Characterization and Clinical Impact of Repetitive Compulsive Behaviors in a Cohort of Psychiatrically Hospitalized Children with ASD

Background: Repetitive compulsive behaviors (RCBs), including stereotypies, self-injurious behaviors and compulsions, are a key clinical feature of autism spectrum disorder (ASD). Diverse RCBs have been reported for patients with different levels of IQ and verbal abilities, but scarce data are available for behaviorally dysregulated patients with ASD requiring hospitalization for acute stabilization.

Objectives: This study aims to characterize RCBs and assess their clinical impact for a sample of youth with ASD hospitalized due to acute behaviors and/or co-morbid psychiatric conditions.

Methods: Youth aged 4-20 years, with an ADOS-confirmed ASD and admitted to a specialized inpatient psychiatry unit, were prospectively enrolled in a six-site study examining patient phenotypes, including RCBs, as part of the Autism Inpatient Collection (AIC). Data regarding the occurrence and nature of RCBs, length of hospital stay, parent stress, and patient community-reported sleep were collected. Caregivers also reported on the following measures at admission: Social Communication Questionnaire (SCQ), Aberrant Behavior Checklist-Community (ABC-C), Self-Injury Subscale of the Repetitive Behaviors Scale-Revised (RBS-R), and the Parent Stress Index-4 –Short Form (PSI-4-SF). They also answered direct questions about patient sleep problems at home, including quality of sleep, awakenings and problems falling asleep. Questions that reflect RCBs from the SCQ (7 items), ABC-C (10 items) and RBS-R (8 items) were chosen and compiled to generate an RCBs composite. Linear regressions estimated the effect of the RCB composite on each of the following outcomes: length of hospital stay, community sleep indicators and parental stress, while controlling for confounders.

Results: Data from 164 patients with ASD who were hospitalized and had complete survey data were analyzed. The mean patient age was 12.8 years (SD=3.22). The sample was 23% female and 87% Caucasian. The RCB composite score was significantly associated with higher parental stress level (p=0.002), poor sleep in the community (p=0.01), more frequent awakenings in the community (p=0.04) and longer length of stay (p=0.007). RCB composite scores were then adjusted for IQ and parental income as appropriate, with results continuing to be significant for parental stress (p=0.001), poor community sleep (p=0.02), difficulty falling asleep in the community (p=0.02), and length of stay (p=0.05).

Conclusions:  RCBs are common and problematic among psychiatrically hospitalized youth with ASD. Results show that RCBs were correlated with clinical outcomes including dysregulated sleep patterns, longer hospitalizations and greater parental stress. These results are independent of the patient’s IQ and parental income level, suggesting that RCBs could be a specific target for intervention to improve overall outcome. While causal connections between RCBs sleep patterns are not yet evident, future research could investigate neurobiological connections between these two clinical phenomena.