OBJECTIVES: The localization of Clinical Outcome Assessment (COA) measures, particularly through Linguistic Validation, has traditionally been a discipline highly reliant on human activities. Recognizing the great potential for progress with current GenAI capabilities, we set out to streamline the Concept Elaboration process by automating it further with the use of GenAI. Our ultimate goal was to test this technology’s capability to improve efficiency and thereby save time and costs in the overall project schedule, while balancing the outputs for completeness and quality via human review.

METHODS: Following our initial literature review on the topic, we performed a qualitative analysis on the Concept Elaboration outputs of multiple COAs of different types and complexity levels generated by a secure GenAI engine. During the investigation, prompts were customized until a suitable output was obtained, in line with the latest COA industry standards and established practices.

RESULTS: Preliminary results of the analysis performed on our sample instruments suggest that, given the right tailor-made prompt, GenAI can successfully perform the extraction of key terminology and can even go a step further to define such concepts within the context of the therapeutic area and disease being evaluated.

CONCLUSIONS: GenAI offers valuable insights to support the Concept Elaboration step and, when customized with the right input, also provides a considerable efficiency improvement in the Linguistic Validation process. Furthermore, during our analysis, it became evident that the GenAI applications can be further extended, as the scope of Concept Elaboration is interdependent with Source Analysis and Translatability Assessment. Therefore, with GenAI support, they have great potential to be merged into one single activity, further condensing the steps needed to perform the full localization process of COAs.