A NOVEL APPROACH- COLLABORATION BETWEEN SPEECH THERAPISTS, CLINICIANS, AND TRANSLATION TEAMS TO GENERATE GLOBAL ASSESSMENTS FOR STROKE PATIENTS
Author(s)
Hayley Simpson-Finch, BA1, Shawn McKown, MA2, Naila Judge, MA3, Katie Miller, MA4, Alejandro Perez, MD5, Russell Tanenbaum, BA6, Veronica Pura, MA7, Mrunmayee Pitkar, MA8;
1IQVIA, Bloxham, United Kingdom, 2IQVIA, Tolland, CT, USA, 3IQVIA, Munford, TN, USA, 4IQVIA, Barcelona, Spain, 5IQVIA, Bogota, Colombia, 6IQVIA, Durham, NC, USA, 7IQVIA, New York, NY, USA, 8Shukra Solutions, Arlington Heights, IL, USA
1IQVIA, Bloxham, United Kingdom, 2IQVIA, Tolland, CT, USA, 3IQVIA, Munford, TN, USA, 4IQVIA, Barcelona, Spain, 5IQVIA, Bogota, Colombia, 6IQVIA, Durham, NC, USA, 7IQVIA, New York, NY, USA, 8Shukra Solutions, Arlington Heights, IL, USA
OBJECTIVES: The NIH Stroke Scale (NIHSS) is a clinician-administered instrument that includes patient-facing exercises measuring impact of a stroke on speech. These include sentences advancing in complexity for repetition and words testing the patient’s use of tongue and lips. Translating directly is not possible, rather equivalent wording is required that tests the same difficulties in articulation, replicating similar phonological characteristics while conveying consistent length, complexity and syllable use. Many languages do not use sounds that mimic the English use of tongue and teeth.
METHODS: Twenty NIHSS languages in a study were unavailable, requiring development across language families including Indo-European, Indo-Aryan and Austronesian, in Latin, Cyrillic, logographic and Indian scripts. As NIHSS includes speech elicitation stimuli including repetition sentences and word prompts, translations were adapted preserving functional equivalence rather than linguistic equivalence. For each language, specialist linguistic validation (LV) linguists proposed candidate stimuli and conducted iterative reviews with an in-country clinician and speech-language pathologist (SLP). Review criteria included phoneme availability, articulatory placement, syllable and word length, complexity progression, and feasibility of standardized clinical administration.
RESULTS: Finding equivalents in some languages for English phonemes and articulatory patterns was challenging, due to differences in alphabets, phoneme inventories and spoken sounds. For example, a source phrase ‘FIFTY-FIFTY’ uses the fricative /f/ not commonly found in Indian languages, requiring replacement. In Russian, the ‘th’ sound in ‘THANKS’ doesn’t exist, and no word replaces the articulation. The ‘f’ sound is similar, but the place of articulation is labiodental not dental. Linguistic exercises in the source testing patient’s speech could not be replicated in many languages; however, clinician and SLP review enabled selection of culturally and clinically appropriate alternatives with comparable functional intent.
CONCLUSIONS: The team comprised of the PM, LV linguists, clinician and SLP worked together for each language ensuring robust cross-linguistic adaptation of NIHSS speech stimuli despite phonological constraints.
METHODS: Twenty NIHSS languages in a study were unavailable, requiring development across language families including Indo-European, Indo-Aryan and Austronesian, in Latin, Cyrillic, logographic and Indian scripts. As NIHSS includes speech elicitation stimuli including repetition sentences and word prompts, translations were adapted preserving functional equivalence rather than linguistic equivalence. For each language, specialist linguistic validation (LV) linguists proposed candidate stimuli and conducted iterative reviews with an in-country clinician and speech-language pathologist (SLP). Review criteria included phoneme availability, articulatory placement, syllable and word length, complexity progression, and feasibility of standardized clinical administration.
RESULTS: Finding equivalents in some languages for English phonemes and articulatory patterns was challenging, due to differences in alphabets, phoneme inventories and spoken sounds. For example, a source phrase ‘FIFTY-FIFTY’ uses the fricative /f/ not commonly found in Indian languages, requiring replacement. In Russian, the ‘th’ sound in ‘THANKS’ doesn’t exist, and no word replaces the articulation. The ‘f’ sound is similar, but the place of articulation is labiodental not dental. Linguistic exercises in the source testing patient’s speech could not be replicated in many languages; however, clinician and SLP review enabled selection of culturally and clinically appropriate alternatives with comparable functional intent.
CONCLUSIONS: The team comprised of the PM, LV linguists, clinician and SLP worked together for each language ensuring robust cross-linguistic adaptation of NIHSS speech stimuli despite phonological constraints.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
CO163
Topic
Clinical Outcomes
Topic Subcategory
Clinical Outcomes Assessment
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Cardiovascular Disorders (including MI, Stroke, Circulatory)