This study aimed to summarize the evidence on sleep alterations in medication-naïve children and adolescents with autism spectrum disorder (ASD).

We systematically searched PubMed/Medline, Embase and Web of Science databases from inception through March 22, 2021. This study was registered with PROSPERO (CRD42021243881). Any observational study was included that enrolled medication-naïve children and adolescents with ASD and compared objective (actigraphy and polysomnography) or subjective sleep parameters with typically developing (TD) counterparts. We extracted relevant data such as the study design and outcome measures. The methodological quality was assessed through the Newcastle-Ottawa Scale (NOS). A meta-analysis was carried out using the random-effects model by pooling effect sizes as Hedges’ g. To assess publication bias, Egger’s test and p-curve analysis were done. A priori planned meta-regression and subgroup analysis were also performed to identify potential moderators.

Out of 4277 retrieved references, 16 studies were eligible with 981 ASD patients and 1220 TD individuals. The analysis of objective measures showed that medication-naïve ASD patients had significantly longer sleep latency (Hedges’ g 0.59; 95% confidence interval [95% CI] 0.26 to 0.92), reduced sleep efficiency (Hedges’ g −0.58; 95% CI −0.87 to −0.28), time in bed (Hedges’ g −0.64; 95% CI −1.02 to −0.26) and total sleep time (Hedges’ g −0.64; 95% CI −1.01 to −0.27). The analysis of subjective measures showed that they had more problems in daytime sleepiness (Hedges’ g 0.48; 95% CI 0.26 to 0.71), sleep latency (Hedges’ g 1.15; 95% CI 0.72 to 1.58), initiating and maintaining sleep (Hedges’ g 0.86; 95% CI 0.39 to 1.33) and sleep hyperhidrosis (Hedges’ g 0.48; 95% CI 0.29 to 0.66). Potential publication bias was detected for sleep latency, sleep period time and total sleep time measured by polysomnography. Some sleep alterations were moderated by age, sex and concurrent intellectual disability. The median NOS score was 8 (interquartile range 7.25–8.75).

We found that medication-naïve children and adolescents with ASD presented significantly more subjective and objective sleep alterations compared to TD and identified possible moderators of these differences. Future research requires an analysis of how these sleep alterations are linked to core symptom severity and comorbid behavioural problems, which would provide an integrated therapeutic intervention for ASD. However, our results should be interpreted in light of the potential publication bias.