Assessment and differential diagnosis in childhood apraxia of speech

Abstract: Introduction and aims: Childhood Apraxia of Speech (CAS) is a disorder of speech motor planning and programming affecting the ability to transform the linguistic code into movements needed for intelligible speech. CAS is difficult to differential diagnose from other speech sound disorders (SSDs) and the diagnostic procedure involves many decisions based on outcomes from assessments before arriving at the appropriate diagnosis/es. Some children with CAS also have coexisting functional difficulties. Consequently, it is important for the clinician to describe and be aware of these difficulties. The overall aim of this project was to investigate speech performance and speech characteristics in 3-to-9-year-old children with typical and disordered speech, to facilitate more confident diagnostic decisions of Swedish-speaking children with CAS. Material and methods: The project consists of four studies with a total of 139 participants in the age range 3:1 to 8:10 years (37 to 106 months). In Study I, a newly developed Swedish test for motor speech skills, Dynamisk motorisk talbedömning (DYMTA), adapted from Dynamic Evaluation of Motor Speech Skills (DEMSS) was described. 94 children (37 to 106 months old) with typical speech- and language development participated (45 boys/49 girls), including 17 bilingual children, and were assessed with DYMTA (DYMTA-A and DYMTA-B) as well as with a comprehensive test battery of language, speech, and orofacial function. In Study II, reliability and validity of DYMTA were analyzed to evaluate its ability to discriminate between children with a diagnosis of CAS from children with other SSDs. 45 children (31 boys/14 girls, 40 to 106 months) were assessed with DYMTA as well as with a comprehensive test battery covering language, speech, and orofacial function. In Study III, the number and types of characteristics associated with CAS were investigated in two different speech samples for each child, one using dynamic assessment (DA) and the other static assessment (SA). 33 children (22 boys/11 girls, 40 to 106 months) with a CAS diagnosis were included and a perceptual assessment of present and absent speech characteristics was performed. In Study IV, a parental questionnaire, specifically designed for this project asking about mental, sensory, motor, and voice functions, was given to caregivers of 33 children with CAS (the same group as in study III). An agglomerative clustering analysis was applied to the individual item-responses to identify groups of children with different patterns of parent-reported functional difficulties. Results: The main result in study I was that Swedish children with typical speech- and language development performed well on tasks in DYMTA already at the age of three. Outcome measures on some subscores and subtests were close to ceiling from five years, while others showed a protracted refinement into early school-age. In Study II, intrarater reliability was found to be strong for the total scores and subscores in both DYMTA-A and DYMTA-B. Interrater reliability (point-by-point and ICC) was strong for total score and all subscores except for the ICC value of the subscore Prosody. The validity estimates showed that the results on DYMTA have good potential to differentiate children with CAS from children with other SSDs. In Study III, the most prominent CAS characteristics apart from inconsistency in the dynamic task were deviant transitionary movements, prosody error, and vowel error, and in the naming task vowel errors, voicing errors and prosody errors. More CAS characteristics were evident in the DA sample (on average 6.8), than in SA sample (5.2). In Study IV the number of parent-reported functional difficulties ranged from 1 to 27 across the participants. The hierarchical cluster analysis identified four groups of children with different functional profiles: 1A) few coexisting difficulties, 1B) coexisting difficulties mainly in voice domains, 2A) coexisting difficulties in mental functions (e.g., attention) and 2B) coexisting difficulties in diverse functions, including motor functions. Conclusions: The findings from this thesis project support benefits of including several tasks and methods addressing speech motor skills in the diagnostic procedure of CAS. The preliminary evidence on validity and reliability of DYMTA showed acceptable results in separating children with CAS and other SSDs. Also, the investigation of detected speech characteristics associated with CAS supports current knowledge on the benefits of a dynamic assessment method, adding information on the child’s speech motor skills when cuing is provided. Apart from the results from assessments on speech performance itself, clinicians should also be aware of various functional profiles and coexisting difficulties evident in some children with CAS.