Sub-groups of spoken language and broader communication skills in a large heterogenous cohort of minimally verbal school-age children: evidence of discrepant profiles

BACKGROUND: Communication and language profiles in neurodevelopmental conditions are characterised by enormous phenotypic heterogeneity. We sought to identify subgroups of Minimally Verbal (MV) children in a school-age transdiagnostic sample. We hypothesised that a cluster with a discrepant profile (strong receptive but low speech production and expressive spoken language skills) would emerge. METHODS: We recruited MV children and their families (n = 193; mean age 7.6 years (sd: 2.5, range 4–13); 73% male). The sample varied in their adaptive skills and range of diagnoses (autism 77%, genetic syndrome 15%). Children took part in a play-based experimenter-child interaction designed to elicit communicative acts such as requesting or sharing attention. Parents completed questionnaires about their child’s developmental profile, communicative and adaptive skills. Additional in-person batteries probed children’s motor, imitative and receptive language skills. The multi-task, multi-informant communication-related variables were then entered into a pre-registered agglomerative hierarchical cluster analysis. RESULTS: Six distinct clusters emerged and were compared in relation to non-social autism symptoms, motor skills, adaptive skills and demographic measures. For four clusters, children’s receptive, expressive, adaptive and motor skills were fairly commensurate and could be described as very low-, low-, mid- or high-skill. Two further clusters described discrepant profiles of ability where speech and spoken language skills were disproportionately lower. Exploratory analyses revealed that children in different clusters differed in terms of their diagnostic profiles, use of Augmentative and Alternative Communication (AAC) and echolalia. LIMITATIONS: Whilst the inclusive, trait-based, transdiagnostic approach taken has high ecological validity, some measures employed were thus necessarily bespoke, adapted or non-normed and reported diagnoses did not undergo systematic validation. CONCLUSION: The hypothesised discrepant profile emerged whereby some MV children had stronger receptive than expressive skills, suggesting motor barriers to speech that necessitate tailored support.