Nurses’ Expectation of Artificial Intelligence to Analyse Patient Stories to Improve Person-Centred Nursing: A Qualitative Descriptive Study

Background In previous work, eight key performance indicators (KPIs) and an associated measurement framework were developed to provide a new perspective on nursing quality. Patient stories are among the four tools used to generate data to evidence the KPIs. To streamline data collection, the Implementing and Measuring Person-Centredness using an App for Knowledge Transfer (iMPAKT) app was developed, which automatically transcribes the stories, while nurses manually analyse them to identify which KPIs are present. This is time-consuming, and with the rise of artificial intelligence (AI), there is potential to automate it. Aims and Objectives This study explored the potential of AI to capture patient experiences through stories and applied directly to the iMPAKT app. Methods This qualitative descriptive study collected data from focus groups with nurses who either have expertise in informatics or work directly with patients, aiming to identify criteria for an AI model capable of semantic search and text mining in collaboration with users. The focus group data were analysed using reflexive thematic analysis. Results Five themes and four principles were identified. Participants emphasised the importance of enabling all patients to participate in data collection while ensuring the solution is user-friendly and considers patient needs. Although participants expressed trust in AI and a desire for a high level of automation, they stressed the importance of maintaining a human-in-the-loop approach, involving both nurses and patients. Ethical considerations, such as anonymising stories and obtaining informed consent, were highlighted to safeguard patient safety and foster a psychologically safe work environment. Participants also discussed the potential for AI to identify disturbing incidents within stories, recognising the organisational responsibilities involved. Conclusions The findings demonstrate a clear intention to develop an inclusive, AI-driven system that removes barriers to patient engagement and emphasises the potential for nurses to contribute to AI development. Ethical responsibilities related to AI remain crucial, as it offers both opportunities and challenges.