Musbahi O, Pouris K, Hadjixenophontos S, Al-Saadawi A, Soteriou I, Cobb JP, Jones GG. Machine learning for patient selection in corticosteroid decision making in knee osteoarthritis: A feasibility model. World J Methodol 2025; 15(4): 105493 [DOI: 10.5662/wjm.v15.i4.105493]
Corresponding Author of This Article
Omar Musbahi, Senior Researcher, Department of Surgery and Cancer - Faculty of Medicine, Imperial College London, 86 Wood Ln, London W12 0BZ, United Kingdom. om112@ic.ac.uk
Research Domain of This Article
Orthopedics
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
World J Methodol. Dec 20, 2025; 15(4): 105493 Published online Dec 20, 2025. doi: 10.5662/wjm.v15.i4.105493
Machine learning for patient selection in corticosteroid decision making in knee osteoarthritis: A feasibility model
Omar Musbahi, Kyriacos Pouris, Savvas Hadjixenophontos, Ahmed Al-Saadawi, Iris Soteriou, Justin Peter Cobb, Gareth G Jones
Omar Musbahi, Justin Peter Cobb, Gareth G Jones, Department of Surgery and Cancer - Faculty of Medicine, Imperial College London, London W12 0BZ, United Kingdom
Kyriacos Pouris, Savvas Hadjixenophontos, Iris Soteriou, Faculty of Medicine and Dentistry, Imperial College London, London W12 0BZ, United Kingdom
Ahmed Al-Saadawi, Faculty of Medicine and Dentistry, Queen Mary University of London, London E1 2AD, United Kingdom
Author contributions: Musbahi O, Cobb JP, and Jones GG conceptualized the study and designed the research methodology; Musbahi O, Pouris K, and Hadjixenophontos S curated the data; Musbahi O and Hadjixenophontos S analyzed the data; Musbahi O, Hadjixenophontos S, and Al-Saadawi A authored the manuscript; Musbahi O, Pouris K, Hadjixenophontos S, Al-Saadawi A, Soteriou I, Cobb JP, and Jones GG were involved in manuscript editing; Cobb JP and Jones GG were involved in supervision; and all authors thoroughly reviewed and endorsed the final manuscript.
Supported by National Institute For Health and Care Research, No. NIHR302632.
Institutional review board statement: Ethical approval was not required for this study as it involved a secondary analysis of publicly-available data on approved cohort studies.
Informed consent statement: Informed consent was not required for this study, as it was already obtained in the approved cohort studies used for analysis.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Data sharing statement: All the data has been published online and is available for use. Statistical code is available from the corresponding author: om112@ic.ac.uk.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Omar Musbahi, Senior Researcher, Department of Surgery and Cancer - Faculty of Medicine, Imperial College London, 86 Wood Ln, London W12 0BZ, United Kingdom. om112@ic.ac.uk
Received: January 24, 2025 Revised: March 29, 2025 Accepted: May 7, 2025 Published online: December 20, 2025 Processing time: 192 Days and 14 Hours
Abstract
BACKGROUND
Relieving pain is central to the early management of knee osteoarthritis, with a plethora of pharmacological agents licensed for this purpose. Intra-articular corticosteroid injections are a widely used option, albeit with variable efficacy.
AIM
To develop a machine learning (ML) model that predicts which patients will benefit from corticosteroid injections.
METHODS
Data from two prospective cohort studies [Osteoarthritis (OA) Initiative and Multicentre OA Study] was combined. The primary outcome was patient-reported pain score following corticosteroid injection, assessed using the Western Ontario and McMaster Universities OA pain scale, with significant change defined using minimally clinically important difference and meaningful within person change. A ML algorithm was developed, utilizing linear discriminant analysis, to predict symptomatic improvement, and examine the association between pain scores and patient factors by calculating the sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and F2 score.
RESULTS
A total of 330 patients were included, with a mean age of 63.4 (SD: 8.3). The mean Western Ontario and McMaster Universities OA pain score was 5.2 (SD: 4.1), with only 25.5% of patients achieving significant improvement in pain following corticosteroid injection. The ML model generated an accuracy of 67.8% (95% confidence interval: 64.6%-70.9%), F1 score of 30.8%, and an area under the curve score of 0.60.
CONCLUSION
The model demonstrated feasibility to assist clinicians with decision-making in patient selection for corticosteroid injections. Further studies are required to improve the model prior to testing in clinical settings.
Core Tip: Historically, the efficacy of corticosteroid injections in knee osteoarthritis has been heavily debated, as patient responses can vary significantly. This study evaluates the feasibility of a machine learning model to identify which patients with knee osteoarthritis will benefit from corticosteroid injections. Data from two cohort studies were combined for analysis. The model generated an accuracy of 67.8% (95% confidence interval: 64.6%-70.9%), F1 score of 30.8%, and an area under the curve score of 0.60. These metrics demonstrate feasibility to assist clinicians with decision-making in patient selection for corticosteroid injections. Further research is required to improve the model prior to testing in clinical settings.