Poster #26 - Can Psycholinguistic Assessments Differentiate Children and Adolescents With DLD From ADHD?

• In Event: Poster Session 09

Abstract

Objective: Although developmental language disorder (DLD) and attention-deficit/hyperactivity disorder (ADHD) are two distinct conditions, the impact of their difficulties in social, academic, and even testing settings can make it difficult to differentiate them. The language difficulties that accompany ADHD can make it especially difficult to separate these individuals from those with non-comorbid DLD. As such, the current study sought to examine whether several well-established assessments could accurately distinguish children and adolescents with DLD from those with ADHD (combined or inattentive) and comorbid DLD + ADHD.

Method: Participants included clinically evaluated children and adolescents with ADHD combined (n = 148), ADHD inattentive (n = 192), DLD (n = 39), ADHD combined + DLD (n = 28), and ADHD inattentive + DLD (n = 34) ages 6-16 years. Receiver operating curves were generated to test the diagnostic power of the Clinical Evaluation of Language Fundamentals – Screening Test (CELFST; Wiig et al., 2013), the Comprehensive Test of Phonological Processing (nonword repetition subtest; Wagner et al., 2013), and the Test of Word Reading Efficiency (sight word and phonemic decoding subtests; Torgesen et al., 2012). Regression with cross-validation was then used to examine how the assessments could be used for classification of the two disorders.

Results: The presence of a language disorder (with or without ADHD) resulted in poorer performance across all tasks. Children and adolescents with ADHD (combined versus inattentive) only significantly differed in sight word efficiency, in favor of those with combined type. While reading efficiency measures were best at discriminating between the two types of ADHD, the CELFST was best at discriminating between pure and comorbid samples of ADHD and DLD. It was also the best predictor of diagnostic status, above and beyond IQ and vocabulary assessments. Cross validation methods further revealed that the models had good generality (root mean square error = .43), such that we predict the results should extend to other samples in predicting DLD status from this combination of tasks.

Conclusion: Our pattern of findings suggests that an additional diagnosis of ADHD in children with DLD does not compound language and literacy difficulties. Our findings diverge with previous studies showing high classification accuracy for nonword repetition tasks (Redmond et al., 2011) but provide evidence for continued use of the CELFST to identify potential language disorders. Future work should investigate which assessments are best at identifying different subtypes of ADHD and DLD as well as comorbid samples. Our findings suggest that assessments of reading efficiency might be a productive starting point. None of the groups could be clearly defined based on their reading or language abilities and thus, additional assessments are needed to identify ADHD and/or DLD status beyond those used here. If replicated, our findings provide a strong empirical basis for the continued use of the CELFST not only to identify DLDs, but to discriminate DLD from ADHD. Our findings also suggest tailored intervention programs will be required that target a child’s strengths and weaknesses relative to language outcomes among developmental language and attention disorders.

Group Authors

N/A

Authors

• Kaitlyn Mary Ann Parks, University of Western Ontario
  Presenting Author

• Marc Joanisse, University of Western Ontario