Identifying Unique Subgroups of Emotional and Behavioral Presentations in a Large Inpatient and Community Sample of Autistic Youth.
Safaa Eldeeb, Jessie Northrup, Caitlin Conner and 4 others
Four distinct emotional-behavioral profiles emerged in 1,311 autistic youth: a severe all-domain group (22%), a reactivity-dominant group (41%), a moderate internalizing group (18%), and a low-symptom group (19%).
The most severe group was 80% inpatient and had the lowest incomes.
Key findings
1Cluster analysis identified four distinct subgroups — Global High (22%, n=289), High-Reactivity Dominant (41%, n=539), Moderate-Internalizing Dominant (18%, n=233), and Global Low (19%, n=250) — and all four differed significantly from one another on every measured subscale (internalizing, externalizing, reactivity, dysphoria).
2In the Global High subgroup, 100% exceeded the clinical cutoff for emotional reactivity, 66% for dysphoria, 91% for externalizing problems, and 69% for internalizing problems; about 80% of this group came from the inpatient sample.
3The survey items that most strongly determined subgroup membership were those capturing intense emotional reactions (explosive outbursts, hard to calm, mood swings) and aggressive behaviors (destroying things, threatening people, disobedient at home), not internalizing or dysphoria items.
Still to come
How it was doneWhat they found
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
All data came from caregiver reports; the autistic youth's own experience was not captured. Participants with incomplete item-level data were excluded, which may reduce representativeness (the full inpatient sample was 1,356 but only 446 had complete CBCL data). The sample is demographically limited: 85.7% white, 92% non-Hispanic/non-Latino, 73% married parents, over 40% with household income above $81,000. The CBCL was normed on a general population; its cutoffs may not map cleanly onto autistic youth, and some items may have elevated base rates in autism for reasons unrelated to psychopathology. The study is cross-sectional; it identifies patterns at one time point and does not track how these profiles change or respond to treatment. Verbal ability was measured approximately (ADOS-2 module or parent report), and other communication modes (gestures, play) were not assessed. No comprehensive medical or psychiatric histories were available for the youth themselves. The clusters have not been validated in a separate sample, and the authors note that future research is needed to determine whether these subgroups actually respond differently to different interventions.
Declared interests
Funded by NIH (R01HD079512), NIMH (T32 MH018951, K23MH127420), Simons Foundation Autism Research Initiative (SFARI #296318), Nancy Lurie Marks Family Foundation, and PCORI. The authors declare no conflicts of interest. No industry sponsor funded, designed, or wrote the study.
The easy way to misread this
Do not read the four subgroups as validated diagnostic categories or as proof that the specific interventions the authors list (DBT, CBT, parent training, behavioral activation) are proven to work for each subgroup. The paper is a descriptive cluster analysis: it shows that four groups exist in this sample and differ on scores, but it does not test whether any treatment produces different outcomes in one group versus another. The authors themselves state that future research is needed to determine whether these subgroups require different approaches or respond differentially to different interventions.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →