Does school-based CBT-I improve adolescent sleep duration and sleep quality?

What actually changes after a school-based CBT-I program?

The clearest signal comes from a meta-analysis pooling eight studies of adolescents aged 10–19 with insomnia symptoms, totaling 323 participants across four countries [4][5]. When researchers compared participants' sleep before and after the intervention, self-reported total sleep time rose by about 24 minutes, the time taken to fall asleep dropped by about 8 minutes, and sleep quality improved by a small-to-moderate amount [4][5]. These are within-group changes, meaning each person was compared to their own baseline rather than to an untreated control group, which makes the numbers more encouraging than definitive.

A separate workshop-based study in 100 adolescents with self-reported sleep problems found a similar pattern: participants reported better sleep efficiency and subjective sleep quality, along with improved daytime vigilance, memory, and anxiety, after four weekly group sessions combining CBT-I with CBT for delayed sleep phase syndrome [3]. That study also measured objective outcomes, but its actigraphy and melatonin data had not yet been analyzed at the time of reporting [3].

Why do self-reports improve while device measurements do not?

The gap between what adolescents say and what devices record is the central puzzle. In the meta-analysis, objective total sleep time rose by about 21 minutes, but the result was not statistically significant, and objective sleep onset latency and sleep efficiency barely moved [4][5]. One plausible explanation is that CBT-I changes how adolescents think about and experience their sleep—reducing worry and bedtime arousal—even when the physiological clock shifts less than they perceive. Another is that the objective measurements came from small samples, so real but modest effects could be missed. The review authors themselves conclude that improvements were observed only for subjective outcomes and describe the evidence base as limited [4][5].

This does not mean the subjective gains are imaginary. Feeling more rested and falling asleep more easily are meaningful outcomes for a teenager. But it does mean that claims about CBT-I literally extending sleep need to be stated carefully until more actigraphy data are available.

What does a school-based CBT-I program actually involve?

The workshop format studied in adolescents was compact: four weekly group sessions covering psychoeducation, behavioral strategies, and personalized sleep hygiene advice, delivered in a school-compatible way [3]. That is far lighter than clinical CBT-I, which typically runs over several weeks with individual sessions. The meta-analysis similarly focused on interventions using evidence-based CBT-I principles delivered in school settings, rather than full clinical protocols [4][5].

A broader systematic review of school-based sleep interventions found promising results across a range of non-pharmacological approaches, but also wide variation in how sleep was measured and inconsistent effects on mental health and academic performance [2]. That review did not isolate CBT-I specifically, so it supports the general idea that schools are a workable setting for sleep programs without confirming that CBT-I is the active ingredient.

How much should we trust these results?

The evidence base is small. Eight studies and 323 adolescents is enough to detect a moderate effect but not enough to settle questions about who benefits most, how long gains last, or whether results hold across different school cultures [4][5]. The review authors explicitly note the limited evidence and call for methodological and implementation guidance [4][5]. The broader school-intervention review adds that assessment tools vary widely and that effects differ across minority groups, pointing to a need for culturally tailored approaches [2].

A digital behavioral intervention that included CBT-I elements but was delivered outside school—through short videos—also produced earlier bedtimes and longer self-reported sleep in a small pilot of 14 adolescents [1]. That study is not school-based, so it does not answer the original question directly, but it hints that the behavioral components of CBT-I may travel well across delivery formats. Whether a school setting adds reach, accountability, or nothing beyond convenience remains an open question.

Sources used in this answer

Related research questions