Sleep and Autism: Current Research, Clinical Assessment, and Treatment Strategies

Abstract

Autism spectrum disorder is associated with a high rate of sleep problems, affecting over 80% of autistic individuals. Sleep problems have pervasive negative effects on health, behavior, mood, and cognition but are underrecognized in autistic children. Problems initiating and maintaining sleep—hallmarks of insomnia—are common. Sleep-disordered breathing and restless legs syndrome have also been described in autism at a higher prevalence than in community populations. The authors describe current research on sleep in autistic children and potential pathophysiologic mechanisms. They describe practical approaches to sleep assessment and synthesize approaches to addressing sleep problems in autistic children.

Clinically significant sleep problems affect over 80% of children with autism spectrum disorder (ASD), a rate two to three times higher than that in children with typical development (e.g., [1, 2]) The National Sleep Foundation identified autistic children as one of the highest priority populations for sleep research (3). However, sleep problems are underreported in ASD, even by developmental-behavioral pediatricians (4). Problems with initiating and maintaining sleep and with early awakening are frequently reported in autistic children and they can lead to insufficient overall sleep duration and poor sleep quality (5–11). These symptoms can be severe enough to meet criteria for insomnia or circadian rhythm disorders (12, 13). Sleep-disordered breathing, particularly obstructive sleep apnea (OSA), and restless legs syndrome (RLS) have also been described in autistic children and occur at a higher prevalence than in community populations (14, 15). The types of sleep problems in ASD may shift over the course of development, from bedtime resistance, sleep anxiety, and night wakings in early childhood to delayed sleep onset and shorter sleep duration in school age (8, 16). In contrast to the usual course of children with typical development, sleep problems in ASD tend to persist rather than resolve by school age (17).

Sleep problems have short- and long-term negative consequences for health, behavior, mood, and cognition in children and adults (17–19). Consistent with literature on neurotypical children as well as those with other psychiatric and health conditions, sleep problems in autistic children are associated with pervasive detrimental effects across multiple domains of function (20–27). Sleep alterations and variability in sleep timing in autistic children is associated with increased externalizing problems (16, 21). Inattention (28, 29), increased hyperactivity (10, 28), and aggression (10, 30) have been associated with sleep problems in ASD. Internalizing problems may also be increased with sleep problems in ASD (16, 31, 32). Anxiety and mood symptoms often co-occur with sleep problems in autistic children (33–35). Sleep problems have been associated with autism severity (10, 21, 36), including increased repetitive behaviors (28, 32, 37), communication problems (32), and social skills deficits (29). The wide range of behavioral and psychiatric problems associated with sleep problems is not surprising, given the known pervasive adverse effects of sleep problems (38). However, the relationship between sleep problems and intellectual disability in autism is inconsistent, and it is evident in some studies (22, 39) but not others (5, 21). Sex differences in sleep in autism are not well understood, which is due in part to the sex ratio in autism favoring males. Recent large database reviews suggest that autistic girls may have more sleep problems than autistic boys (40), and parent reports on a widely used questionnaire found that, among children ages 6–12, almost 85% of autistic girls had sleep problems compared with significantly lower levels in autistic boys and typically developing boys and girls (41). It is important to note that most of the available information on sleep in autism is based on parent report, and objective measures are needed to verify and extend these findings.

Potential Pathophysiologic Mechanisms

Although the high prevalence of sleep disturbances in autistic children is well established, the pathophysiologic relationships between sleep and autism are not as clearly understood. The biopsychosocial model of sleep in ASD posits that sleep problems result from a combination of alterations to brain architecture and function, psychological and behavioral characteristics, and factors in the home environment (20). It is widely recognized that each of these factors can be transactional; that is, they can be both the cause and the consequence of sleep problems. Genetic factors may confer increased risk for developing both ASD and sleep problems (22, 42). We found that problems with falling asleep in the first year of life can precede an autism diagnosis and are associated with altered patterns of brain development (43). It is possible that sleep problems and autism share common etiologies (44). Sleep problems may be a core feature of ASD, but there are likely to be different mechanisms depending on the nature of these disturbances (insomnia vs. sleep-disordered breathing vs. movement disorders). In this section, we explore the potential pathophysiologic mechanisms underlying sleep disorders commonly seen in autistic children.

Insomnia

When considering potential pathways that cause difficulties with initiating and maintaining sleep, three foci emerge: primary or predisposing factors that affect circadian and sleep regulation; secondary or precipitating factors such as mood, anxiety, and pain; and tertiary or perpetuating factors such as poor sleep hygiene.

A robust body of literature on autistic individuals and animal models of ASD suggests that dysregulation of melatonin, a hormone regulating sleep-wake cycles, is linked to sleep-wake disruptions in individuals with ASD, leading to difficulties with initiating and maintaining sleep (i.e., insomnia) (45). Lower levels of melatonin may result from abnormal production (46–48). Genetic polymorphisms resulting in abnormal melatonin metabolism (49, 50) may lead to either increased or decreased levels of melatonin during both wake and sleep. Mutations in circadian regulatory genes (22, 51–54), associated with sleep-wake dysregulation and divergent cortisol profiles, have also been described. Abnormalities of other neurotransmitters critical to sleep activation and wake inhibition (55), such as gamma-aminobutyric acid (GABA) (56), serotonin (57, 58), acetylcholine (59), and glutamate (60), have been described in those with ASD.

Changes in sleep architecture (61, 62) have also been seen in autistic children on polysomnography, suggesting that fundamental differences in sleep circuits may be present in some autistic children. Sleep studies have shown sleep macrostructure alterations (63), including reduction in REM sleep with increases in non-REM sleep. Microstructural abnormalities (11) on electroencephalogram analysis using cyclic alternating patterns (CAPs) have also been described, with autistic children having lower CAP rates.

Secondary factors, co-occurring conditions that are independently associated with sleep problems, can also be associated with insomnia in autistic children. Anxiety occurs at high rates in autistic individuals (for a review, see [64]), and attention-deficit hyperactivity disorder (ADHD) co-occurs in approximately 28% of autistic individuals (65, 66). It is not known whether anxiety- and ADHD-linked insomnia is additive, synergistic, or amplified in autistic children who have sleep problems, but the combination of autism with anxiety or ADHD can be a phenotype of severe and refractory insomnia. Chronic pain is also associated with increased likelihood of refractory insomnia in pediatric populations, including autism (67, 68). Pain may be underestimated in ASD because of communication challenges or poor interoception in some autistic children. It is important to rule out potential sources of unreported pain (69), such as ear infections, abdominal discomfort from reflux or constipation, oral pain related to unrecognized dental issues, “growing pains,” or amplified pain from sensory integration difficulties in autistic children who have sleep problems.

Tertiary factors tend to be specific to the home sleep environment. Examples include poor sleep hygiene, irregular sleep schedules, lack of routine, and inconsistent caregiver responses (70). In ASD, these are often further complicated by difficulties that children face with integrating environmental cues, circadian signals, and poor communication skills (71). Caregiver responses to these challenges can be impaired by caregiver exhaustion (72). Treatment—largely, behavioral modifications and caregiver education—help to attenuate insomnia in ASD (71).

Sleep-Related Breathing Disorders (SRBD)

SRBD is an umbrella term that includes multiple respiratory abnormalities, but most commonly refers to obstructive sleep apnea (OSA; [14, 73]). OSA is a common condition of childhood, with a prevalence of up to 4%, and studies have shown that children with ASD have a higher prevalence of OSA compared with their typically developing peers (14). Manifestations of untreated OSA in childhood range widely by age and development, but many symptoms overlap with ADHD and some features of ASD. The pathophysiologic mechanisms of OSA in ASD likely involve anatomic and neuromotor tone differences within the airway that lead to obstructive sleep-disordered breathing (14, 74, 75). Reflux and dysphagia are also more prevalent in individuals with ASD and may contribute to sleep-disordered breathing (76).

Some studies suggest that the symptom burden of SRBD in autistic children may be higher than in children without ASD, making diagnosis and treatment more complex (77). The presence of ASD can make the diagnosis and treatment of SRBD more challenging. Tolerance of in-laboratory polysomnography, the gold standard for diagnosing SRBD, is often low in autism. The behavioral and sensory sensitivities commonly seen in ASD can reduce the acceptance and effectiveness of interventions such as continuous positive airway pressure (CPAP) therapy. However, desensitization programs can improve the ability to participate in a sleep study and improve CPAP adherence (78, 79). Treatment of sleep-disordered breathing can lead to the improvement of target symptoms, such as daytime behavioral struggles, in autistic children (80). Improvements are not reliably predictable across children, and, at times, families may feel that the improvements do not reach clinical significance (80). It is important to discuss this possibility with caregivers and children when weighing the benefits and risks of engaging in diagnosis and treatment of SRBD.

Fundamental congenital or acquired neuroanatomic differences may result in irregular control of breathing, leading to central sleep apnea or hypoventilation (81). Many individuals with genetic syndromes, such as Joubert syndrome, Down’s syndrome, Smith-Magenis syndrome, and Prader-Willi syndrome (82–85), have functional and behavioral features that are consistent with ASD. It is not known whether the SRBD in these syndromes plays a role in ASD phenotype, or whether treatment of SRBD changes ASD severity in these syndromes. Treatment for SRBD needs to be prioritized for individuals with syndromic autism, because SRBD can be life-threateningly severe in some cases.

Movement Disorders

Sleep disturbances such as periodic limb movement disorder (PLMD) or restless sleep disorder (RSD) (15, 86) are more prevalent in pediatric autism than previously thought. RLS (87) is related to similar pathophysiologic pathways as PLMD and RSD, but disruptive movements occur during wakefulness, most prominently at night before sleep. Iron deficiency has been identified as a contributor to the symptoms, mainly by altering dopaminergic pathways involved in motor activity (87). Hence, a decrease in bioavailable iron, even if not overtly anemic, may result in PLMD, RLS, or RSD (88). Given significant dietary and nutritional limitations in some autistic children, iron deficiency should be assessed and treated when children present with sleep problems.

Understanding the pathophysiology behind sleep disorders in autistic children is crucial for developing tailored interventions and improving the overall well-being and quality of life for those affected by this neurodevelopmental disorder. Further research and increased clinical efforts are essential to address this challenging issue comprehensively.

Sleep Assessment

Sleep History

Both the caregiver and autistic child should be assessed in a comfortable, nonthreatening environment with minimal distraction. Separating the caregiver interview from the child interview or examination may be beneficial for adolescents with emerging independence, young children who are highly disruptive, or caregivers who do not feel comfortable discussing their child’s problems in front of the child. A sleep history can be approached categorically, such as with the BEARS screening tool (89): BEARS is an acronym for five key assessment questions related to bedtime problems, excessive daytime sleepiness, awakenings at night, regularity and duration of sleep, and sleep-disordered breathing. The screening tool can be used with caregivers of children ages 2 to 12 years or directly with school-age children ages 6 to 12 and adolescents ages 13 to 18. Alternatively, the clinician can conduct a review of a typical 24-hour cycle reviewing details of sleep-wake patterns, habits, routines, and difficulties. Recognizing the manifestations of “sleepiness” across the developmental spectrum is critical. For example, sleep-deprived prepubertal children will tend to become more hyperactive and impulsive compared with postpubertal adolescents, who are more likely to be overtly sleepy. Hence, an inquiry such as “Is your child sleepy?” regarding a young child may lead to misclassification. In autism, a caregiver report should be used when the child’s verbal ability precludes self-report. However, the caregiver report may be limited by a lack of reliable information (e.g., sleeping in different rooms) or (mis)judgment stemming from the parents’ own sleep deprivation, stress, or perceptions. In autistic children presenting with sleep problems, a history of medical and psychiatric disorders that often co-occur with sleep disorders should also be assessed.

Sleep-Specific Assessment Tools

Sleep diaries and actigraphy can be very useful for establishing daily sleep-wake patterns and providing an objective estimate of total sleep time and sleep efficiency in the home environment. Autistic children often have no problems tolerating wrist or ankle actigraphy, and with appropriate desensitization, actigraphy for most children can be successful. However, actigraphy requires specialized equipment, scoring, and interpretation, so it may not be available in a typical pediatric practice. Sleep diaries can provide specific information about daytime and nighttime sleep-related challenges and behaviors, they have the advantage of not relying on retrospective memory, and they can be collected over a longer time frame to capture nightly variability (90).

Standardized questionnaires such as the Children’s Sleep Habits Questionnaire (91) were developed for research purposes and not intended for widespread clinical use. Overnight, in-lab polysomnography closely evaluates breathing, brain wave activity, and body movements and is considered the gold standard for diagnosis of SRBD. However, polysomnography can prove difficult for many autistic children given the unusual sensory input (i.e., multiple leads) and unfamiliar routines and environment of testing. With desensitization, pediatric-specific expertise, and lots of preparation, more children with ASD can tolerate this test. Home sleep apnea testing is not yet routine, nor is it recommended for children under age 18.

Sleep Interventions

Sleep Education

Educating caregivers about how to prepare the sleep environment is usually the first step in addressing sleep problems. Most of the information that caregivers of autistic children need about sleep is consistent with the information needed by caregivers of neurotypical children. Caregivers need to provide the structure for healthy sleep, because the home sleep environment provides important entrainment cues (92, 93). The home sleep environment includes the physical environment (e.g., noise, light, temperature). A cool (60°–68°F; [94]), quiet bedroom, free of arousing stimuli such as computers and TVs, is important for healthy sleep. Caregivers need to know the total amount of sleep and recommended sleep timing for their child’s age group. Providing consistent routines and consistent timing for bedtime and awakening in the morning can be a challenge, but it is a key to improved sleep for many autistic children. Strategies to avoid media use in the 1 to 2 hours before bedtime should be discussed. It is difficult for autistic children to change routines and habits after they are established; therefore, if possible, establishing a healthy routine early in development is recommended. One autism-specific caution is that parents should strive to include small changes to the bedtime routine so children get used to some level of flexibility in the types and order of the activities before bedtime. This can prevent evening meltdowns when changes to the routine occur. In addition, parents should be coached to keep the bedtime routine under 30 minutes so it does not extend and interfere with timely sleep.

Sleep education research suggests that educational materials alone may not be sufficient to improve sleep in autistic children (ages 2–10 years; [91]). When sleep education involves meeting with a clinician as well as providing materials, it is more efficacious (95–98). Individual and group education delivery appear equally effective (96). Sleep education research is in its early stages and has methodological limitations (it relies on parent reports as primary outcomes and uses pre-post designs). Future randomized studies are needed to determine the impact of sleep education on sleep problems in autism.

Behavioral Interventions

Clinical guidelines advocate for behavioral sleep interventions in ASD as an initial approach to treatment for insomnia and behavioral sleep problems (69, 71, 99). Behavioral sleep interventions can improve sleep problems in autistic children, including total sleep, night wakings, resistance at bedtime, and daytime behaviors, with relatively low time commitment (100). Behavioral sleep interventions are differentiated from sleep education by the use of strategies implemented at home by caregivers tailored to the child’s specific sleep challenge. A detailed review of the content and approaches in behavioral sleep interventions is beyond the scope of this paper; additional information is provided elsewhere (101–103). Behavioral interventions are lower cost, have fewer side effects, and are more durable than pharmacological interventions (3). Even brief group-based behavioral sleep workshops for parents of autistic children have shown evidence of improving sleep problems and daytime behaviors (104). Behavioral parent training programs that teach parents about behavioral strategies (e.g., reinforcement, time-out) without a specific focus on sleep have not shown efficacy in improving sleep in autistic children (105).

Pharmaceutical Interventions

Melatonin, which is considered by the U.S. Food and Drug Administration as a supplement, has demonstrated efficacy and safety in autistic children (106, 107). However, long-term outcome studies are lacking. A randomized placebo-controlled trial with autistic children ages 4 to 10 years suggested that controlled-release melatonin and cognitive-behavioral therapy (CBT) were both effective in treating insomnia in this population (108). The combination of melatonin and CBT demonstrated greater efficacy than either single-component treatment group. The optimal dose of melatonin has not been determined, although studies have reported benefit with 1 to 10 mg and, perhaps, most benefit between 1 and 5 mg (106, 107). Other categories of medications besides melatonin or melatonin agonists, such as alpha agonists, antihistamines, antidepressants, and antipsychotics, have been reported in case reports and series with varying efficacy on sleep latency and sleep quality but have not been robustly studied (109).

OSA Interventions

Treatment of OSA leads to varying degrees of improvement, depending on individual differences in the child and OSA severity. OSA treatment can improve quality of life, behavior, cognition, academic performance, breathing, and sleep (110). First-line treatment of OSA in children with adenotonsillar hypertrophy is adenotonsillectomy (AT; [73]). However, “cure” rates after AT are lower than previously believed. Treatment of persistent OSA after AT is an area of very active exploration. The second-line mainstay of treatment used to be CPAP, but this is challenging for many children, perhaps even more so in those with ASD, and may require desensitization to be successful. Novel treatments include drug-induced sleep endoscopy–directed surgeries in the upper airway and hypoglossal nerve stimulators (111).

Restless Sleep Interventions

For RLS, PLMD, and RSD, optimization of the sleep environment and adequate sleep opportunity are foundational (112, 113). Then, first-line pharmaceutical treatment involves iron supplementation if ferritin levels are below thresholds recommend for RLS management. Oral iron supplementation is generally safe but may be associated with worsening of constipation or may be poorly tolerated because of taste. Intravenous iron supplementation can be considered if oral iron supplementation is not successful, but safety is important to consider (114). Overall, iron supplementation can be very effective at decreasing symptoms of restlessness across all three disorders. Specifically, ferric carboxymaltose formulations have been reported as effective in autistic children (113).

Conclusions

Sleep problems result from a combination of primary (e.g., molecular, neurobiological), secondary (e.g., psychological, behavioral), and tertiary (e.g., home sleep environment) factors. These factors can be both a cause and a consequence of sleep problems, but a fundamental question is whether sleep problems and ASD share a common etiology. Sleep problems affect a large majority of autistic children and are associated with pervasive detrimental effects across multiple domains of function. Problems with sleep in autistic children can be severe enough to meet the criteria for insomnia or circadian rhythm disorders (115–117) (summarized in Box 1). Assessing and treating sleep problems in autistic children can improve sleep, health, and daytime functioning for children and their caregivers.

BOX 1. Abbreviated sleep disorder diagnostic criteria

Insomnia (115): The child has difficulty with either their sleep quality or their quantity. This might mean that the child has difficulty falling asleep and/or difficulty staying asleep; the child may wake up early in the morning and be unable to fall asleep again. The child’s sleep difficulty comes with next-day consequences, such as difficulty concentrating at school. Sleep difficulty must occur at least three nights per week for at least 3 months.

Sleep-Related Breathing Disorders (SRBD) (116): Abnormalities of respiration during sleep including:

1. Obstructive sleep apnea

   1. Upper airway narrowing or closure during sleep while respiratory effort continues

2. Central sleep apnea disorders

   1. Reduction or cessation of airflow due to absent or reduced respiratory effort

3. Sleep-related hypoventilation disorders

   1. Abnormal increase in arterial carbon dioxide (CO₂) during sleep (For children, CO₂ must be greater than 50 mmHg for more than 25% of the total sleep time.)

4. Sleep-related hypoxemia disorder

   1. Sustained periods of significantly reduced oxyhemoglobin saturation during sleep

Sleep-Related Movement Disorders (116): Group of disorders characterized by relatively simple, usually stereotyped, movements that disturb sleep or its onset.

Restless Legs Syndrome (RLS) (116): An uncomfortable urge to move legs that is worse during inactivity, is relieved by movement, and occurs mostly in the evening, the symptoms of which cause daytime impairment.

Periodic Limb Movement Disorder (PLMD) (116): Frequent periodic limb movements on sleep study (more than five per hour in children) that cause insomnia, sleep disturbance, and/or daytime impairment.

Restless Sleep Disorder (RSD) (117): Frequent large muscle movements in sleep (five per hour) that cause sleep disturbance and daytime impairment.

Circadian Rhythm Sleep-Wake Disorders (116): Alterations of the circadian time-keeping system or its entrainment mechanisms, or a misalignment of the endogenous circadian rhythm and the external environment.