1044 Relationship Between Core Autism Symptoms and Sleep Disturbances in Youth with Autism: A Latent Class Analysis
Kristina Lenker,Susan Calhoun,Julio Fernandez-Mendoza,Susan Mayes
DOI: https://doi.org/10.1093/sleep/zsae067.01044
IF: 6.313
2024-04-20
SLEEP
Abstract:Abstract Introduction Previous studies have used cluster analysis to clarify diagnostic heterogeneity of autism, but have been limited to identifying subgroups of autism on the basis of core symptoms rather than sleep problems. The present study examined the relationship between core autism symptoms and sleep problems in children and adolescents with autism spectrum disorder (ASD). Methods 1466 patients (1–18y, M=6.4±3.6; IQs of 8–146, M=88.9±27.25; 81.2% male, 89.0% white) diagnosed with ASD. Dimension reduction via principal component analysis (PCA) was performed on the 10 sleep items from the Pediatric Behavior Scale. Latent class analyses (LCA) was used to determine phenotypes characterized by core ASD symptoms (social interaction, perseveration, somatosensory disturbance, atypical communication/ development, mood, and selective attention/safety awareness), and sleep dimensions accounting for age, gender, IQ, and medication use. Results Three clusters with distinguishable sleep factors were retained from the PCA, with eigenvalues >1, with adequate goodness-of-fit (TLI=0.91;RMSEA=0.075); disturbed sleep (trouble falling asleep, restless sleep, wakes often, nightmares, and talks/walks/or cries in sleep), insufficient sleep (sleeps less, wakes early), and, hypersomnolence (drowsy/sleepy, sluggish/slow-moving, sleeps more). LCA revealed a 3-class (AIC=29164.00;BIC=29594.00;sBIC=29334.00) and 4-class model (AIC=28986.00;BIC=29489.00;sBIC=29184.00), with adequate fit. Based on IC measures, the 4-class model yielded the best model fit, with acceptable entropy (0.856). Based on conceptual grounds and considering all model fit statistics, the 4-class model was chosen. Using Class 3 (N=708;50.7%) as the reference group, Class 1 (N=71;5.1%) was categorized by more problems with social interactions, disturbed sleep, hypersomnolence, older age and increased medication use. Class 2 (N=367;26.3%) was categorized by less severe autism symptoms, particularly problems with selective attention/safety awareness, overall sleep problems, older age and higher IQ. Class 4 (N=251;18.0%) was categorized by more problems with perseveration and somatosensory disturbance, disturbed and insufficient sleep, younger age and increased medication use. Gender was not a significant covariate. Conclusion Using LCA we found four distinct patterns of core autism symptoms and sleep problems differing in terms of age, IQ, and medication use. This study provides strong evidence for phenotyping and targeting sleep as a standard part of therapeutic intervention in individuals with autism. Support (if any) Department of Health and Human Services, Health Resources and Services Administration, R4349152
neurosciences,clinical neurology