Predictors of Inpatient Psychiatric Hospitalization for Children and Adolescents with Autism Spectrum Disorder.

The study was cross-sectional, so it shows associations with hospitalization but cannot show that any factor caused hospitalization. The hospitalized participants were assessed during their hospital stay, so mood, sleep, and social-affect scores may reflect the acute crisis rather than a stable risk profile. The sample was relatively small compared with administrative database studies, and the authors say the findings need validation in larger datasets. Psychiatric conditions were measured differently in the two groups, caregiver report for the never-hospitalized group and clinician report for the hospitalized group, and not by structured interview. The ADOS-2 social-affect severity score may be partly influenced by the hospitalized group's psychiatric presentation at the time of testing. Intellectual disability and nonverbal status were more common in the hospitalized group but were not significant independent predictors, so they cannot be treated as separate risk markers in this model. Several services and supports that could affect hospitalization risk, such as outpatient psychiatric care, behavioral services, school setting, day programming, and other family services, were not measured.

The authors declared no conflicts of interest.

Do not read mood disorder, sleep problems, social-affect severity, adaptive functioning, or caregiver marital status as proven causes of hospitalization. The study was cross-sectional, and hospitalized participants were assessed during their inpatient stay, so these factors may reflect the acute crisis rather than predict it beforehand.