Leveraging AI to Drive Timely Improvements in Patient Experience Feedback: Algorithm Validation

• Published in: JMIR medical informatics Vol. 13; p. e60900
• Main Authors: Khanbhai, Mustafa, Carenzo, Catalina, Aryasinghe, Sarindi, Manton, David, Mayer, Erik
• Format: Journal Article
• Language: English
• Published: Canada JMIR Publications 10.07.2025
• Subjects: Algorithms; Artificial Intelligence; Automatic text analysis; Automation; Classification; England; Feedback; Humans; Language; Large language models; Libraries; Machine Learning; Multimedia; Natural Language Processing; New Methods; Original Paper; Participatory Medicine & E-Patients; Patient Engagement and Empowerment; Patient Satisfaction; Python; Quality Improvement; R&D; Real time; Research & development; Responsible Health AI; Sentiment analysis; Servers; Software; Support vector machines; Text categorization; Tools, Programs and Algorithms; England; United Kingdom > UK; free-text analysis; natural language processing; patient experience; friends and family test; quality improvement; patient feedback; health informatics; machine learning; artificial intelligence; algorithm
• ISSN: 2291-9694; 2291-9694
• DOI: 10.2196/60900

Understanding and improving patient care is pivotal for health care providers. With increasing volumes of the Friends and Family Test (FFT) data in England, manual analysis of this patient feedback poses challenges for many health care organizations. This underscores the importance of automated text analysis, particularly in predicting sentiments and themes in real time. Leveraging machine learning and natural language processing, this study explores the utility of a supervised algorithm to systematically test and refine the algorithm's cross-contextual performance in diverse health care settings, addressing variations in population characteristics, geographical locations, and care settings, ultimately driving improvements based on patient feedback. The text analytics algorithm initially developed in a large acute trust in London was further tested in 9 health care organizations with diverse care settings across England. These trusts varied in technical capacity and resource, population demographics, and FFT free text datasets. Testing and validation of the algorithm were performed, including manual coding of a subset of retrospective comments. Technical infrastructure, including coding environments and packages for algorithm testing and deployment, was optimized. The algorithm was iteratively trained using bag of words from anonymized data, tailored to accommodate contextual variations, and tested for change in algorithm performance while simultaneously rectifying issues identified. The algorithm demonstrated satisfactory overall accuracy (>75%) in predicting themes and sentiments embedded within free-text responses across a variety of care settings and population demographics. While the algorithm yielded strong and reusable models in relatively stable environments, such as adult inpatient care settings, the initial accuracy was notably lower in organizations providing services such as pediatrics and mental health. However, the accuracy of our algorithm significantly improved when individual trust coding templates were applied. Thematic saturation was reached after the fifth organization was recruited, and no further coding was required for the last 4 organizations. Subsequently, a framework and pipeline for deployment of the algorithm were developed to provide a standardized approach for implementation and analysis of FFT free text, ensuring ease of use. This study represents a significant step forward in leveraging free-text FFT data for valuable insights in diverse health care settings through the testing and development of a robust supervised learning text analytics algorithm. The disparity in some care settings was anticipated, given that the lexicon and phraseology used was inherently different from those prevalent in adult inpatient care (where the algorithm was developed). However, these challenges were addressed with further coding and testing. This approach enhanced the accuracy and reliability of the algorithm, encouraged inter- and intraorganizational collaboration, and shared learning.