Table 1 Main findings about sleep problems in ASD

Schreck et al. [13]

ASD children (N = 55, range 5–12 years, mean age = 8.2 years)

A database of parent report of sleep problems of children based on a self-report demographic form, the GARS, and the BEDS

Sleep problems (fewer hours of sleep per night, increased sensitivity to environmental stimuli in the bedroom) and the diagnostic characteristics of autism (social skill deficits, communication problems, developmental sequence) disturbances) may be related

Limitations: absence of a control group of healthy individuals

Sikora et al. [14]

ASD children (N = 1193, M = 1014, F = 179, range = 4–10 years)

Measures included Children’s Sleep Habits Questionnaire, Vineland Adaptive Behavior Scales, Survey Interview Form, Second Edition, and Child Behavior Checklist

Even if sleep is negative related with internalizing and externalizing behaviour, it could be differently related with the acquisition of adaptive skills

Limitations: few differences were clinical significant; information about sleep and daytime behaviour was registered through parents reports; use of quartile scores as the cutoff between mild and moderate to severe sleep problems may not be valid; even though the study highlights the bidirectional relationship between sleep and daytime behaviour, other factors are supposed to be implicated; disproportion between male and female (M/F = 1014/179)

Strengths: large sample size

Goldman et al. [15]

ASD children (N = 1859, range 3–18 years, mean age 80.1 ± 42.3 months)

CSHQ, PCQ

Sleep problems persist through adolescence in ASD with differences in types of problems experienced

Limitations: data are subjective and based on parental report; the cross-sectional nature of these data limits evaluation of a temporal relationship between age and sleep; the CSHQ and PCQ have not been validated with adolescents; not differentiation between males and females

Strengths: large sample size

Ballester et al. [20]

ASD adults (N = 41, M = 31, F = 10, mean age = 33 ± 6 years)

Typically developing adults (N = 51, M = 21, F = 30, mean age = 33 ± 5 years)

ACM recording wrist temperature, motor activity, body position, sleep, and light intensity

Poorer sleep conditions in adults with autism (increased sleep latency and number/length of night awakenings) resulted in decreased sleep efficiency

Limitations: these problems are life-long conditions, not only childhood related; parents or legal guardians of adults with ASD and ID with disturbed sleep were more likely to participate compared to those without sleep problems; ACM had not been validated in ASD to study sleep; while the comparison group was healthy, ASD one was medicated with polypharmacy; disproportion in M/F ratio particularly in the first group

Strengths: it is a controlled study with similar numerosity between groups

Baker et al. [22]

ASD adults (N = 36, M = 17, F = 19, IQ > 80, range = 21–44 years, 20 of them met criteria for insomnia and/or a circadian rhythm sleep–wake disorder) and age and sex-matched controls (N = 36, M = 17, F = 19 4 of them met criteria for the disorders above)

14-day actigraphy assessment and questionnaire battery

It has emerged, for the first, time that sleep problems are associated with unemployment in adults with autism spectrum disorder

Strengths: it is a controlled study with similar M/F ratio in both groups. Limitations: studies with larger ASD samples would be useful

Deserno et al. [23]

ASD adults with subjective QoL as outcome (N = 598, M = 310, F = 288, range = 17–83 years) ASD adults with objective QoL as outcome (N = 544, M = 270, F = 274, range = 17–82 years)

For objective QoL: five subscales of the Autism Quotient (Baron-Cohen, Wheelwright, Skinner, Martin, & Clubley, 2001), five subscales of the Sensory Perception Quotient (Tavassoli, Hoekstra, & Baron-Cohen, 2014), seven items of the Insomnia Severity Index (Bastien, Vallières, & Morin, 2001). For subjective QoL: an item assessing how satisfied participants were with their own life

Sleep problems are highly influential in predicting long-term QoL in ASD individuals

Limitations: a specific set of symptom data and environmental factors was used for determination of multivariate pathways; individuals have a long-term level of happiness to which they always spontaneously return after life events of either valence; the authors were limited by online survey context and were unable to verify diagnosis and IQ of participants; the study sample has a large age range (17–83); there is not a control group

Strengths: similar groups composition in terms of age and M/F ratio

Horiuchi et al. [16]

Autistic toddlers (N = 26, M = 20, F = 6) and non-autistic toddlers (N = 400, M = 184, F = 216)

Total range:17–19 months

Total mean age: 18 months

Japanese version of the M-CHAT-JV and the CASC

Autistic traits are associated with sleep problems in toddlers. As a result, daytime sleepiness might be a visible symptom that enables the earlier detection of ASD in children

Limitations: parents of toddlers who attended nursery school (199/426) had few information about their children’s daytime behavior; it is difficult to assess sleepiness in toddlers; few data are available on the reliability and validity of the CASC, and the previous studies that have used the CASC with toddlers; some items are not suitable for toddlers; disproportion between ASD and control group and in M/F ratio in ASD toddlers

Strengths: controlled study

Türkoğlu et al. [17]

ASD drug-naïve children (N = 46, M = 38, F = 8, range: 4–17 years, mean age 7.89)

AuBC, CSHQ, CCQ

Children with ASD during the home confinement reported higher chronotype scores and autism symptom scores compared to the normal non-hone confinement state. The sleep problems of the children with ASD during the home confinement period mediated the relationship between chronotype score and severity of autism symptoms

Limitations: lack of a control group; small sample size; family members were not screened for psychopathology, post-confinement follow-up with participants was not done; disproportioned M/F ratio

Miike et al. [18]

ASD children from K-Development Support Center for Children (K-ASD, N = 121, M = 94, F = 27), ASD children from H-Children’s Sleep and Development Medical Research Center (H-ASD, N = 56, M = 40, F = 16), Children from recruited from four nursery schools in T-city (control) (N = 203, M = 104, F = 99)

Questionnaires to assess parent(s) of children with ASD and controls investigating: maternal lifestyle during pregnancy, neonatal sleep patterns, status of parent(s) in the child-rearing years

Neonatal sleep–wake rhythm abnormalities, especially in irritable-type neonates, are important precursors for future ASD development and so it is important to pay much more attention to the maternal role in fetal chronobiology formation and to circadian rhythm formation

Limitations: disproportioned M/F ratio in ASD groups

Strengths: it is a controlled study

Yavuz-Kodat et al. [19]

ASD children (N = 52, M = 41, F = 11, range = 2.75–9.57 years, mean age = 5.39 ± 1.50 years)

Sleep and circadian rest–activity rhythms were objectively measured with actigraphy and subjectively with the Children’s Sleep Habits Questionnaire. Behavioral difficulties were assessed using the ABC-C

Problem behaviors were strongly accounted for by both sleep and circadian rhythm disturbances. Particularly, the longest continuous sleep episode is a novel clinically meaningful sleep parameter to consider, especially in children with severe sleep disorders

Assessment of both sleep and circadian rhythms with actigraphy is a strength because it is an objective measure, but also it is a limitation due to the fact that it remains a proxy of the circadian timing system. Another limitation: the lack of a control group

Jovevska et al. [21]

ASD group (N = 297, mean age = 34.36 ± 15.24) and comparison group (N = 233, mean age = 33.01 ± 15.53)

PSQI to examine sleep quality, SoL, total night sleep, and sleep efficiency. Other predictors of sleep quality: autistic traits, mental health condition, medication, employment, and sex

Autistic adolescents and adults, particularly females, remain vulnerable to sleep problems. Times where the risk is highest are early and middle adulthood

Limitations: sleep was measured using a self-report, retrospective questionnaire; the study was cross-sectional and so long-term trends for sleep quality in the context of chronicity or aging are not captured; autistic adults with intellectual disability were not included; about half the autistic sample was female, but the generally accepted male:female ratio in autism spectrum disorder is 4:1; the sample consisted of volunteers who responded to an advertisement, and not all participants had responses for all variables

Strengths: a large part of ASD population included (18–80 years old); large sample size, including an age-matched comparison group; the first study which examines male–female differences in sleep in autistic adolescents and adults