleep opportunity. Later start times generally increase sleep and reduce social jet lag. Adolescents can also develop genuine insomnia, and recent randomized-trial meta-analyses show CBT-I can meaningfully improve it.[1-5]

The strongest editorial principle is:

Do not confuse a late biological clock, insufficient sleep opportunity and insomnia. They can produce the same exhausted teenager while requiring different interventions.

Related reading

• Night Owl vs Delayed Sleep Phase Disorder
• Delayed Sleep Phase vs Insomnia
• Melatonin Timing vs Dose
• Morning Light and Sleep Timing
• Blue Light, Screens and Sleep
• Weekend Catch-Up Sleep
• Caffeine and Sleep Timing
• CBT-I vs Sleep Supplements