A large-scale study reveals why some autistic children remain minimally verbal despite intensive early intervention, and what factors may shape more personalized support.
Study: Proportion and Profile of Autistic Children Not Acquiring Spoken Language Despite Receiving Evidence-Based Early Interventions. Image credit: Prostock-studio/Shutterstock.com
A recent study in the Journal of Clinical Child and Adolescent Psychology investigated the proportion and profile of children on the autism spectrum who remain minimally verbal despite receiving early interventions targeting spoken language.
The research examines patterns of child characteristics and intervention features that are associated with which minimally verbal autistic children are more or less likely to develop speech after early intervention.
Verbal communication in autistic children
Approximately one in three school-aged children diagnosed with autism spectrum disorder (ASD), which is commonly referred to as autism, do not communicate using spoken language. Although spoken language differences in autism usually vary, children with limited spoken language by school age may represent a distinct, clinically meaningful subgroup.
As children who do not acquire spoken language beyond a minimal level during preschool years are at an increased risk for lifelong disabilities, spoken language has been included as a key target in autism intervention since pioneering work in the field. Even though various early interventions have shown positive effects on spoken language, a significant number of autistic children remain unresponsive to these strategies.
To date, few large-scale studies have systematically examined this variable clinical outcome of early interventions. Limited knowledge of the proportion and profile of children who do not acquire spoken language despite receiving targeted interventions is a major barrier to progress in the field. Identifying factors associated with poorer clinical outcomes in some autistic children is critical for individualizing interventions and optimizing outcomes while setting realistic expectations.
Large multi-site dataset tracks language outcomes after intervention
The current large-scale retrospective study investigated the proportions and profiles of autistic children who did not acquire spoken language despite receiving targeted interventions. An aggregate dataset was assembled comprising child-level data for 707 pre-schoolers on the autism spectrum who had received well-characterized, evidence-based early interventions.
All participants were between 15 and 68 months old at baseline, diagnosed with autism, underwent standardized spoken language tests at pre- and post-intervention, and received evidence-based intervention for 6–24 months from trained clinicians or educators who delivered 10+ hours per week of direct intervention.
Participants received one of four intervention types: Early Start Denver Model (ESDM), a comprehensive intervention facilitating skill development through child-led joint activity routines; Naturalistic Developmental Behavioral Interventions (NDBIs), which blend teaching episodes within naturalistic contexts; Early Intensive Behavioral Intervention (EIBI), using structured, adult-directed instruction such as Discrete Trial Training; or TEACCH, focused on environmental accommodations and visual supports for autistic learners.
Factors Predicting Language Progress in Autistic Children
The study's primary analysis focused on 470 children. At the start, 293 displayed minimal language skills (Phase 1), and 177 could use single words (Phase 2). About 233 children progressed to combining words, while 237 children did not.
The type of intervention children received did not affect their progress. To understand which factors mattered, researchers analyzed multiple variables, including intervention type, the number of hours per week children received, the duration of the intervention, the child’s age at the start, sex, autism severity, adaptive functioning, cognitive abilities, and motor imitation skills.
Because many variables were interrelated, researchers used a statistical technique to group them into a smaller number of independent underlying factors, allowing clearer interpretation of which patterns were associated with language outcomes. The analysis identified five main factors that captured 90 % of the information. The first factor combined child characteristics with intervention details, such as age at start, adaptive skills, cognitive abilities, duration, and intensity. The second factor captured additional patterns in the data.
The mixed-effects logistic regression included sex, intervention type, and the first five derived factors as fixed effects, with site as a random effect. Results showed that neither sex nor intervention type predicted outcomes. However, the first two factors were significantly associated with children's language skill development.
Children who did not progress generally started intervention later and received shorter-duration services, although at higher weekly intensity. More importantly, these children had greater autism severity, lower adaptive functioning, lower cognitive abilities, and weaker motor imitation skills at the beginning of treatment.
Among 293 children who were non-speaking at the start, about two-thirds (195 children) learned to use single words or more, while one-third (98 children) remained non-speaking. Again, the type of intervention did not affect outcomes. Researchers analyzed what factors were associated with progress.
Researchers found that neither sex nor intervention type predicted whether children learned to speak. The key factors were motor imitation skills, autism severity, and age at the start of intervention. Children who remained non-speaking had weaker imitation skills, more severe autism symptoms, and, on average, began intervention at a slightly younger age than those who progressed, a pattern the authors caution should not be interpreted as evidence that earlier intervention is ineffective, but rather may reflect developmental readiness at very young ages.
When looking only at younger children (ages 48 months and younger), the results were similar to the main analysis. Children who did not progress had more severe autism, lower adaptive functioning and cognitive abilities, weaker imitation skills, and started intervention later than those who advanced.
Findings highlight need for individualized early intervention strategies
About one-third of children with limited or no spoken language did not progress despite receiving evidence-based interventions. Intervention type did not predict outcomes; however, baseline child characteristics, including motor imitation skills, autism severity, cognitive abilities, and adaptive functioning, were most strongly associated with language progress.
The current study findings suggest that early interventions may benefit from greater individualization. Future research should test whether earlier access, longer intervention duration, or more targeted support for prelinguistic skills such as imitation and social communication can improve outcomes for children who show profiles associated with limited spoken-language progress, alongside timely support for alternative and augmentative communication approaches when spoken-language gains are limited.
Download your PDF copy now!
Reproductive history is associated with women's lifespan and biological aging