ADVERTISEMENT

Issue Navigator

Volume 14 No. 11
Earn CME
Accepted Papers





Review Articles

Cognitive and Behavioral Interventions to Improve Sleep in School-Age Children and Adolescents: A Systematic Review and Meta-Analysis

Lie Åslund, MSc1,4; Filip Arnberg, PhD1,2; Marie Kanstrup, PhD1,3; Mats Lekander1,4
1Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden; 2Department of Neuroscience, Psychiatry, Uppsala University, Uppsala, Sweden; 3Functional Area Medical Psychology, Functional Unit Behavior Medicine, Karolinska University Hospital, Stockholm, Sweden; 4Stress Research Institute, Stockholm University, Stockholm, Sweden

ABSTRACT

Study Objectives:

Sleep problems are common in children and adolescents and can aggravate comorbid disorders. This meta-analysis examined the effect of cognitive and behavioral sleep interventions (with four or more treatment sessions) from randomized controlled trials on school-age children and adolescents.

Methods:

In a systematic literature search, six randomized controlled trials were identified (n = 528; mean age = 14.6 years; female = 63%) that reported total sleep time (TST), sleep onset latency (SOL), wake after sleep onset, and daytime sleepiness from ratings and actigraphy.

Results:

After intervention, no effect was seen on self-reported TST, but when measured with actigraphy, an effect favoring the intervention group was observed (+11.47 minutes, P = .05). SOL decreased in the intervention group compared to the control group after intervention as measured by both sleep diaries (−9.31 minutes, P = .007) and actigraphy (−19.48 minutes, P < .0001). Effect sizes ranged from small to large. No effect was found for wake after sleep onset or daytime sleepiness. Short-term (4 to 8 weeks) follow-up data from four studies indicated maintained positive effects on SOL: sleep diaries −15.85 minutes (P = .01) and actigraphy −23.67 minutes (P < .0001). At follow-up, the effects on wake after sleep onset from ratings (−14.41 minutes, P = .001) and actigraphy (−7.54 minutes, P = .01) were significant, favoring the intervention group (moderate to large effect sizes). No effect on TST was indicated.

Conclusions:

Cognitive and behavioral sleep interventions are indicated to improve sleep in school-age children and adolescents. However, because treatment protocols were heterogeneous and risk of bias high, results should be interpreted with caution. Large and rigorous trials are needed.

Citation:

Åslund L, Arnberg F, Kanstrup M, Lekander M. Cognitive and behavioral interventions to improve sleep in school-age children and adolescents: a systematic review and meta-analysis. J Clin Sleep Med. 2018;14(11):1937–1947.


INTRODUCTION

Sleep disorders are common in children and adolescents. The most common disorders relate to difficulties initiating and/or maintaining sleep, usually referred to as insomnia. Insomnia is defined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) as difficulties falling asleep, staying asleep, and/or suffering from early awakenings for at least 3 nights/wk over at least 3 months with associated clinically significant functional impairment.1 In school-age children (6 to 11 years), the prevalence of symptoms of disordered sleep range between 10% in the general population and as high as 75% for children with neurodevelopmental disorders.2,3 A longitudinal study on the lifetime prevalence of adolescent insomnia reported the median age of onset as 11 years, with prolonged sleep onset as the most common symptom.4 In adolescents (12 to 20 years), approximately 10% meet criteria for insomnia diagnoses, and more than 33% report “at least some” insomnia symptoms.5 However, patterns of sleep change as a part of normal development, with a decline in slow wave sleep observed in adolescence that seems associated with underlying changes in the brain structure such as a decline in synaptic density.6 In this age group, changes in sleep electroencephalography (EEG) have been linked to altered sleep schedules and sleepiness.7 Although such changes are part of normal development, social and academic demands can contribute to suboptimal sleep, with mood-related disturbances as possible consequences.6 Adolescents are at increased risk for disrupted circadian rhythm, including an extreme manifestation of rhythm disruption known as delayed sleep phase disorder (DSPD).8 The DSM-5 defines DSPD as normal sleep that is delayed in its timing with respect to the individual's sleep onset and rising times, leading to daytime sleepiness and loss of function. As many as 16% of adolescents and young adults may be afflicted with DSPD, and DSPD can have negative consequences for both mental health and academic performance.9 Developmental transition into adolescence can increase independence regarding sleep habits, often resulting in a reduction in parental control, later bedtimes, and insufficient sleep.

The long-term consequences of poor sleep are many, including increased risk for diabetes10 and cardiovascular mortality.11 Sleep loss also affects cognitive and affective functioning in all ages. In school-age children, shorter sleep duration and inconsistent sleep schedules are associated with behavior difficulties,12 a link that seems to persist into adolescence when several risk behaviors and health outcomes are associated with late bedtimes.13 For example, poor sleep resulting in sleepiness is associated with impaired executive functioning14 and an increase in irritability, inattention, and lack of motivation.15 Studies have shown a strong co-occurrence of sleep disorders and depression in this age group, as 77% of girls and 69% of boys aged 7 to 15 years suffering from depression also report symptoms of sleep disturbances.16 In adolescents, shorter sleep duration is not only associated with more acute illnesses (eg, common cold and flu)17 but also with negative effects on cognition and emotion. In this age group, experimental sleep restriction is associated not only with self-reported negative affective functioning but also with negative affective behavior in social contexts primed for peer conflict.18 Thus, poor sleep is related to both somatic and mental health issues. In the relationship between sleep and mood-related disabilities, anxiety disorders and depression are indicated as the most common cooccurring mental disorders.19 It seems that the period from late childhood to early adolescence is a particularly sensitive time as studies have shown that children in this age range have an increased risk of the development of sleep problems and anxiety.20 In this age group, as many as 90% of children with anxiety disorders also report sleep-related problems.21 School-age children with disordered sleep have an increased risk for onset of major depression, and sleep problems in 8-year-old children seem to predict depression 2 years later without evidence of the opposite scenario.22 In addition to sleep problems, more than 50% of adolescents with insomnia fulfill criteria for at least one comorbid psychiatric disorder.4 In a study on sleep and depression in this age group, 66% of individuals with insomnia also presented symptoms of depression.23 In adults, failure to treat insomnia is a risk factor for depression recurrence,24 further highlighting the need to establish well-supported treatment guidelines.

Several meta-analyses have found that cognitive behavioral therapy (CBT) for treatment of sleep disorders in adult populations has shown positive effects on sleep parameters.2528 Cognitive behavioral therapy for insomnia (CBT-I) refers to a standardized protocol that aims to improve the participant's sleep by targeting both cognitive and behavioral components of the disorder.29 CBT-I techniques for adults and adolescents include the promotion of behavioral change through the use of sleep restriction and stimulus control, changing negative thoughts about sleep through cognitive interventions and relaxation.30,31 The intervention can be delivered in multiple forms such as face-to-face individual meetings or in small groups, via an Internet platform, or as school-based sleep education programs. In younger children, cognitive and behavioral interventions for sleep disorders include the active participation of caregivers.32,33 These interventions include parental education and behavioral interventions that encourage parents to gradually restrict unwanted sleep behaviors (ie, ignoring negative behavior and gradually restricting parental presence at the child's sleep onset), establishing positive bedtime routines, using bedtime fading (ie, moving the child's bedtime to match the time of falling asleep), and scheduling night awakenings (ie, waking the child before his/her typical nocturnal awakenings to help with sleep consolidation).

There is a lack of knowledge about the effect of cognitive and behavioral sleep interventions in school-age children and adolescents as these issues have only recently been examined. In a 2014 systematic review and meta-analysis, Meltzer and Mindell33 found that interventions for pediatric insomnia for both controlled and within-subject studies yielded small to large effect sizes. However, Meltzer and Mindell only found weak evidence for older children and adolescents due to a lack of studies, and between-group effects could not be estimated for this age group. Recently, studies have suggested a positive effect of cognitive and behavioral interventions on adolescents regarding sleep26 and associated functions such as daytime sleepiness, negative daytime behavior, and emotional dysregulation. Blake et al. recently conducted a meta-analysis of cognitive behavioral sleep interventions in adolescents34 to address whether existing sleep interventions improve sleep in this age group. They found positive effects with several sleep parameters and measures of sleepiness and comorbid anxiety. Because the study mainly focused on within-group differences, studies with a noncontrolled design were included. Between-group effects were calculated but not interpreted due to large heterogeneity. Since these studies, new randomized controlled trial (RCT) studies have been published. Given the strengths of the RCT design, the evidence base would be strengthened by the examination of the efficacy of controlled studies on sleep of school-age children and adolescents. Therefore, this analysis aims to update and evaluate the current knowledge of the effect of cognitive and behavioral sleep interventions of four or more treatment sessions for school-age children and adolescents, using a controlled research design, on objective (actigraphy) and self-reported (sleep-wake diary and questionnaire) sleep outcomes. In addition, this study aims to investigate the effect of treatment on daytime sleepiness.

METHODS

Criteria for Inclusion and Exclusion

Studies were selected if they examined the effect of cognitive and/or behavioral interventions for sleep in school-age children and adolescents. Specifically, studies were included if they fulfilled the following criteria: (1) the participants' age range was between 6 and 20 years; (2) the study evaluated a cognitive and/or behavioral sleep intervention; (3) the study was published in a peer-reviewed journal in English; (4) was an RCT; (5) the intervention comprised a minimum of four sessions, corresponding to the minimum number of sessions found most effective in adult CBT-I35 (most programs were delivered in between four to eight sessions);29 and (6) the intervention was evaluated by objective (actigraphy) and/or self-reported (sleep-wake diary and questionnaire) measures.

Studies that combined cognitive behavioral sleep interventions with treatment modules for other disorders (eg, anxiety and depression) were not excluded as long as these modules targeted similar psychological mechanisms as the sleep interventions (eg, changing specific behaviors/thoughts regarding the disorder). Studies that included interventions that targeted other mechanisms were excluded (eg, aiming to affect the sleep cycle using bright light). Studies were included even if not all participants fulfilled criteria for a sleep disorder or if the participants had comorbid disorders. The selected articles were read in full and data were extracted by two independent reviewers for meta-analytic evaluation.

Search Strategy

A systematic literature search was conducted in 2014 and updated in October 2016 by two librarians at the Karolinska Institutet University Library in the following databases: Medline (Ovid), PsycInfo (Ovid), Web of Science Core Collection, Embase (embase.com), Cinahl (Ebsco), Cochrane Library (Wiley), SveMed+, and ClinicalTrials.gov. The MeSH terms identified for searching Medline (Ovid) were adapted in accordance to corresponding terms in Embase, PsychInfo, and Cinahl. Each search concept was also complemented with relevant free-text terms and the terms were, if appropriate, truncated and/ or combined with proximity operators. No restrictions in language, year, or publication type were applied. The complete search strategies are available in the supplemental material.

Measurement Domains

The main outcomes for sleep were selected from those recommended by Buysse et al.36 The outcomes combined both sleep and wake measures using self-reported (sleep-wake diary, questionnaire) and objective (actigraphy) measures of total sleep time (TST), sleep onset latency (SOL), and wake after sleep onset (WASO). The secondary outcome was daytime sleepiness (Pediatric Daytime Sleepiness Scale).37 When sleep-wake diaries were used, they were kept parallel to actigraphy by participants themselves with parental support when needed. Other studies used questionnaires to collect self-reported sleep data. Table 1 provides more detailed information on sleep outcome measures across studies.

Characteristics of included studies.

jcsm.14.11.1937.t01.jpg

table icon
Table 1

Characteristics of included studies.

(more ...)

Quality Ratings

To assess the risk of bias across the studies, we relied on standardized quality scoring criteria. Using the Cochrane Collaboration Risk of Bias Tool for RCTs, two study authors (LÅ & FA) independently assessed randomization bias, allocation bias, blinding bias, incomplete outcome data, and selective reporting bias.38

Data Analysis

Analyses were performed using the software program Review Manager.39 The effects of cognitive behavior interventions on sleep and sleepiness are reported in original units when meaningful (ie, hours or minutes) and as a standardized effect size (Cohen d).40 The effect was considered small if d < 0.2, moderate if 0.5, and large if 0.8.41 Between-group effect sizes were computed using random effects models. If studies did not report raw participant data, means and standard deviations were estimated. The results compared the difference between intervention and control groups at two times: post-intervention and short-term follow-up (< 8 weeks). As one study42 included two intervention groups and only one control group, one half of the control group was allocated to each of the comparisons with the intervention groups according to the Cochrane standards.38 The heterogeneity of the studies was assessed by the I2 statistic,43 which indicates what proportion of the variation in observed effects is due to the variation in true effects. The heterogeneity was considered low if I2 was 25%, moderate if 50% and high if 75%.43

RESULTS

Search Results

The search yielded 5,814 abstracts that after exclusion by title/ abstract resulted in 57 studies. Figure 1 shows the details of the selection procedure. Ultimately, six studies were included in the analysis.

Flow chart of inclusion procedure.

RCT = randomized controlled trial.

jcsm.14.11.1937a.jpg

jcsm.14.11.1937a.jpg
Figure 1

Flow chart of inclusion procedure.

(more ...)

Characteristics of Included Trials

Table 1 provides a summary of the six included studies. The studies were published in Australia (n = 4), the Netherlands (n = 1), and the United States (n = 1). In total, the studies included 528 participants at baseline, ranging in age from 7 to 20 years (mean age = 14.6, female 63%). Mean number of sessions was 6.2 (range 4 to 10). Three studies used wait-list comparisons. Two studies used active controls where the control group participated in organized sessions but not using CBT-I–specific techniques. One study used treatment-as-usual as comparison, which in this case meant attending regular psychology classes at school.

Characteristics of Included Interventions

Table 1 provides the treatment setting and duration for each study. Across the studies, six separate sleep intervention components were identified (Table 2). Of these six components, two can be described as educational (sleep education and sleep hygiene), two as promoting behavioral change (sleep restriction/ bedtime fading and stimulus control), one targeting changing thought content (cognitive therapy), and one using techniques for relaxation and acceptance (relaxation/mindfulness). These treatment modules are classic components in evidence-based CBT-I for adults.29 Two studies also included CBT treatment modules to address anxiety and depression.

Characteristics of included interventions.

jcsm.14.11.1937.t02.jpg

table icon
Table 2

Characteristics of included interventions.

(more ...)

Quality Ratings of Included Trials

Table 3 provides a summary of the quality ratings. All studies included a sufficient description of participant characteristics. The description of the intervention that was offered to participants in the intervention group was deemed satisfactory in all studies. Two studies were found to have an overall low risk of bias whereas the other four studies had at least one parameter with a high risk of bias. Not all studies reported treatment adherence. As the information that was relevant to quality rating was sometimes missing, the risk of bias could not always be fully assessed. To summarize, the risk of bias across these six studies was regarded as high.

Risk-of-bias assessment across studies.

jcsm.14.11.1937.t03.jpg

table icon
Table 3

Risk-of-bias assessment across studies.

(more ...)

Results of Meta-Analysis

Post-Intervention Effects

As seen in Figure 2, there was no group difference in self-reported TST. For the objective TST (Figure 2), the participants in the intervention group slept on average 11.49 minutes longer as compared to those in the control group (P = .05). Self-reported SOL (Figure 3) was significantly lower in the intervention group as the participants fell asleep 9.31 minutes faster than those in the control group (P = .007). Similarly, objective SOL (Figure 3) was shorter in the intervention group (mean = −19.48 minutes, P < .0001). There were no differences in self-reported or objective measures of WASO (Figure 4) or in daytime sleepiness (Figure 5). The heterogeneity of the post-intervention effects for the self-reported measures was low for SOL, moderate for TST, and high for WASO and daytime sleepiness. The heterogeneity was low for the objective measures.

Meta-analysis of the post-effect of the intervention on self-reported (top) and objective (bottom) total sleep time.

I2 for measures = 55% (self-reported) and 0% (objective). CI = confidence interval, SD = standard deviation.

jcsm.14.11.1937b.jpg

jcsm.14.11.1937b.jpg
Figure 2

Meta-analysis of the post-effect of the intervention on self-reported (top) and objective (bottom) total sleep time.

(more ...)

Meta-analysis of the post-effect of the intervention on self-reported (top) and objective (bottom) sleep onset latency.

I2 for measures = 20% (self-reported) and 15% (objective). CI = confidence interval, SD = standard deviation.

jcsm.14.11.1937c.jpg

jcsm.14.11.1937c.jpg
Figure 3

Meta-analysis of the post-effect of the intervention on self-reported (top) and objective (bottom) sleep onset latency.

(more ...)

Meta-analysis of the post-effect of the intervention on self-reported (top) and objective (bottom) wake after sleep onset.

I2 for measures = 72% (self-reported) and 3% (objective). CI = confidence interval, SD = standard deviation.

jcsm.14.11.1937d.jpg

jcsm.14.11.1937d.jpg
Figure 4

Meta-analysis of the post-effect of the intervention on self-reported (top) and objective (bottom) wake after sleep onset.

(more ...)

Meta-analysis of the post-effect of the intervention on daytime sleepiness.

I2 = 82%. CI = confidence interval, SD = standard deviation.

jcsm.14.11.1937e.jpg

jcsm.14.11.1937e.jpg
Figure 5

Meta-analysis of the post-effect of the intervention on daytime sleepiness.

(more ...)

Short-Term Follow-Up Intervention Effects

Because there was considerable variation in the time to follow-up, only short-term follow-up data (4 to 8 weeks post-intervention) from four of the studies were included in the meta-analysis. No differences were found between the groups on self-reported or objective measures of TST (Figure S1 in the supplemental material). A statistically significant effect was found on self-reported SOL (Figure 6) with participants in the intervention group falling asleep 15.85 minutes faster than those in the control group (P = .01). The effect of objective SOL was also significant (Figure 6) with a mean of −23.67 minutes for the intervention group compared to the control group (P < .0001). Likewise, a group effect of intervention on WASO was observed, both based on self-reported (mean = −14.41 minutes, P = .001) and objective (mean = −7.4 minutes, P = .01) measures (Figure 7). There was no group effect on daytime sleepiness at follow-up (Figure S2 in the supplemental material). The heterogeneity of the short-term follow-up intervention effects for the self-reported measures was low for WASO and moderate for SOL. The heterogeneity was low for the objective measures.

Meta-analysis of the short-term follow-up effect of the intervention on self-reported (top) and objective (bottom) sleep onset latency.

I2 for measures = 53% (self-reported) and 0% (objective). CI = confidence interval, SD = standard deviation.

jcsm.14.11.1937f.jpg

jcsm.14.11.1937f.jpg
Figure 6

Meta-analysis of the short-term follow-up effect of the intervention on self-reported (top) and objective (bottom) sleep onset latency.

(more ...)

Meta-analysis of the short-term follow-up effect of the intervention on self-reported (top) and objective (bottom) wake after sleep onset.

I2 for measures = 10% (self-reported) and 0% (objective). CI = confidence interval, SD = standard deviation.

jcsm.14.11.1937g.jpg

jcsm.14.11.1937g.jpg
Figure 7

Meta-analysis of the short-term follow-up effect of the intervention on self-reported (top) and objective (bottom) wake after sleep onset.

(more ...)

Sensitivity Analysis

To address the large variation in the included studies both regarding treatment components and age of the participants, we performed sensitivity analyses on posttreatment comparisons. Three analyses were performed: (1) excluding the two studies that included CBT for anxiety and depression44,45; (2) excluding the study with the youngest participants (mean age = 9.3 years)46; and (3) only including participants in whom sleep disorders were diagnosed as two studies reported results from subgroups with DSPD.47,48 No substantial differences were found between these and original analyses (data not shown).

DISCUSSION

This meta-analysis synthesizes the current literature on cognitive and behavioral interventions on insomnia and sleep difficulties for school-age children and adolescents on sleep and daytime fatigue. By including recently published RCTs and evaluating between-group comparisons, the findings provide an updated estimate of the treatment effect with less risk of bias as compared to previous meta-analyses33,34 of uncontrolled effect sizes. There were significant differences in TST and SOL at posttreatment favoring the sleep-specific cognitive behavioral interventions as compared to wait-list and active control groups. The effect sizes range from 0.21 to 0.81. The difference in improvement in wakefulness in bed parameters (objective and self-reported SOL) was nominally stronger (effect sizes 0.4 and 0.81) than the difference in sleep parameters (objective TST effect size 0.21). These results are generally consistent with previous findings in adults.25 However, the effects on WASO and daytime sleepiness were not significant at post-treatment. These findings differ from findings in adult populations where statistically significant effects post-intervention (with small to moderate effect sizes) have been found for these outcomes.25 Regarding adolescents, a within-sleep condition meta-analysis by Blake et al.34 suggested that CBT-I can result in statistically significant improvements in TST, SOL, and WASO after treatment. The current results extend their preliminary conclusions, corroborating the efficacy of cognitive and behavioral interventions on these outcomes. Between 4 and 8 weeks after treatment, the improvement was still evident for SOL but not for TST. However, the differences in improvement between groups in WASO was now significant both for self-reported (effect size: 0.54) and objective (effect size: 0.42) measures. Thus, the effect of cognitive and behavioral interventions for the sleep of school-age children and adolescents still seems to be retained for most measures, or is even larger at the short-term follow-up, in a manner similar to what has been suggested in treatment of insomnia in adults25,26 and in within-group analyses in adolescents.34 As noted by Koffel et al.25 regarding adults, a continued improvement in some sleep parameters could possibly be due to CBT-I techniques such as stimulus control and sleep restriction. These techniques aim to limit the participants' sleep window in order to increase the drive to sleep and thus reduce SOL and WASO. However, it should be noted that the positive effect on TST that was present for the intervention group posttreatment was not significant at follow-up. Continued research on treatment of sleep problems among school-age children and adolescents may shed light on the long-term stability on sleep parameters, with a clear need for longer follow-up periods.

It is also important to highlight the difference between statistical and clinical significance and to further investigate which effects both regarding types of measures and effect size that would make a participant perceive an intervention as useful. The key indicators of good sleep quality in adults, adolescents, and school-age children are similar (sleep efficiency > 85%, SOL < 30 minutes, WASO < 20 minutes and night awakenings < 2),49 but what does not indicate good sleep quality differs slightly between age groups. In adolescents, WASO > 50 minutes and 3 or more night awakenings do not indicate good sleep quality, as compared to WASO > 40 minutes and 4 or more night awakenings in school-age children and adults.49 Because guidelines for pediatric sleep are different by age group, age should be considered for a more sensitive evaluation of clinically meaningful change when evaluating treatment effects. To avoid insomnia remission that reflects an increased tolerance of sleep symptoms despite continued sleep disturbances and impairment,50 qualitative methods could be of use in future studies to explore possible reasons for the discrepancy between participants' perceptions of their sleep quality, presence of insomnia symptoms, and objective sleep parameters.

Regarding the differences in effect sizes between objective and self-reported data, the effect of objective measurements can be smaller than their self-reported counterparts.25 Four included studies used both objective and self-reported data, and the analysis revealed this pattern was not observed either at posttreatment or at follow-up. Self-reported measures showed small to moderate size effects, whereas the effect sizes for the objective measures were moderate to large. The reasons for this are unclear, although one may speculate that the inaccuracy in self-reported perceptions of sleep that is found in adult populations51 is also present in children. Therefore, objective positive treatment effects could thus be underestimated in sleep-wake diaries and questionnaires. As noted previously, child/adolescent perception of good sleep quality might not correspond to objective sleep measures and objectively noted improvements might need time to be experienced. However, the stability of this finding should be examined in further analyses. Interestingly, the heterogeneity of the effects was generally larger for self-reported than for objective outcome measures, pointing toward another possible origin for the differences between these types of measures, one that becomes noticeable only in meta-analyses. Additionally, the heterogeneity was generally lower in this study as compared to previous within-group analyses,34 indicating lesser variability in effect estimation when analyzing controlled studies.

Because we only included RCTs, there is a risk that we might have excluded studies that made an important contribution to understanding the effect of cognitive and behavioral interventions on the sleep of school-age children and adolescents. The scarcity of RCTs in this field precluded analyses of homogeneous subsets of trials and so the included studies were heterogeneous regarding participant characteristics, treatment components, treatment delivery, and treatment length. As we limited this meta-analysis to only include studies with four or more treatment sessions, these results do not reflect the information on the effect of very brief cognitive and behavioral interventions for school-age children and adolescents that might contribute to the understanding of treatment effects. Because no meta-analysis has specifically examined the effect of such brief interventions in this age group, it is unclear whether interventions of traditional or abbreviated length would be most efficacious for children and adolescents. The participants' age ranged from 7 to 20 years due to the postulation that CBT interventions offered to school-age children resemble those available to adolescents and adults as compared to interventions targeting infant and toddlers that primarily target mal-adaptive parent behaviors. It should be noted that most studies included adolescent participants and that the results presented here should be interpreted with caution regarding school-age children. The results may also be affected by developmental issues other than age, such as the ability to pursue an uncomfortable but goal-oriented behavior over time (eg, follow instructions for sleep restriction) or to use various cognitive strategies to solve problems and adapt more functional strategies. A difference in level of caregiver involvement could also have affected the results, where younger children (as opposed to older teenagers) would benefit from more support to achieve their treatment goals. Clearly, more research is needed to shed light on whether and how age moderates the treatment effects. It is also noted that a direct comparison of these treatment effects to the effect of CBT-I in adults has to be done with caution because of the possible differences in delivered treatment.

The participants reported levels of sleep dysfunctions ranging from normal sleep to severe sleep disorders such as primary insomnia and delayed sleep phase disorder. Although sensitivity analyses were performed to investigate possible influences of such factors, the results should be interpreted with caution because of study heterogeneity. Differences in treatment delivery format further contributed to the study of heterogeneity. However, CBT-I in adults seems to be effective irrespective of whether the mode of delivery is adult individual treatment, group therapy, telephone consultations, or bibliotherapy.52,53 When comparing the effect of adult individual treatment, group therapy, and telephone consultations, CBT-I was effective at improving sleep parameters in all groups with improvements remaining after 6 months.52 Research on group CBT-I has shown moderate to large effect sizes for several sleep parameters with effects retained over time.25 In studies of internet-delivered CBT-I, the effects were large for insomnia diagnosis, sleep parameters, and comorbid anxiety.54 Some questions regarding follow-up measures should also be highlighted. Because the time of follow-up varied, a reduced number of studies were used in this comparison. Therefore, the results should be interpreted with caution. Although effect sizes were moderate to large for some sleep parameters, it is unclear whether the effect sizes are retained in the long term. Studies with longer follow-up periods are needed. To ensure that positive results posttreatment are not lost over time, future studies should address aspects of participant motivation by investigating the long-term acceptability of the treatments.

Furthermore, the risk of bias across studies was considered high, partly as a consequence of difficulties with blinding in psychological treatment research. The risk of bias could be lowered in future studies by using active control participants who would receive an equivalent number of sessions with nontreatment-specific sleep content that does not promote behavioral change (eg, general information about sleep architecture and the believed function of sleep). To improve the evidence base for cognitive and behavioral sleep interventions for school-age children and adolescents, issues related to bias, transparency, and heterogeneity should be addressed in future studies. Half of the current studies42,44,45 were preregistered on ClinicalTrials.gov, indicating a need to present protocols and analysis strategy before data collection.

Because insomnia is diagnosed using self-reported indicators of poor sleep quality, self-reported measures are considered to be the most adequate measures of intervention effect in adults.55,56 However, an objective (actigraph and/or EEG) measure of sleep should be used as a complement to self-reported measures57 because such measures do not completely overlap.51 Daytime assessments, such as self-reported sleepiness, could also provide complementary information on sleep quality because this measure is a sensitive indicator of insufficient sleep (as measured by EEG).58 To develop effective CBTI-interventions in future care, it is important to understand how objective improvements in sleep quality are linked to perceived amelioration of insomnia symptoms and treatment satisfaction. Qualitative and participatory methods can add to the in-depth understanding of treatment results, experiences from treatment, and which outcomes are perceived as particularly relevant for youths. The use of both self-reported and objective measures could provide data that can be used to improve treatment protocols in a way that could not be obtained if only one type of measure were used. To date, no RCT on the effect of child or adolescent cognitive behavioral sleep interventions has used polysomnography, perhaps because of a combination of cost and feasibility issues. However, this technique could further contribute to the objective evaluation of CBT-I treatment effects in this age group.

Although this meta-analysis did not examine effects of cognitive and behavioral sleep interventions on comorbid mood disorders, the overlap between sleep disorders and psychiatric conditions, such as depression and anxiety, raises the question of underlying shared mechanisms between these conditions. Although CBT-I primarily targets sleep, a meta-analysis on adults with insomnia and comorbid mood disorders showed improvement in both depression and anxiety.59 Exploration of the directional relationship between sleep problems and depression in children and adolescents has led some researchers to propose a model with initial sleep disturbances leading to depression rather than the opposite scenario.60 Research on adults with insomnia who also suffer from depression show that treatment of the sleep disorder also improves symptoms of depression, but the opposite is not true.61 These findings suggest that insomnia is more than a symptom of depression and needs to be addressed specifically in treatment, perhaps even before addressing comorbid psychiatry due to the positive transdiagnostic treatment effect of CBT-I.61,62

CONCLUSIONS

The current meta-analysis suggests that cognitive and behavioral interventions of four sessions or more can improve sleep of school-age children and adolescents, particularly regarding wakefulness in bed variables such as SOL and WASO. The short-term follow-up treatment effects also suggest that these interventions can improve sleep over time with effects similar to those found in the adult populations. However, because of the heterogeneity of included studies regarding participants, protocols, and interventions, the risk of bias is high. Therefore, results should be interpreted with caution. As this field further develops, more stringent study designs should be applied in order to allow for more informative comparisons and syntheses. With a more solid base of evidence, treatment recommendations can be developed to ensure effective treatments for children and adolescents with disordered sleep.

DISCLOSURE STATEMENT

All authors have read and approved the final version of the manuscript. The authors have no financial relationships relevant to this article. Funding was provided by the Doctoral School in Health Care Sciences at Karolinska Institutet (LÅ, MK), Stockholm County Council-ALF project (LÅ, ML), Stockholm Stress Center (LÅ, ML, FA), Functional Area Medical Psychology at Karolinska University Hospital (MK), and the Swedish Brain Foundation (ML). The authors report no conflicts of interest.

ABBREVIATIONS

CBT

cognitive behavioral therapy

CBT-I

cognitive behavioral therapy for insomnia

DSM

Diagnostic and Statistical Manual of Mental Disorders

DSPD

delayed sleep phase disorder

EEG

electroencephalography

RCT

randomized controlled trial

SOL

sleep onset latency

TST

total sleep time

WASO

wake after sleep onset

ACKNOWLEDGMENTS

The authors thank Carl Gornitzki and Klas Moberg at the Karolinska Institutet University Library, Stockholm, for conducting the literature search.

REFERENCES

1 

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington DC: American Psychiatric Association Publishing; 2013.

2 

Quach J, Hiscock H, Ukoumunne OC, Wake M. A brief sleep intervention improves outcomes in the school entry year: a randomized controlled trial. Pediatrics. 2011;128(4):692–701. [PubMed]

3 

Souders MC, Mason TB, Valladares O, et al. Sleep behaviors and sleep quality in children with autism spectrum disorders. Sleep. 2009;32(12):1566–1578. [PubMed Central][PubMed]

4 

Johnson EO, Roth T, Schultz L, Breslau N. Epidemiology of DSM-IV insomnia in adolescence: lifetime prevalence, chronicity, and an emergent gender difference. Pediatrics. 2006;117(2):e247–e256. [PubMed]

5 

Dohnt H, Gradisar M, Short MA. Insomnia and its symptoms in adolescents: comparing DSM-IV and ICSD-II diagnostic criteria. J Clin Sleep Med. 2012;8(3):295–299. [PubMed Central][PubMed]

6 

Colrain IM, Baker FC. Changes in sleep as a function of adolescent development. Neuropsychol Rev. 2011 3;21(1):5–21. [PubMed]

7 

Feinberg I, Campbell IG. Sleep EEG changes during adolescence: an index of a fundamental brain reorganization. Brain Cogn. 2010;72(1):56–65. [PubMed]

8 

Crowley SJ, Acebo C, Carskadon MA. Sleep, circadian rhythms, and delayed phase in adolescence. Sleep Med. 2007;8(6):602–612. [PubMed]

9 

Gradisar M, Crowley SJ. Delayed sleep phase disorder in youth. Curr Opin Psychiatry. 2013;26(6):580–585. [PubMed Central][PubMed]

10 

Gottlieb DJ, Punjabi NM, Newman AB, et al. Association of sleep time with diabetes mellitus and impaired glucose tolerance. Arch Intern Med. 2005;165(8):863–867. [PubMed]

11 

Eaker ED, Pinsky J, Castelli WP. Myocardial infarction and coronary death among women: psychosocial predictors from a 20-year follow-up of women in the Framingham Study. Am J Epidemiol. 1992;135(8):854–864. [PubMed]

12 

Biggs SN, Lushington K, van den Heuvel CJ, Martin AJ, Kennedy JD. Inconsistent sleep schedules and daytime behavioral difficulties in school-aged children. Sleep Med. 2011;12(8):780–786. [PubMed]

13 

McGlinchey EL, Harvey AG. Risk behaviors and negative health outcomes for adolescents with late bedtimes. J Youth Adolesc. 2015;44(2):478–488. [PubMed]

14 

Anderson B, Storfer-Isser A, Taylor HG, Rosen CL, Redline S. Associations of executive function with sleepiness and sleep duration in adolescents. Pediatrics. 2009;123(4):e701–e707. [PubMed Central][PubMed]

15 

Carskadon MA. Sleep's effects on cognition and learning in adolescence. Prog Brain Res. 2011;190:137–143. [PubMed]

16 

Liu X, Buysse DJ, Gentzler AL, et al. Insomnia and hypersomnia associated with depressive phenomenology and comorbidity in childhood depression. Sleep. 2007;30(1):83–90. [PubMed]

17 

Orzech KM, Acebo C, Seifer R, Barker D, Carskadon MA. Sleep patterns are associated with common illness in adolescents. J Sleep Res. 2014;23(2):133–142. [PubMed]

18 

McMakin DL, Dahl RE, Buysse DJ, et al. The impact of experimental sleep restriction on affective functioning in social and nonsocial contexts among adolescents. J Child Psychol Psychiatry. 2016;57(9):1027–1037. [PubMed]

19 

Johnson EO, Chilcoat HD, Breslau N. Trouble sleeping and anxiety/depression in childhood. Psychiatry Res. 2000;94(2):93–102. [PubMed]

20 

McMakin DL, Alfano CA. Sleep and anxiety in late childhood and early adolescence. Curr Opin Psychiatry. 2015;28(6):483–489. [PubMed Central][PubMed]

21 

Chase RM, Pincus DB. Sleep-related problems in children and adolescents with anxiety disorders. Behav Sleep Med. 2011;9(4):224–236. [PubMed]

22 

Gregory AM, Rijsdijk FV, Lau JY, Dahl RE, Eley TC. The direction of longitudinal associations between sleep problems and depression symptoms: a study of twins aged 8 and 10 years. Sleep. 2009;32(2):189–199. [PubMed Central][PubMed]

23 

Amaral MO, de Almeida Garrido AJ, de Figueiredo Pereira C, Master NV, de Rosario Delgado Nunes C, Sakellarides CT. Quality of life, sleepiness and depressive symptoms in adolescents with insomnia: A cross-sectional study. Aten Primaria. 2017;49(1):35–41. [PubMed]

24 

Ohayon MM, Roth T. Place of chronic insomnia in the course of depressive and anxiety disorders. J Psychiatr Res. 2003;37(1):9–15. [PubMed]

25 

Koffel EA, Koffel JB, Gehrman PR. A meta-analysis of group cognitive behavioral therapy for insomnia. Sleep Med Rev. 2015;19:6–16. [PubMed]

26 

Morin CM, Culbert JP, Schwartz SM. Nonpharmacological interventions for insomnia: a meta-analysis of treatment efficacy. Am J Psychiatry. 1994;151(8):1172–1180. [PubMed]

27 

Lichstein KL, Wilson NM, Johnson CT. Psychological treatment of secondary insomnia. Psychol Aging. 2000;15(2):232–240. [PubMed]

28 

Mitchell MD, Gehrman P, Perlis M, Umscheid CA. Comparative effectiveness of cognitive behavioral therapy for insomnia: a systematic review. BMC Fam Pract. 2012;13:40. [PubMed Central][PubMed]

29 

Edinger JD, Means MK. Cognitive-behavioral therapy for primary insomnia. Clin Psychol Rev. 2005;25(5):539–558. [PubMed]

30 

Epstein DR, Sidani S, Bootzin RR, Belyea MJ. Dismantling multicomponent behavioral treatment for insomnia in older adults: a randomized controlled trial. Sleep. 2012;35(6):797–805. [PubMed Central][PubMed]

31 

Morin CM, Hauri PJ, Espie CA, Spielman AJ, Buysse DJ, Bootzin RR. Nonpharmacologic treatment of chronic insomnia. An American Academy of Sleep Medicine review. Sleep. 1999;22(8):1134–1156. [PubMed]

32 

Sadeh A. Cognitive-behavioral treatment for childhood sleep disorders. Clin Psychol Rev. 2005;25(5):612–628. [PubMed]

33 

Meltzer LJ, Mindell JA. Systematic review and meta-analysis of behavioral interventions for pediatric insomnia. J Pediatr Psychol. 2014;39(8):932–948. [PubMed]

34 

Blake MJ, Sheeber LB, Youssef GJ, Raniti MB, Allen NB. Systematic review and meta-analysis of adolescent cognitive-behavioral sleep interventions. Clin Child Fam Psychol Rev. 2017;20(3):227–249. [PubMed]

35 

van Straten A, van der Zweerde T, Kleiboer A, Cuijpers P, Morin CM, Lancee J. Cognitive and behavioral therapies in the treatment of insomnia: A meta-analysis. Sleep Med Rev. 2018;38:3–16. [PubMed]

36 

Buysse DJ, Ancoli-Israel S, Edinger JD, Lichstein KL, Morin CM. Recommendations for a standard research assessment of insomnia. Sleep. 2006;29(9):1155–1173. [PubMed]

37 

Drake C, Nickel C, Burduvali E, Roth T, Jefferson C, Pietro B. The pediatric daytime sleepiness scale (PDSS): sleep habits and school outcomes in middle-school children. Sleep. 2003;26(4):455–458. [PubMed]

38 

Higgins JP, Green S; The Cochrane Collaboration. Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 [updated March 2011]; 2011.

39 

Review Manager (RevMan) [Computer program]. Version 5.3. Copenhagen: The Nordic Cochrane Centre TCC; 2014.

40 

Cohen J. A power primer. Psychol Bull. 1992;112(1):155–159. [PubMed]

41 

Matthey S. p < .05 — but is it clinically significant? Practical examples for clinicians. Behavior Change. 1998;15(2):140–146

42 

de Bruin EJ, Bogels SM, Oort FJ, Meijer AM. Efficacy of cognitive behavioral therapy for insomnia in adolescents: a randomized controlled trial with internet therapy, group therapy and a waiting list condition. Sleep. 2015;38(12):1913–1926. [PubMed Central][PubMed]

43 

Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557–560. [PubMed Central][PubMed]

44 

Blake M, Waloszek JM, Schwartz O, et al. The SENSE study: Post intervention effects of a randomized controlled trial of a cognitive-behavioral and mindfulness-based group sleep improvement intervention among at-risk adolescents. J Consult Clin Psychol. 2016;84(12):1039–1051. [PubMed]

45 

Clarke G, McGlinchey EL, Hein K, et al. Cognitive-behavioral treatment of insomnia and depression in adolescents: A pilot randomized trial. Behav Res Ther. 2015;69:111–118. [PubMed Central][PubMed]

46 

Paine S, Gradisar M. A randomised controlled trial of cognitive-behaviour therapy for behavioural insomnia of childhood in school-aged children. Behav Res Ther. 2011;49(6-7):379–388. [PubMed]

47 

Cain N, Gradisar M, Moseley L. A motivational school-based intervention for adolescent sleep problems. Sleep Med. 2011;12(3):246–251. [PubMed]

48 

Moseley L, Gradisar M. Evaluation of a school-based intervention for adolescent sleep problems. Sleep. 2009;32(3):334–341. [PubMed Central][PubMed]

49 

Ohayon M, Wickwire EM, Hirshkowitz M, et al. National Sleep Foundation's sleep quality recommendations: first report. Sleep Health. 2017;3(1):6–19. [PubMed]

50 

Pillai V, Roth T, Drake CL. Towards quantitative cut-offs for insomnia: How current diagnostic criteria mischaracterize remission. Sleep Med. 2016;26:62–68. [PubMed Central][PubMed]

51 

Zhang L, Zhao ZX. Objective and subjective measures for sleep disorders. Neurosci Bull. 2007;23(4):236–240. [PubMed]

52 

Bastien CH, Morin CM, Ouellet MC, Blais FC, Bouchard S. Cognitive-behavioral therapy for insomnia: comparison of individual therapy, group therapy, and telephone consultations. J Consult Clin Psychol. 2004;72(4):653–659. [PubMed]

53 

Jernelöv S, Lekander M, Blom K, et al. Efficacy of a behavioral self-help treatment with or without therapist guidance for co-morbid and primary insomnia--a randomized controlled trial. BMC Psychiatry. 2012;12:5[PubMed Central][PubMed]

54 

Lancee J, van Straten A, Morina N, Kaldo V, Kamphuis JH. Guided online or face-to-face cognitive behavioral treatment for insomnia: a randomized wait-list controlled trial. Sleep. 2016;39(1):183–191. [PubMed Central][PubMed]

55 

Mundt JM, Crew EC, Krietsch K, et al. Measuring treatment outcomes in comorbid insomnia and fibromyalgia: concordance of subjective and objective assessments. J Clin Sleep Med. 2016;12(2):215–223. [PubMed Central][PubMed]

56 

Lichstein KL, Durrence HH, Taylor DJ, Bush AJ, Riedel BW. Quantitative criteria for insomnia. Behav Res Ther. 2003;41(4):427–445. [PubMed]

57 

Sadeh A. The role and validity of actigraphy in sleep medicine: an update. Sleep Med Rev. 2011;15(4):259–267. [PubMed]

58 

Akerstedt T, Anund A, Axelsson J, Kecklund G. Subjective sleepiness is a sensitive indicator of insufficient sleep and impaired waking function. J Sleep Res. 2014;23(3):240–252. [PubMed]

59 

Ye YY, Zhang YF, Chen J, et al. Internet-based cognitive behavioral therapy for insomnia (ICBT-i) improves comorbid anxiety and depression-a meta-analysis of randomized controlled trials. PloS One. 2015;10(11):e0142258. [PubMed Central][PubMed]

60 

Lovato N, Gradisar M. A meta-analysis and model of the relationship between sleep and depression in adolescents: recommendations for future research and clinical practice. Sleep Med Rev. 2014;18(6):521–529. [PubMed]

61 

Blom K, Jernelöv S, Kraepelien M, et al. Internet treatment addressing either insomnia or depression, for patients with both diagnoses: a randomized trial. Sleep. 2015;38(2):267–277. [PubMed Central][PubMed]

62 

Blom K, Jernelöv S, Ruck C, Lindefors N, Kaldo V. Three-year follow-up comparing cognitive behavioral therapy for depression to cognitive behavioral therapy for insomnia, for patients with both diagnoses. Sleep. 2017;40(8)


Supplemental Material

Supplemental Material

(.pdf | 705 KB)