Machine learning clinical decision support for interdisciplinary multimodal chronic musculoskeletal pain treatment

Aug 5, 2026

Upload by Himbo

Chronic musculoskeletal pain affects a large proportion of people globally and reduces mobility, social participation, work capacity, and quality of life. Interdisciplinary multimodal pain treatment programs can support patients in behavior change and in achieving value-based goals for better pain management. Using 2019–2021 Centre for Integral Rehabilitation data (n=2,364), a multidimensional machine learning framework was developed to predict outcomes across five clinical domains using 13 measures. The resulting stratified prognostic patient profile enables consistent outcome assessment and may support clinician-aided patient selection and goal setting.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages18

Reading Time45 min

AccessFree to download

Introduction

Clinical problem and rationale for multimodal interdisciplinary treatment

Methods

Data source and outcome domains

Feature selection and model validation

Results

Algorithm performance and patient-level prognostic profile

Discussion

Clinical implications for patient selection and goal setting

TYPE Original Research PUBLISHED 09 May 2023 DOI 10.3389/fpain.2023.1177070 EDITED BY Abhinav Kaushik, Stanford University, United States REVIEWED BY Sayane Shome, Stanford University, United States Eleni G. Hapidou, McMaster University, Canada *CORRESPONDENCE Rob J. E. M. Smeets [r.smeets@maastrtichtuniversity.nl](r.smeets@maastrtichtuniversity.nl) RECEIVED 01 March 2023 ACCEPTED 07 April 2023 PUBLISHED 09 May 2023 CITATION Zmudzki F and Smeets RJEM (2023) Machine learning clinical decision support for interdisciplinary multimodal chronic musculoskeletal pain treatment. Front. Pain Res. 4:1177070 . doi: 10.3389/fpain.2023.1177070 COPYRIGHT © 2023 Zmudzki and Smeets. This is an openaccess article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Machine learning clinical decision support for interdisciplinary multimodal chronic musculoskeletal pain treatment Fredrick Zmudzki1,2 and Rob J. E. M. Smeets3,4,5* 1Époque Consulting, Sydney, NSW, Australia, 2Social Policy Research Centre, University of New South Wales, Sydney, NSW, Australia, 3Department of Rehabilitation Medicine, Care and Public Health Research Institute (CAPHRI), Faculty of Health, Life Sciences and Medicine, Maastricht University, Maastricht, Netherlands, 4CIR Rehabilitation, Eindhoven, Netherlands, 5Pain in Motion International Research Group (PiM), Brussels, Belgium Introduction: Chronic musculoskeletal pain is a prevalent condition impacting around 20% of people globally; resulting in patients living with pain, fatigue, restricted social and employment capacity, and reduced quality of life. Interdisciplinary multimodal pain treatment programs have been shown to provide positive outcomes by supporting patients modify their behavior and improve pain management through focusing attention on speciﬁc patient valued goals rather than ﬁghting pain. Methods: Given the complex nature of chronic pain there is no single clinical measure to assess outcomes from multimodal pain programs. Using Centre for Integral Rehabilitation data from 2019–2021 (n = 2,364), we developed a multidimensional machine learning framework of 13 outcome measures across 5 clinically relevant domains including activity/disability, pain, fatigue, coping and quality of life. Machine learning models for each endpoint were separately trained using the most important 30 of 55 demographic and baseline variables based on minimum redundancy maximum relevance feature selection. Five-fold cross validation identiﬁed best performing algorithms which were rerun on deidentiﬁed source data to verify prognostic accuracy. Results: Individual algorithm performance ranged from 0 .49 to 0 . 65 AUC reﬂecting characteristic outcome variation across patients, and unbalanced training data with high positive proportions of up to 86% for some measures. As expected, no single outcome provided a reliable indicator, however the complete set of algorithms established a stratiﬁed prognostic patient proﬁle. Patient level validation achieved consistent prognostic assessment of outcomes for 75 .3% of the study group (n = 1,953) . Clinician review of a sample of predicted negative patients (n =81) independently conﬁrmed algorithm accuracy and suggests the prognostic proﬁleis potentially valuable for patient selection and goal setting. Discussion: These results indicate that although no single algorithm was individually conclusive, the complete stratiﬁed proﬁle consistently identiﬁed patient outcomes. Our predictive proﬁle provides promising positive contribution for clinicians and patients to assist with personaliz

FAQs

Why is interdisciplinary multimodal treatment important for chronic musculoskeletal pain?

It helps patients modify behavior and manage pain by focusing attention on specific value-based goals rather than solely trying to eliminate pain, with positive sustained outcomes even when other modalities have failed.

How was the machine learning decision support framework built?

It used Centre for Integral Rehabilitation data from 2019–2021 (n=2,364) to train models for 13 outcome measures across five clinically relevant domains, using selected demographic and baseline variables.

Can clinicians rely on a single predicted outcome?

No single outcome provided a reliable indicator, but the complete set of algorithms formed a stratified prognostic patient profile that consistently assessed outcomes for most participants.

Aug 5, 2026

Upload by Himbo