Up to 20 % of the pediatric population [7] suffers from a diagnosable anxiety disorders according to the most recent DSM-5 classification [8], not considering previously included diagnoses such as post-traumatic stress disorder (PTSD) and obsessive-compulsive disorders (OCDs).

The incidence of children encountering significant stressful anxiety without otherwise meeting conditions for being properly classified among current diagnostic criteria for anxiety disorders is thought to be quite elevated.

A relatively recent report [9] estimates a prevalence of transient sleep problems in 85 % against 50 % of chronically impaired sleep, in children with anxiety disorders.

Typical habits of anxious children vary with age and cultural diversities, ranging from bedtime fears and rituals to requests for transitional objects, crying, and co-sleeping.

Nocturnal awakenings are the most consistent marker of all anxiety disorders [2].

Family habits and parental conflicts or psychopathology significantly interfere with anxiety symptoms, favoring the development of sleep disorders. In particular, lack of structure and inconsistent parenting styles may negatively influence the development of correct maturational skills such as self-soothing at bedtime, thus depriving the child of early opportunities to achieve this imperative ability [10]. Nighttime fears and recurrent nightmares may be emblematic of traumatic abusive experiences, albeit generally present in most anxiety disorders.

In addition, sleep-related symptoms in anxiety disorders vary across the life span, being expressed as bedtime fears and refusal in younger children or as disruptive nightmares with prominent sleep fragmentation in adolescents [2].

The severity of sleep problems correlates with functional impairment within anxiety disorders, with early sleep disruption holding a predictive negative value for the development of anxiety disorders but not for depression in later years [10, 11]. Longitudinal community-based studies give rise to the concern that anxious children may often underreport their sleep problems in comparison to depressed kids [12], thus rendering parental reports and objective findings crucial to diagnosis [13].

Separation anxiety disorder (SAD) is by far the most common anxiety disorder in very young children, accounting for most of the referrals in the field. Ninety-seven percent of these children experience sleep problems, most commonly initial insomnia and bed refusal without the presence of a significant attachment figure. These children also experience more awakenings and frequent enuretic episodes besides typical disorders of arousal such as sleepwalking and night terrors [14].

Generalized anxiety disorder (GAD) is, instead, characterized by more generic worries about school and home in this age group, with a lifetime prevalence estimated of approximately 5 % [8]. Sleep problems are overly common, being reported by 90 % of the pediatric population with GAD. To simplify, they could be summarized as generating a “hyperarousal” state leading to difficulties in both initiating and maintaining sleep, nightmares, as well as difficulties waking up in the morning and daytime somnolence [9, 15]. A few objective studies offer conflicting results about sleep structure in anxiety disorders. Forbes et al. [13] found a prolonged SL, more awakenings, and reduced slow-wave sleep (SWS) when comparing healthy and depressed subjects. Alfano et al. [16] also indicated a reduced REM latency in the control group. In fact, a recent meta-analysis [17] reported very similar objective findings in polygraphically recorded anxious and depressed children.

Children with PTSD have been exposed to and reexperience a traumatic event. They manifest increased stimuli reaction, avoidance of trauma-related stimuli, hyperarousal, and, sometimes, dissociative behavior. Approximately 14 % of children exposed to traumatic events develop PTSD. Trauma may be related to physical or sexual abuse or to disastrous experiential events such as hurricanes, earthquakes, and terrorist attacks. All PTSD subjects are five times more likely to show sleep disruption after 3 years from the event [18]. A few actigraphic studies confirmed sleep fragmentation and poor quality sleep [19] and significantly increased SL with enhanced nocturnal activity [20].

OCD has a lifetime prevalence of 1–2 %, 0.8 % within the pediatric population [8]. Bedtime routines may be extenuatingly long and interfere with sleep onset especially, but also with sleep continuity by promoting a high rate of waking after sleep onset (WASO). Also TST is inversely related to the severity of compulsions [21]. The one and only, very dated, PSG study in OCD adolescents revealed decreased sleep efficiency (SE) with increased SL [22]. The importance of a genetic component has been demonstrated by exposing high-risk (one parent with a diagnosis of social anxiety) versus normal-risk children to different emotional facial stimuli. Frontal, temporal, and limbic areas were selectively overactivated during exposure in high-risk versus typical-risk children. These are the same areas that are thought to be potentially responsible for nightmares and hyperarousal during sleep [23].

Major depressive disorder (MDD) presents with pervasive sadness, loss of interest, and pleasure leading to significant impairments in social and academic life [8]. Additional symptoms in children include irritability, behavioral dysregulation, and failure to gain weight. MDD prevalence increases with age from 1 % in early years up to 8 % in adolescence, often with a recurrent course [2]. Two-ninth of the descriptive features of MDD in the DSM-5 are related to sleep alterations such as insomnia or, conversely, hypersomnia. The latter is rare in pediatric depression; however, when co-occurring with insomnia, it carries a negative and dire prognosis [24]. Insomnia is usually associated with psychomotor agitation and restlessness, rumination, and worries, whereas hypersomnia correlates with psychomotor delay, hopelessness, and decreased vital energy. All of these symptoms strictly reflect mood oscillations which are influenced in a bidirectional way. In fact, in a prospective cohort study run on a community-based sample, insomnia at baseline increased by two- to threefolds the risk for MDD; conversely, MDD increased the occurrence of subsequent insomnia by the same measure. Early persistent sleep problems hold a negative prognostic value, predicting the development of anxiety and mood disorders later in adult life [25].

In examining special subgroups with suicidal ideation, 87 % of the sample was reported to suffer from persistent sleep problems [26]. Dated objective PSG measures carried out in pediatric populations revealed reduced REM and increased SL, increased REM density, and reduced and fragmented SWS [27, 28]. A more recent meta-analysis, however, found a significant difference of increased sleep latency in over 31 % of depressed children compared to normal controls. In addition, intra- and interhemispheric temporal coherence was decreased in the same percentage of patients [17]. Actigraphy, on the other hand, was only able to detect a blunted diurnal activity with delay of reaction phase in depressed children [29].

Maturational and gender-related factors are also powerful modulators of sleep features. Females, in fact, do not differ from healthy controls, whereas males exhibit the shortest REM latencies, fewer SWS, and highest arousals and transitional phases [30]. Greater REM density and longer SL were significantly associated with hospitalization and suicidality [31]. Adolescent female patients show the most drastic PSG changes compared to males. Trazodone and fluoxetine have both been employed for the treatment of adolescents with MDD, alone or in combination [32]. Hypnotic agents should be used judiciously and for short periods in this age group, so as to avoid excessive daytime sedation and worsening of comorbid disorders [33].

Pediatric patients experience different bipolar symptoms from adults. Especially in prepubescent children, rapid or even continuous cycling is common with both manic and depressive overlapping. The hallmark of bipolar disorder is mania co-occurring with a decreased need for sleep along with other key symptoms including grandiosity, hypersexuality, and racing thoughts and ideas. Few studies report sleep features in early-onset bipolar disorder, and they all reflect a core symptom: decreased need for sleep paralleling the most severe mood episodes [34].

Only two studies collected PSG data. Rao et al. [35] found increased transitional phases and reduced SWS, but no significant differences as far as REM sleep. On the other hand, another study that assessed children via the Child Behavior Checklist revealed increased WASO and lower REM sleep compared to the control group [36]. Two actigraphic studies [37, 38] showed opposite results: decreased SE and duration with prolonged SL versus longer TST and less activity in between episodes coupled with a subjective report of unrefreshing fragmented sleep. Shorter SL and longer TST were also observed in unaffected children with a familial risk for bipolar disorders [38].

Fortunately, the incidence of early schizophrenia in children under the age of 15 is very low, being less than 1 in 10,000 children. The DSM-5 [8] does not include sleep disturbances among the diagnostic features. However, insomnia is the most common side effect in young schizophrenics. Co-occurrence of insomnia and excessive daytime sleepiness (EDS) predicted psychotic episodes in adolescents and potential risk of psychosis based on specific structured rating instruments [39]. Sleep dysfunction in a subsequent study on “ultrahigh-risk” (UHR) adolescents was found to correlate more with negative rather than positive symptoms [40]. No PSG studies are available so far in children with schizophrenia. In adults, SWS reduction and disrupted architecture are thought to be trait markers of the disease, correlating with severity of psychotic symptoms and lasting over remission [41]. Reduced SE and TST along with increased SL and decreased REM density and latency comprise other important and confirmed features [42].

Several non-pharmacological approaches are available for the treatment of sleep problems in pediatric psychiatric disorders. They include cognitive behavioral therapy (CBT) often combined with medications [43], sleep hygiene, and behavioral intervention to address maladaptive sleep habits. For very young children, correct seeking-and-reward consequence systems to promote adaptive behaviors work best [44]. Avoiding presleep frightening TV contents and creating self-soothing rituals may quell bedtime anxiety. Sleep consolidation, instead, is promoted by delaying bedtime, as suggested by bed restriction therapy (BRT), in order to favor sleep pressure in keeping with the sleep homeostasis predicate [45, 46]. More specific techniques may be employed to avoid negative presleep worries and rumination by planning a session of positive relaxing thoughts through imagery distraction [47].

Autism spectrum disorders (ASDs) refer to a gamut of developmental disorders impacting communication and social skills, characterized by the expression of restricted repetitive stereotyped behaviors. This category includes autistic and Asperger disorders in addition to pervasive developmental disorder (PDD) not otherwise specified. A substantial increment in ASD diagnosis has been observed over the last decades, with an estimated prevalence rate of almost 70/10,000, likely due to increased awareness, genetic and environmental factors, and the extension of ASD diagnosis to less severe and uncommon forms of the disorder. Sleep problems are often experienced by ASD subjects (see Table 17.2). Children comprise up to 80 % of the whole ASD population. ASD seems to represent an independent risk factor for the occurrence of sleep disorders, besides that conferred by intellectual deficit alone [48]. Insomnia is the most common complaint from parents of ASD children. Initial as well as maintenance and early morning insomnia may coexist in varying associations. Mostly behavioral insomnia of childhood, either limit setting or sleep association type, seems to recur in these children where behavioral problems perpetuate innate neurobiological deficits linked to serotonin and glutamate domains [33]. Circadian rhythmicity is also dysfunctional due to an abnormal melatonin regulation, thus contributing to a delayed phase shift with melatonin rising in the morning rather than at nighttime. Several objective abnormalities have been detected in these children’s PSGs regarding microstructural aspects of both REM and NREM sleep [49]. They also exhibit reduced TST and REM latency, with a lower cyclic alternating pattern (CAP) rate in SWS due to selectively reduced percentage of A1 subtypes. This parallels what is seen in other mental retardation (MR) disorders where IQ and cognitive abilities negatively relate to the A1 percentage. A previous study from the same group [50] compared ASD sleep with that of normal children and children with MR and fragile X syndrome, showing reduced REM latency and increased transitional (ST1) phases in ASD compared to normal controls, whereas sleep findings almost overlapped with those observed in fragile X subjects. Interestingly, mental abilities correlated with tonic variables such as TST and WASO, whereas communicational skills and activity levels were significantly related to REM variables. In other words, cognition separates from verbal/communication skills, the former being mainly affected by alterations of SWS and the latter by REM sleep alteration. Within the ASD gamut, children with high-functioning Asperger disorder exhibit significantly higher A1 subtype percentage, akin to normal controls, compared to autistic children [51], showing a positive correlation of A1 percentage with verbal IQ and performance IQ as far as their duration is concerned. ASD does not seem to be an independent risk factor for obstructive sleep apnea (OSA), unless local risk factors such as adenotonsillar hypertrophy or craniofacial malformations are at stake. For this reason, PSG is warranted whenever an organic primary sleep disorder is suspected. This also includes the possible occurrence of PLMS, which several authors report as highly prevalent (nearly 50 %) in ASD children compared to controls. ASD children have been also reported to exhibit low serum ferritin levels [52] compared to normal subjects, and, as known, this is an independent risk factor for sleep-related movement disorders including PLMS and RLS.

Gastrointestinal problems linked to a deficient serotonin metabolism are highly prevalent in ASD [53] and may impact both sleep and iron absorption. ASD children also exhibit several parasomnias including arousal disorders, enuresis, and nightmares. REM without atonia, akin to what is seen in older adults with REM behavior disorder (RBD), has been reported in a case series [54]. Bedtime clonazepam, as in typical RBD, improved both daytime and nighttime behavior in these children.

Treating insomnia in ASD children may be a challenge [55]. Complete versus gradual extinctions are the most used behavioral techniques. Behavioral interventions such as chronotherapy [56], massage, and dental appliances [57] for the management of OSA often require additional pharmacotherapy. One to 3 mg of melatonin represents the safest and most efficient treatment for the majority of ASD-related sleep problems. Extended release melatonin may be selectively indicated for sleep maintenance insomnia [58].

ADHD has an increasing prevalence, recently estimated around 12 %, especially in the Western world. According to the DSM-5 [8], ADHD refers to an impairment in three major areas: attention, hyperactivity, and impulsivity. Three major clinical/phenomenological subtypes are generally recognized: predominantly hyperactive-impulsive type (H), predominantly inattentive (I), and a combined type (C). Both gender and age play a major role in the phenotypic expression of this disorder, with an estimated 1:10 male prevalence, associated with more disruptive symptoms in this gender and a tendency to subside, but not completely disappear, by adult age.

Sleep disturbance is an important hallmark of ADHD with over 80 % of affected children reporting inadequate or altered sleep [59]. Most data deal with subjective children or parental reports and/or with actigraphic data, whereas only few data of PSG recordings are available. Several meta-analytic reviews [60–62] were published over the last decade, dealing with the many confounding factors affecting most of the relevant studies including gender, age, comorbid disorders, medication and diversity of methods, and data collection.

Subjective reports of sleep complaints in ADHD include initial insomnia with delayed sleep onset, which has been interpreted in some cases as delayed sleep phase. Endogen circadian alterations [63] along with forced ultradian cycling [64] have been quoted as possible responsible mechanisms.

More often, though, an increased sleep duration along with multiple nocturnal awakenings and parasomnias has been described [65]. In addition, increased EDS compared to normal controls has been reported [65, 66]. This might be related to sleep-disordered breathing (SDB) with snoring and mild apneas or to narcoleptic-like traits [67, 68]. Despite the fact that ADHD symptoms, in particular attention deficits, are common in narcolepsy, no conclusive evidence has been reached of narcoleptic traits in ADHD children [69].

Restlessness and increased number of movements during sleep in ADHD types H and C are almost unanimously reported [62, 70–72]. Multiple studies report an increased prevalence of PLMS in ADHD [67, 73]. RLS has instead been reported in up to 44 % of ADHD children [74] with a tendency to decrease with age. In fact, only a 20 % prevalence was reported in young adults [75]. Common underlying genetic and pathophysiological alterations in both ADHD and RLS may be related to iron deficiency [76, 77] and dopaminergic transmission [78].

Among sleep-related movement disorders (SRMDs), bruxism [71] and rhythmic movement disorders [79] also seem to occur with discrete frequency, the latter mostly in the inattentive (I) ADHD subgroup, with a tendency to persist beyond the usual age range in relation to its common occurrence [80].

According to some authors [71, 81, 82], there is an increased prevalence of parasomnias in ADHD children, whereas others report no differences in their prevalence compared to pediatric controls [70].

One possible explanation of these conflicting results, especially in relation to disorders of arousals (DOA), could be linked to the possible co-occurrence of SDB as a major precipitant of night terrors, sleepwalking, and confusional arousals. DOA have been reported in up to 50 % of an ADHD cohort via clinical interview [71] with confusional arousals being most common. According to both Gau et al. [81] and Silvestri et al. [71], cognitive deficits rather than behavioral symptoms are more indicative of children with DOA as opposed to children with increased nocturnal hyperactivity.