Presentation (CTI)

OBJECTIVES: Linguistic validation (LV) aims to ensure well-comprehended, culturally relevant COA translations for use in global clinical trials. An important LV role is identifying solutions for difficult-to-understand or translate terminology. The terms “ashamed”, “self-conscious”, and “embarrassed” often co-occur within COAs and even individual items; however, cognitive interview feedback suggests these terms overlap conceptually and in translation, with potential implications for patient comprehension and data poolability.
METHODS: Translators (n=81) representing 38 countries and 40 languages were surveyed on the terms “ashamed”, “self-conscious”, and “embarrassed”. Questions focused on their language-specific meaning, conceptual overlap, familiarity, frequency, and difficulty of translation and interpretation.
RESULTS: Linguists provided 51 unique translations for “ashamed”, 59 for “self-conscious”, and 65 for “embarrassed”. 32/81 linguists representing 22 languages reported conceptual overlap between the terms, greatest for “embarrassed” and “ashamed” (13/81 identical translations; 62% said these are always or contextually conceptually identical in their language). “Self-conscious” was deemed the most difficult concept to translate (30% endorsing “difficult” or “very difficult”), followed by “embarrassed” (13%) and “self-conscious” (11%). 89% and 83% of linguists indicated translations for “ashamed” and “embarrassed” respectively would be familiar to patients, falling to 60% for “self-conscious”. Free responses confirmed “self-conscious” as the most difficult term conceptually and for translation (N=15), and emphasized context dependence of meanings / translations and need for elaboration of all terms (N=10). Thematic analysis (TA) of linguist provided definitions revealed 5 dominant themes for each term, while confirming 5/10 conceptually overlapping areas for “ashamed” and “embarrassed”. TA results were used to generate optimized term definitions for use as PRO elaborations.
CONCLUSIONS: Survey results reveal conceptual overlap between the target terms, with significant cultural and geographic variation in familiarity and translatability. These data necessitate careful consideration of these terms’ use contexts within COAs, and, ideally, elaboration / separation of the concepts to avoid data validity issues.