What patients like or dislike in physicians: Analyzing drivers of patient satisfaction and dissatisfaction using a digital topic modeling approach. Issue 3 (May 2021)

• Record Type: Journal Article
• Authors: Shah, Adnan Muhammad; Yan, Xiangbin; Tariq, Samia; Ali, Mudassar
• Is Part Of: Information processing & management. Volume 58:Issue 3(2021)
• Publication Date: 2021-05
• Subjects: Patient satisfaction -- Patient dissatisfaction -- Text mining -- Topic modeling -- LDA -- Sentiment analysis; Information storage and retrieval systems -- Periodicals; Information science -- Periodicals; Systèmes d'information -- Périodiques; Sciences de l'information -- Périodiques; Information science; Information storage and retrieval systems; Periodicals; 658.4038
• DOI: 10.1016/j.ipm.2021.102516
• Languages: English
• ISSNs: 0306-4573

Abstract

Highlights: The current research proposed a text mining approach to investigate the drivers of patient satisfaction and dissatisfaction across different types of diseases. Drawing on Herzberg's two-factor theory, this research identified the key topics of patient satisfaction and dissatisfaction expressed in online doctor reviews. The text mining method based on combining Sentinet and LDA was applied to disclose the semantics of patients' healthcare experiences. The classification results reveal that the proposed model that analyzes patients' opinions toward different aspects of care outperformed other state-of-the-art models. Abstract: A large volume of patients' opinions—as online doctor reviews (ODRs)—are available online in order to access, analyze, and improve patients' perceptions about the quality of care; however, this development needs to be explored further. Drawing on the two-factor theory, this paper aims to mine ODRs to explore the different determinants of patient satisfaction (PS) and patient dissatisfaction (PD) toward the United Kingdom healthcare services. This study collects reviews from a publicly available medical website Iwantgreatcare.org from January 2014 to December 2018, followed by the text mining method based on combining SentiNet and LDA to disclose the semantics of patients' healthcare experiences. The proposed method found latent topics across the high-risk and low-risk disease category that revealed new insights into what patients value when consulting a physician and what they dislike. For high-risk and low-risk diseases, the determinants of PS were more specific to the hospital business process (hospital environment, location, hospital cafeteria servicescape, parking availability, and medical process, etc.) and doctor-related aspects (physician knowledge, competence, and attitudes, etc.). In contrast, patients' concerns were most commonly related to their treatment experience and staff bedside manners for both disease categories. Finally, the classification results revealed that the proposed model, which analyzes patient opinion toward different aspects of care, outperformed other state-of-the-art models, with the highest classification F1 -score of 88%. The study findings provide a clue for doctors, hospitals, and government officials to enhance PS and minimize PD by addressing their needs and improve the quality of care across different types of diseases, particularly in the current pandemic era of COVID-19.