Bilingual children who stutter: Convergence, gaps, and directions for research (Choo & Smith, 2020).
This systematic review provides updated information on the literature in regard to bilingual children who stutter as the last systematic review occurred in 2011.
Articles of interest were searched for on Scopus, Science Direct, PubMed, Google Scholar, and EricEbsco according to PRISMA guidelines; without a limitation on year published. Terms used during searches included stuttering, stutter, child, children, multilingualism, and bilingualism. Inclusionary terms required addressing children who stutter, children who are bilingual who stutter, and empirical articles published in peer-reviewed journals. Exclusionary items included addressing irrelevant or ambiguously related to stuttering, only adults or monolinguals, and not published in English. A total of 50 articles qualified for this review, of which 39 were new since the 2011 systematic review and were overall categorized into the following categories: epidemiology, profile of the bilingual child, treatment and assessment, and culture.
The articles categorized into epidemiology supported a difference in prevalence of stuttering between sexes, especially for older children (i.e. between the ages of 6 to 21 years, males demonstrated a higher ratio of stuttering as compared to females). Additionally, data reviewed in regard to age of onset also supported a difference between sexes (i.e. females were found to have an earlier age of onset as compared to males). Positive familial history of stuttering and sexually dimorphic patterns for recovery were also found to mirror monolingual trends. The articles used to build a profile of bilingual children who stutter supported the presence of stuttering, a combination of stuttering-like disfluencies and typical or other disfluencies, across languages. Data on comorbidity, developmental domains, and stuttering severity were inconclusive or not robust. The articles categorized into treatment and assessment supported the challenging nature of identifying and treating stuttering in bilingual children. There is data to support the use of the Stuttering Severity Instrument with bilingual English language learners, but diagnostic agreement amongst monolingual and bilingual clinicians was poor (e.g. “Agreement was achieved in only 50% of the cases…” pp. 105741). Additionally, behavior treatment programs were found to commonly report relapses with one program reporting 80% of the treated children experiencing relapse 21 months after treatment. Transference of treatment outcomes in the non-dominant language to the heritage language were found in several studies. Lastly, the articles addressing cultural impacts supported a complex interaction between cultural values and norms on attitudes, expectations, and experiences of culturally diverse children who stutter. This highlights the need for culturally responsive practices during the assessment and treatment process.
Limitations of the articles reviewed include the absence of a standardized definition of bilingualism utilized across studies and presence of underpowered populations used. Additionally, representation of languages across studies was limited as most articles focused on Spanish-English speakers. Therefore, the findings in this systematic review are preliminary as much research is needed to fill in the current knowledge gaps concerning bilingualism and stuttering.
An introductory examination of speech disfluencies in Spanish-English bilingual children who do and do not stutter during narratives (Rincon, Johnson, & Byrd, 2020).
This article studies the frequency and type of stuttering-like and nonstuttering-like disfluencies found in narrative samples elicited in Spanish and English with Spanish-English children who stutter as compared to Spanish-English children who do not stutter.
Participants used in this study consisted of six Spanish-English (SE) speaking children who were evenly split into groups of children who stutter (CWS) and children who do not stutter (CWNS); age range of participants was 5-7;5 years; months and all were recruited from the Houston, Texas area. Inclusion criteria for participants were the presence of at least 20% input and output in both languages, clinician perceptual judgment and/or caregiver concern of stuttering, and no reported concerns in regard to speech, language and hearing skills.
Narrative samples were elicited in story retell form from Frog Where Are You? in English and Spanish for 15-20 minutes with a total of 300 words transcribed using Systematic Analysis of Language Transcripts (SALT) software. SALT software calculated mean length of utterance (MLU) for words demonstrated a range from 4.26-9.43 for SE-speaking CWS. Additionally, stuttering-like disfluencies (SLD) per words spoken demonstrated a range from 3.00-7.00% in English and 3.33-10.00% in Spanish for this group. Contrastively, MLU for words demonstrated a range from 4.33-8.07 and SLD per words spoken demonstrated a range from 2.00-4.33% in English and 2.00-3.33% in Spanish for SE-speaking CWNS. Data from these groups supported findings in other studies. Recommendations include assessing speech disfluencies in both languages with the understanding that SE children will present with higher frequencies of typical disfluencies in addition to SLD (e.g. audible sound prolongation, inaudible sound prolongation [blocks]). Therefore, relying on frequency criteria established with monolingual English-speaking CWS should be used with caution as SE-speaking CWNS vary in meeting all, one, or none of the frequency criteria.
Limitations for this study include its underpowered pool of participants, lack of statistical analyses of individual differences (e.g. across groups for age, gender, language dominance, etc.), and the reliance on perceptual judgment to diagnose stuttering. Additionally, cognitive and affective measurements were not considered, speech disfluency coding and language transcribing reliability measurements were unconventional, and results are only preliminary for SE-speaking children.
Classifying disfluencies in preschool- and school-age Spanish-English bilinguals who do not stutter: An exploratory study (Rojas & Irani, 2020).
This preliminary study focused on identifying the speech disfluencies and language skills of Spanish-English speaking children who do not stutter. Additionally, this study aimed to identify differences in speech disfluencies of typically developing bilingual children in relation to language, mazing, and demographic variables.
29 SE-speaking children in prekindergarten through fourth grade (4.9-10.1 years in age) participated in this study. Language history and function were obtained via a survey completed by the caregivers. Language proficiency was screened with the Bilingual English-Spanish Oral Screener for children in prekindergarten through first grade and caregiver report of special education or concern for children in second through fourth grade. Narrative retell language sample analyses were also conducted in both languages for all children. The resulting language use across participants ranged from primarily Spanish to relatively equal use of Spanish and English in the home; all participants attended an English language immersion program. Inclusionary criteria included an absence of academic, speech, language, or stuttering history or concerns; passing the BESOS screening and ability to provide a typical narrative language sample for both languages as compared to SALT 18 software.
Results confirm previous findings in regard to higher percentages of mazing behaviors and normal disfluencies in Spanish relative to English. Consequently, clinicians are cautioned about using the stuttering criteria established for monolingual English children as the typically developing children in this study met or exceeded an overall 3% normal disfluency and stuttering-like disfluency frequency in their samples. Recommendations align with other findings in regard to obtaining narrative retell language samples in both languages, preferably with language sample analysis that can calculate the percentage of mazed words as this language measure yielded significant results (e.g. >10% in both languages for all children across all grades) and percentage of normal disfluencies. Additionally, clinical judgment should be exercised as the percentage of mazed words can slightly inflate frequency count as compared to traditional measures of stuttering frequency.
Limitations include a small n for each represented grade, lack of a language screener used for participants in grade 2-4, the use of grade matching to the SALT 18 database, and lack of formal and informal testing to exclude bilingual children who stutter from the participant pool. Furthermore, inclusion of more participants who are primarily dominant in Spanish or primarily dominant in English was also identified as a limitation.
Assessing bilingual children: Are their disfluencies indicative of stuttering or the by-product of navigating two languages? (Byrd, 2018).
This tutorial reviews best practices for assessing bilingual children referred for stuttering concerns according to the literature available in 2018.
Recommendations for best practices in this tutorial echo the systematic review and other articles annotated for this blog post. Use of frequency criteria established with monolingual English-speaking children who stutter with bilingual children can lead to false-positive identification, especially for children lacking language proficiency in their heritage and non-heritage languages. Additionally, perceptual judgments of stuttering are problematic even for bilingual clinicians matched to children for heritage language. Contrastively, caregiver concern, which is derived from perceptual judgments, have been demonstrated to be a reliable indicator for further evaluation. Differential diagnosis of a bilingual child who demonstrates disfluency types and frequencies associated with mazes from a bilingual child who demonstrates disfluency types and frequencies associated with stuttering (e.g. dysrhythmic phonation, atypical tension in typical disfluency behaviors, audible sound prolongation, inaudible sound prolongation [blocks]) is critical to preventing over- and under- identification of stuttering. Furthermore, there is insufficient data to support bilingualism as a risk factor for the onset and development of stuttering; “...deferring exposure to another language until ‘the critical period’ (as argued by some) for second-language learning has passed may compromise the child’s ability to acquire that language.” pp. 326-327. Lastly, gathering language samples and information in regard to language history, function, and proficiency for both languages are recommended best practices.
Limitations are congruent with other articles in that the literature in regard to bilingualism and stuttering is severely lacking, typically underpowered, absent of a definition for bilingualism, and absent of the affective as well as cognitive factors that are accepted as characteristics of persistent stuttering.
References
Byrd, C. T. (2018). Assessing bilingual children: Are their disfluencies indicative of stuttering or the by-product of navigating two languages? Seminars in Speech and Language, 39(4), 324-332. DOI: https://doi.org/10.1055/s-0038-1667161. Choo A. I., & Smith S. (2020). Bilingual children who stutter: Convergence, gaps, and directions for research. Journal of Fluency Disorders, 63. DOI: https://doi.org/10.1016/j.fludis.2019.105741. Rincon, C., Johnson, K., & Byrd, C. T. (2020). An introductory examination of speech disfluencies in Spanish-English bilingual children who do and do not stutter during narratives. Perspectives of the ASHA Special Interest Groups, 5(1), 131-141. DOI: https://doi.org/10.1044/2019_PERSP-19-00040. Rojas, R. & Irani, F. (2020). Classifying disfluencies in preschool- and school-age Spanish-English bilinguals who do not stutter: An exploratory study. Perspectives of the ASHA Special Interest Groups, 5, 119-130. DOI: https://doi.org/10.1044/2019_PERSP-19-00164.
