EUVIMEDEUVIMED entdecken
CrossrefInfokarte

Cluster Analysis and Predictive Factors Influencing Physiotherapists’ Recommendations for Exercise Therapy for Musculoskeletal Pain

Annika Griefahn, Florian Avermann, Christoff Zalpour, Robert Percy Marshall

physioscience · 2026-08-14

DOI: 10.1055/a-2797-4664 ↗
AbstractBibliografische Angaben

Abstract

Abstract Background The decision as to whether patients with musculoskeletal pain are suitable for exercise therapy is central to safe and effective physiotherapy care. Assessments by physiotherapists can provide important information about contraindications. Objective The study examined three research questions: (1) relationship between pain-related clusters and physiotherapeutic classification of contraindications for exercise therapy alone, (2) characteristic differences between the clusters in terms of pain-related parameters, and (3) predictive factors for the presence of a contraindication. Method In a retrospective secondary data analysis, data from 80 patient cases were evaluated that had been classified in a previous study by both a rule-based AI system and 20 physical therapists with regard to the presence of a contraindication for exercise therapy alone. To identify pain-related patterns, a k-means cluster analysis based on 24 pain-related variables was performed. Differences between the clusters were analyzed using ANOVA and post-hoc tests. A binary logistic regression was used to identify significant predictors for the presence of contraindication. Results 3 clusters were identified. Significant predictors for the presence of a contraindication for exercise therapy were high pain intensity (numerical rating scale 7–10), persistent pain (especially at night), multiple pain regions, and pain descriptors such as “pressing” or “burning.” The final regression model showed high predictive power (Nagelkerke’s R² = 0.90). Conclusion Certain pain-related parameters can reliably predict the presence of a contraindication for exercise therapy alone. Integrating these factors into AI-supported systems has the potential to increase patient safety, avoid inappropriate therapies, and support personalized treatment decisions. Registration of primary study OSF, https://doi.org/10.17605/OSF.IO/YCNJQ, 30.12.2021

Wissen, das zu Ihrer Arbeit passt.

Vom kostenlosen Einstieg über die persönliche Recherche bis zur vollständigen Research Suite.

Tarife ansehen →

Sind alle Verlagsvolltexte enthalten?

Nein. Bibliografische Nachweise, Abstracts und Volltexte sind unterschiedliche Inhalte. Ob ein Volltext verfügbar ist, hängt von der Quelle, Open Access und Ihren vorhandenen Lizenzen ab. Ein EUVIMED-Tarif ersetzt keine zusätzlichen Verlagsrechte.

Tarife ansehen →