Weerasinghe, Kusal, Olawade, David ORCID: https://orcid.org/0000-0003-0188-9836, Teke, Jennifer, Msiska, Maines and Boussios, Stergios ORCID: https://orcid.org/0000-0002-2512-6131 (2025) The Impact of AI-Driven Chatbot Assistance on Protocol Development and Clinical Research Engagement: An Implementation Report. Journal of Personalized Medicine, 15 (7). p. 269.

Preview

Text jpm-15-00269.pdf - Published Version Available under License Creative Commons Attribution. Download (544kB) | Preview

Abstract

Background: The integration of artificial intelligence (AI) into healthcare research has the potential to enhance research capacity, streamline protocol development, and reduce barriers to engagement. Medway NHS Foundation Trust identified a plateau in homegrown research participation, particularly among clinicians with limited research experience. A generative AI-driven chatbot was introduced to assist researchers in protocol development by providing step-by-step guidance, prompting ethical and scientific considerations, and offering immediate feedback. Methods: The chatbot was developed using OpenAI’s GPT-3.5 architecture, customised with domain-specific training based on Trust guidelines, Health Research Authority (HRA) requirements, and Integrated Research Application System (IRAS) submission protocols. It was deployed to guide researchers through protocol planning, ensuring compliance with ethical and scientific standards. A mixed-methods evaluation was conducted using a qualitative-dominant sequential explanatory design. Seven early adopters completed a 10-item questionnaire (5-point Likert scales), followed by eight free-flowing interviews to achieve thematic saturation. Results: Since its launch, the chatbot has received an overall performance rating of 8.86/10 from the seven survey respondents. Users reported increased confidence in protocol development, reduced waiting times for expert review, and improved inclusivity in research participation across professional groups. However, limitations in usage due to free-tier platform constraints were identified as a key challenge. Conclusions: AI-driven chatbot tools show promise in supporting research engagement in busy clinical environments. Future improvements should focus on expanding access, optimising integration, and fostering collaboration among NHS institutions to enhance research efficiency and inclusivity.

Item Type:	Article
Status:	Published
DOI:	10.3390/jpm15070269
School/Department:	London Campus
URI:	https://ray.yorksj.ac.uk/id/eprint/12205

University Staff: Request a correction | RaY Editors: Update this record