Parent Reported Development and Withdrawal Informs Differential Diagnosis of Autism Spectrum Disorder Versus Developmental Language Disorder in Children Under 6

Background:  Efforts to improve rates of identification of Autism Spectrum Disorder (ASD) at younger ages have largely focused on differentiating children with ASD from neurotypicals children. In clinical practice, among children under age 6 presenting with concern for ASD who are not ultimately diagnosed with ASD, one of the most common diagnoses is Developmental Language Disorder (DLD). Easily administered parent-report measures may be helpful in facilitating differential diagnosis of ASD versus DLD. However, no measures for this specific purpose are currently available, and it remains unknown whether existing measures may be helpful in this regard.

Objectives:  The current study aimed to identify aspects of parent-rated development and emotional/behavioral symptoms that may facilitate differentiation of ASD (with or without comorbid DLD) from DLD (without ASD) among children under age 6 presenting with concern for ASD.

Methods:  Medical record review was conducted for 90 children ages 19 months to 5 years, 5 months (M_age = 3 years, 4 months) who were evaluated within a multidisciplinary autism evaluation clinic within the University of Michigan Health System Department of Neurology. Of these, 54 children were diagnosed with ASD (with or without DLD) and 28 were diagnosed with DLD (without ASD). Record review included results of parent report questionnaires (Child Development Inventory [CDI], Child Behavior Checklist [CBCL], and Social Communication Questionnaire [SCQ]). Mann-Whitney U tests were conducted to examine group differences in CBCL and SCQ standardized scores. For CDI, difference scores were calculated by subtracting age equivalence scores from chronological age. Chi square tests were conducted to determine group differences in the probability of falling within the delayed range (Z score ≤ -1.5) on the CDI.

Results:  As expected, SCQ scores were significantly higher for children with ASD than DLD (U = 389, p < .005). On the CDI, children with ASD had greater difference scores than DLD (U = 455, p < .05) and were more likely than DLD to be classified as delayed in general development (χ² = 7.8, p < .05). Specifically, children with ASD were more likely to be classified as impaired in the domains of social development (χ² = 7.05, p < .05) and language comprehension (χ² = 5.22, p < 0.05). Additionally, difference scores for children with ASD were significantly greater than DLD for both receptive (U = 438, p < .05) and expressive language (U = 455, p < .05). Children with ASD were also rated as significantly more withdrawn on the CBCL than DLD (U = 418, p < .05).

Conclusions:  Parent ratings on the withdrawal subscale of the CBCL and social, receptive and expressive language subscales of the CDI may add useful information in the differential diagnosis of ASD versus DLD among children under age 6 presenting with concern for ASD. Development of a parent-report measure including items indexing social and language development as well as withdrawal may provide greater clinical utility in this regard.