Using sentiment analysis to review patient satisfaction data located on the internet
Journal of Health Organization and Management
ISSN: 1477-7266
Article publication date: 13 April 2015
Abstract
Purpose
The purpose of this paper is to test the usefulness of sentiment analysis and time-to-next-complaint methods in quantifying text-based information located on the internet. As important, the authors demonstrate how managers can use time-to-next-complaint techniques to organize sentiment analysis derived data into useful information, which can be shared with doctors and other staff.
Design/methodology/approach
The authors used sentiment analysis to review patient feedback for a select group of gynecologists in Virginia. The authors utilized time-to-next-complaint methods along with other techniques to organize this data into meaningful information.
Findings
The authors demonstrated that sentiment analysis and time-to-next-complaint techniques might be useful tools for healthcare managers who are interested in transforming web-based text into meaningful, quantifiable information.
Research limitations/implications
This study has several limitations. For one thing, neither the data set nor the techniques the authors used to analyze it will account for biases that resulted from selection issues related to gender, income, and culture, as well as from other socio-demographic concerns. Additionally, the authors lacked key data concerning patient volumes for the targeted physicians. Finally, it may be difficult to convince doctors to consider web-based comments as truthful, thereby preventing healthcare managers from using data located on the internet.
Practical implications
The report illustrates some of the ways in which healthcare administrators can utilize sentiment analysis, along with time-to-next-complaint techniques, to mine web-based, patient comments for meaningful information.
Originality/value
The paper is one of the first to illustrate ways in which administrators at clinics and physicians’ offices can utilize sentiment analysis and time-to-next-complaint methods to analyze web-based patient comments.
Keywords
Acknowledgements
The authors have not received any funding for this project. Neither of them had any conflicts of interest when they submitted the first copy of the report to the Journal of Health Organization and Management. During the interim period between the review and the revisions, Anthony Hopper joined Rapid Improvement, Inc. The company uses sentiment analysis and AI to parse patient satisfaction data. The authors are grateful to Dr Farrokh Alemi for his helpful suggestions and critiques. They might not have completed this project without his encouragement and commentary. Dr Alemi is currently the Chief of Performance Improvement at the District of Columbia Veterans Administration Medical Center. He is also the chief scientist for Rapid Improvement, Inc., a position he held before the authors began work on this report.
Citation
Hopper, A.M. and Uriyo, M. (2015), "Using sentiment analysis to review patient satisfaction data located on the internet", Journal of Health Organization and Management, Vol. 29 No. 2, pp. 221-233. https://doi.org/10.1108/JHOM-12-2011-0129
Publisher
:Emerald Group Publishing Limited
Copyright © 2015, Emerald Group Publishing Limited