Scale Up Multilingualism in Health Emergency Learning: Developing an Automated Transcription and Translation Tool.

World Health Organization's (WHO) emergency learning platform OpenWHO provided by Hasso Plattner Institut (HPI) delivered online learning in real-time and in multiple languages during the COVID-19 pandemic. The challenge was to move from manual transcription and translation to automated to increase the speed and quantity of materials and languages available. TransPipe tool was introduced to facilitate this task. We describe the TransPipe development, analyze its functioning and report key results achieved. TransPipe successfully connects existing services and provides a suitable workflow to create and maintain video subtitles in different languages. By the end of 2022, the tool transcribed nearly 4,700 minutes of video content and translated 1,050,700 characters of video subtitles. Automated transcription and translation have enormous potential as a public health learning tool, allowing the near-simultaneous availability of video subtitles on OpenWHO in many languages, thus improving the usability of the learning materials in multiple languages for wider audiences.

Effects of face masks on language comprehension in bilingual children.

Due to the COVID-19 pandemic, many children receive language input through face coverings. The impact of face coverings for children's abilities to understand language remains unclear. Past research with monolingual children suggests that hearing words through surgical masks does not disrupt word recognition, but hearing words through transparent face shields proves more challenging. In this study, we investigated effects of different face coverings (surgical masks and transparent face shields) on language comprehension in bilingual children. Three-year-old English-Mandarin bilingual children (N = 28) heard familiar words in both English and Mandarin spoken through transparent face shields, surgical masks, and without masks. When tested in English, children recognized words presented without a mask and through a surgical mask, but did not recognize words presented with transparent face shields, replicating past findings with monolingual children. In contrast, when tested in Mandarin, children recognized words presented without a mask, through a surgical mask, and through a transparent face shield. Results are discussed in terms of specific properties of English and Mandarin that may elicit different effects for transparent face shields. Overall, the present findings suggest that face coverings, and in particular, surgical masks do not disrupt spoken word recognition in young bilingual children.

Characterizing Communicative Participation in Multilingual Jamaican Preschoolers.

PURPOSE: This study characterized communicative participation and related aspects of functional communication for Jamaican Creole (JC)-English-speaking preschoolers with and without functionally defined speech sound disorders (fSSDs). This study included parent reports and direct assessment measures from an existing corpus of baseline data collected prior to the COVID-19 pandemic. METHOD: The communicative participation of typically developing (TD; n = 226) bilingual JC-English-speaking preschoolers and those with fSSDs (n = 39) was documented using the Focus on the Outcomes of Communication Under Six (FOCUS). Functional speech intelligibility was recorded using the Intelligibility in Context Scale (ICS) in English and JC (ICS-JC). Objective measures of speech production were collected through direct child assessment in both languages and then transcribed and calculated for percent of consonants (PCC), vowels (PVC), and phonemes correct (PPC). Within-group relationships were explored using association testing, and differences between groups were explored through multivariate analyses. RESULTS: FOCUS scores and ICS and ICS-JC scores were found to be minimally to moderately related for Jamaican preschoolers in the TD group (r = .28-.34, p < .002) and strongly related in the fSSD group (r = .56-.60, p < .002). No relations were observed between the FOCUS scores and PCC/PVC/PPC in either language. There was a statistically significant difference between all FOCUS scores for Jamaican preschoolers in the TD and fSSD groups (p ≤ .002). CONCLUSIONS: These findings provide additional evidence for using the FOCUS beyond documenting change in communicative participation to support clinical decision-making in planning and developing speech-language interventions. This study also documents an important characterization of JC-English-speaking children with and without fSSDs, offering data on children's abilities that can be used in future comparisons of communicative participation and speech functioning observed during the COVID-19 pandemic and beyond.

SuperSpeech: Multilingual Speech and Language Maintenance Intervention for Vietnamese-Australian Children and Families via Telepractice.

PURPOSE: The aim of this pilot feasibility study was to evaluate the effectiveness of the group VietSpeech SuperSpeech program targeting speech skills and home language maintenance via telepractice. METHOD: In Stage 1, using a case-control design, 30 Vietnamese-English-speaking children were assessed in English and Vietnamese, and parents completed questionnaires about speech and language competency and practices. During Stage 2, children were allocated to intervention (n = 14) or control (n = 16) conditions. COVID-19 restrictions resulted in changes including nonrandom allocation. Online group intervention targeting speech, home language maintenance, and multilingualism as a superpower was delivered 1 hr/week for 8 weeks. For Stage 3, assessments were undertaken approximately 10 weeks after the pre-intervention assessment. RESULTS: Parents in the intervention group significantly increased encouragement of their children to speak Vietnamese. The intervention group significantly increased intelligibility in English. Growth of Vietnamese vocabulary was faster for the control group. There was a moderate effect of intervention for children's perception of being happy talking in Vietnamese and English. There was no significant mean change from pre- to post-intervention compared with the control group for measures of speech sound accuracy in Vietnamese or English, Vietnamese intelligibility, English vocabulary, or hours of Vietnamese spoken each week. CONCLUSIONS: This study presents preliminary evidence that this 8-hr online group program targeting speech skills and home language maintenance had some impact on Vietnamese-Australian children's speech and home language maintenance. Further research involving a randomized trial is warranted.

From a Distance: Comparison of In-Person and Virtual Assessments With Adult-Child Dyads From Linguistically Diverse Backgrounds.

PURPOSE: Our proof-of-concept study tested the feasibility of virtual testing using child assessments that were originally validated for in-person testing only. METHOD: Ten adult-child dyads were assigned to complete both in-person and virtual tests of language, cognition, and narratives. Child participants fell between the ages of 4 and 8 years; adult participants were speech-language clinicians or researchers with experience in administering child assessments. Half of child participants were Spanish-English bilinguals, and half were monolingual English speakers. RESULTS: Results showed similar performance across in-person and virtual modalities on all assessments. Recommendations for adapting, administering, and scoring virtual measures with linguistically diverse children are discussed. CONCLUSIONS: Although additional research on virtual assessment is needed, our results open opportunities for appropriate remote assessment, particularly for bilingual children, who may not have in-person access to speech-language pathology services.

Is Bilingual Receptive Vocabulary Assessment via Telepractice Comparable to Face-to-Face?

PURPOSE: This study investigated the effect of delivery method (face-to-face or telepractice), time, home language, and language ability on bilingual children's receptive vocabulary scores in Spanish and English. METHOD: Participants included bilingual children with (n = 32) and without (n = 57) developmental language disorders (DLD) that were assessed at 2 time points about 1 year apart. All children participated in face-to-face assessment at Time 1. At Time 2, 41 children were assessed face-to-face and 48 children were assessed using telepractice. RESULTS: Delivery method was not a significant predictor of receptive scores in either Spanish or English. Spanish and English receptive vocabulary increased over time in both children with and without DLD. Children with DLD had lower receptive vocabulary raw scores than children with typical development. Children who spoke English-only at home had significantly higher English receptive scores than children who spoke Spanish-only or both Spanish and English at home. CONCLUSIONS: Face-to-face and telepractice assessments seem to be comparable methods for the assessments of Spanish and English receptive skills. Spanish and English receptive skills increased over time in children with and without DLD. SUPPLEMENTAL MATERIAL: https://doi.org/10.23641/asha.17912297.

Parental Report via a Mobile App in the Context of Early Language Trajectories: StarWords Study Protocol.

Social sciences researchers emphasize that new technologies can overcome the limitations of small and homogenous samples. In research on early language development, which often uses parental reports, taking the testing online might be particularly compelling. Due to logistical limitations, previous studies on bilingual children have explored the language development trajectories in general (e.g., by including few and largely set apart timepoints), or focused on small, homogeneous samples. The present study protocol presents a new, on-going study which uses new technologies to collect longitudinal data continuously from parents of multilingual, bilingual, and monolingual children. Our primary aim is to establish the developmental trajectories in Polish-British English and Polish-Norwegian bilingual children and Polish monolingual children aged 0-3 years with the use of mobile and web-based applications. These tools allow parents to report their children's language development as it progresses, and allow us to characterize children's performance in each language (the age of reaching particular language milestones). The project's novelty rests on its use of mobile technologies to characterize the bilingual and monolingual developmental trajectory from the very first words to broader vocabulary and multiword combinations.

Birthing Experiences of Spanish Speakers During the COVID-19 Pandemic in NYC.

Patient-provider language discordance is known to negatively affect patient experiences. This study describes the birth experience during COVID-19 among monolingual Spanish and bilingual Spanish/English speakers. Qualitative videoconference interviews with 15 monolingual Spanish and 15 bilingual Spanish/English patients that gave birth during the COVID-19 pandemic at a NYC tertiary-care hospital. 93% of participants had a positive birthing experience. Common themes were quality of care, birth outcome, and supportive staff. 80% of patients lacked a support person postpartum. 27% of Spanish-only speakers felt that an interpreter should have been provided but was not, and 47% felt the experience would have been different if they spoke English. The patient birth experience is tied to birth outcomes and quality of care and remained positive during the beginning of the COVID-19 pandemic. Restrictions on support people during the intra- or postpartum impacted the birthing experience more than provider language discordance.

Multilingual Approach to COVID-19 Online Learning Response on OpenWHO.org.

In pursuit of equitable access to emergency-related knowledge, the World Health Organization (WHO) translates COVID-19 and other infectious disease courses into multiple languages on its open-access online learning platform OpenWHO.org. Languages spoken by vulnerable or underserved populations in low- and middle-income countries and in outbreak-prone and affected areas are prioritized. Accessing learning in preferred languages enhances uptake and comprehension. In this study, we assess and compare the initial enrollment levels and global reach of these multilingual courses. On average, OpenWHO's 38 COVID-19 courses have each been translated into 4.8 languages. The platform hosts courses in 55 different languages with 10.4 million words translated. The findings identify which available languages were most utilized for COVID-19 learning to inform course production and outreach strategies. Languages were used differently across geographic regions, calling for localized learning offerings. A streamlined multilingual publishing scheme, ensuring quick and effective delivery of diverse languages, is critical to achieving greater equity of access to knowledge.

Speak like a Native English Speaker or Be Judged: A Scoping Review.

This review aims to examine the discrimination and prejudices toward the accent of non-native English speakers and cyberbullying as the ripple effect of these negative consequences. Following Arksey and O'Malley's framework of conducting a scoping review, 60 studies from 2012 to 2021 were retrieved from the ERIC and Google Scholar databases. The studies were reviewed from two aspects: (1) psychological impact on speakers with a non-native English accent, (2) attitudes toward non-native English accents from the victim's and perpetrator's perspectives. The findings suggested that speaking with a non-native English accent drew negative cognitive, affective, and behavioral experiences. Biases toward non-native English accents were due to the general derogatory perception of an accent and the comprehensibility of speakers' accent and pronunciation. "Accent acceptability" can be inculcated at all levels of education, not only through multicultural education but also through the concerted effort of policy makers and practitioners to seriously address this social issue. Accent awareness can dispel unwarranted and undesirable judgements of non-native English accent speakers. Future studies should be conducted on the effects of social and mental health experiences, particularly of non-native ESL and EFL teachers, given that this may be the only profession required to teach "live" during the pandemic and thus be subjected to public praise or ridicule.

Challenges in Service Provision to Children and Families from Linguistically Diverse Backgrounds.

The purpose of this study was to examine and describe experiences and perceptions of speech-language pathologists (SLPs) serving children and families from culturally and linguistically diverse populations. Twenty SLPs were interviewed using an online audio platform regarding common practices, resources and supports, challenges, and communication. Participants' responses to the semi-structured questions were transcribed and analyzed to identify themes in experiences. Results suggested six overall themes including: considerations for assessment; cultural impact; linguistic access and barriers; professional preparedness; impact of COVID-19; and helpful tips and resources. The discussion includes recommendations and resources to address obstacles.

Multistage BiCross encoder for multilingual access to COVID-19 health information.

The Coronavirus (COVID-19) pandemic has led to a rapidly growing 'infodemic' of health information online. This has motivated the need for accurate semantic search and retrieval of reliable COVID-19 information across millions of documents, in multiple languages. To address this challenge, this paper proposes a novel high precision and high recall neural Multistage BiCross encoder approach. It is a sequential three-stage ranking pipeline which uses the Okapi BM25 retrieval algorithm and transformer-based bi-encoder and cross-encoder to effectively rank the documents with respect to the given query. We present experimental results from our participation in the Multilingual Information Access (MLIA) shared task on COVID-19 multilingual semantic search. The independently evaluated MLIA results validate our approach and demonstrate that it outperforms other state-of-the-art approaches according to nearly all evaluation metrics in cases of both monolingual and bilingual runs.

Evaluating Local Multilingual Health Care Information Environments on the Internet: A Pilot Study.

For foreign-born populations, difficulty in finding health care information in their primary language is a structural barrier to accessing timely health care. While such information may be available at a national level, it may not always be relevant or appropriate to the living situations of these people. Our objective was to explore the quality of online multilingual health information environments by pilot-testing a framework for assessing such information at the prefectural level in Japan. The framework consisted of five health care domains (health system, hospitals, emergency services, medical interpreters, and health insurance). Framework scores varied considerably among prefectures; many resources were machine-translated. These scores were significantly associated with foreign population proportion and the number of hospitals in each prefecture. Our multilingual health care information environment (MHCIE) framework provides a measure of health access inclusivity, which has not been quantified before. It is adaptable to other international contexts, but further validation is required.

Bridging the communication gap between radiographers and patients to improve chest radiography image acquisition: A multilingual solution in the COVID-19 pandemic.

INTRODUCTION: A chest X-ray (CXR), taken in full inspiration, is important to ensure pathology in the lungs will not be missed. To achieve this, effective communication on breathing instructions for patients is crucial. During the COVID-19 pandemic, radiographers in Sengkang General Hospital (SKH) were challenged when performing CXR for the patients whose native language is not English. Most of these patients were foreign workers living in the same dormitory which had formed the largest COVID-19 cluster in Singapore. These dormitory residents found it difficult to understand and adhere to breathing instructions, resulting in a suboptimal degree of inspiration when the CXRs were taken. This may ultimately affect the diagnostic value of the radiographs. This paper aims to share and evaluate how radiographers tackled this issue and continued to acquire fully-inspired CXR for the dormitory residents despite the language barrier. METHODS: Using a combination of online survey and retrospective analysis of the rejection rates of CXR done over the period of early April to early June, a team of radiographers evaluated the effectiveness of using audio recordings in managing the issue of not achieving a fully inspired CXR for patients due to language barrier. RESULTS: The rejection rate for CXR due to suboptimal inspiration decreased from 26% to 9% upon implementation of the audio recordings. 92.3% of the CXRs taken within this period also fulfilled the criteria of a fully-inspired CXR, as evidenced by having at least 9 posterior ribs seen above the right hemi-diaphragm. Survey results found a fairly balanced number of radiographers who agreed and disagreed that a fully-inspired CXR was achieved for most of their patients after utilisation of translation manuals and audio recordings. CONCLUSION: After the implementation of audio recordings, the decrease in rejection rate of CXR and an audit which demonstrated that CXR quality was upheld had proven that the radiographers successfully achieved fully-inspired CXR for suspected COVID-19 patients. This confirmed that using pre-recorded audio instructions was an efficient intervention albeit being a one-way communication, leads to more accurate imaging results, aligning with existing literature on communication experiences between radiographers and patients. Moreover, the decreased rejection rate of CXRs had increased department efficiency consequently reducing departmental expenses in the long run. IMPLICATIONS OF PRACTICE: Given that we have an ageing population and the vast majority of the elderly converse in their various dialects, positive feedback from radiographers presented opportunities to expand the translation manual and audio recordings to include local dialects. These can be seamlessly integrated in CXR and other procedures in the hospital setting. To ensure that the translations are culturally sensitive, attention should be paid to the translation process of instructions into other languages and local dialects by enlisting the help of native speakers.

openEHR Archetype Use and Reuse Within Multilingual Clinical Data Sets: Case Study.

BACKGROUND: Despite electronic health records being in existence for over 50 years, our ability to exchange health data remains frustratingly limited. Commonly used clinical content standards, and the information models that underpin them, are primarily related to health data exchange, and so are usually document- or message-focused. In contrast, over the past 12 years, the Clinical Models program at openEHR International has gradually established a governed, coordinated, and coherent ecosystem of clinical information models, known as openEHR archetypes. Each archetype is designed as a maximal data set for a universal use-case, intended for reuse across various health data sets, known as openEHR templates. To date, only anecdotal evidence has been available to indicate if the hypothesis of archetype reuse across templates is feasible and scalable. As a response to the COVID-19 pandemic, between February and July 2020, 7 openEHR templates were independently created to represent COVID-19-related data sets for symptom screening, confirmed infection reporting, clinical decision support, and research. Each of the templates prioritized reuse of existing use-case agnostic archetypes found in openEHR International's online Clinical Knowledge Manager tool as much as possible. This study is the first opportunity to investigate archetype reuse within a range of diverse, multilingual openEHR templates. OBJECTIVE: This study aims to investigate the use and reuse of openEHR archetypes across the 7 openEHR templates as an initial investigation about the reuse of information models across data sets used for a variety of clinical purposes. METHODS: Analysis of both the number of occurrences of archetypes and patterns of occurrence within 7 discrete templates was carried out at the archetype or clinical concept level. RESULTS: Across all 7 templates collectively, 203 instances of 58 unique archetypes were used. The most frequently used archetype occurred 24 times across 4 of the 7 templates. Total data points per template ranged from 40 to 179. Archetype instances per template ranged from 10 to 62. Unique archetype occurrences ranged from 10 to 28. Existing archetype reuse of use-case agnostic archetypes ranged from 40% to 90%. Total reuse of use-case agnostic archetypes ranged from 40% to 100%. CONCLUSIONS: Investigation of the amount of archetype reuse across the 7 openEHR templates in this initial study has demonstrated significant reuse of archetypes, even across unanticipated, novel modeling challenges and multilingual deployments. While the trigger for the development of each of these templates was the COVID-19 pandemic, the templates represented a variety of types of data sets: symptom screening, infection report, clinical decision support for diagnosis and treatment, and secondary use or research. The findings support the openEHR hypothesis that it is possible to create a shared, public library of standards-based, vendor-neutral clinical information models that can be reused across a diverse range of health data sets.