Predictors of Social Communication Competence in a General Population of Children

Background:

Population estimates of prevalence may underestimate autistic characteristics of lesser severity, for two main reasons. First, cases are usually ascertained from secondary screening, based on an initial selection of children with severe and obvious symptoms. Mild or moderate deficits in social and communicative competence may be missed, especially if associated with marked comorbidity such as conduct problems and ADHD.  Second, methods of ascertainment are designed to maximise the distinction between valid cases of autism, defined according to conventional criteria, and to exclude conditions that do not quite reach diagnostic significance. In a recently reported study, the parent-report Social and Communication Disorders Checklist (SCDC) was administered to participants (n = 8,094) in the Avon Longitudinal Study of Parents and Children, when those children were approximately 92 months of age. We correlated impairment severity with independent clinical diagnoses of ASD, cognitive abilities, and teacher-rated maladaptive behavior. SCDC scores were continuously distributed in the general population; boys had mean scores 30% higher than girls. Social communicative deficits were associated with functional impairment at school, especially in the domains of hyperactivity and conduct disorders.

Objectives:

We aimed to ascertain predictors of social communication competence in a 24 month follow-up study of the original cohort,

Methods:

In this further study, the SCDC was administered again, to 6520 (80.6%) of the original sample, when they were approximately 116 months of age. Predictors (measured at 92 months) included: i) cognitive abilities; ii) teacher ratings of maladaptive behaviour on the Strengths and Difficulties questionnaire. Independent clinical identification of cases of ASD in the ALSPAC sample was obtained by 11 years of age.
Results:

Mean SCDC scores for children subsequently identified as having an autistic disorder (prevalence 0.69%) were almost identical at both time points (14.7/24). The strongest predictors of being in the highest decile of the distribution of SCDC scores at 116 months (independent of prior SCDC score) were: teacher rated hyperactivity, conduct disorder and peer problems at 92 months (all p<0.001), independent of verbal IQ and gender.

Conclusions:

We concluded from our earlier cross-sectional study that social and communicative deficits are of prognostic significance, in terms of behavioural adjustment at school, for both boys and girls. These new findings indicate the converse is also true. Children presenting with conduct problems at school in early adolescence may have undetected social communication difficulties.