Clinical Report: An Algorithm Utilizing Decision Trees for Systematic Risk Assessment

Overview

This study presents a decision tree model for risk stratification of rare inflammatory rheumatic diseases in a tertiary referral setting. The model achieved an apparent classification accuracy of 81.5%.

Background

Rare inflammatory rheumatic diseases often present with heterogeneous and multisystemic symptoms, complicating early diagnosis. Timely and accurate diagnosis is crucial, as delays can adversely affect patient outcomes. Structured approaches that integrate clinical and laboratory data may enhance diagnostic precision in these complex cases.

Data Highlights

Key Findings

• A confirmed rheumatologic diagnosis was established in 52.0% of patients evaluated.
• Fatigue and generalized pain were the most prevalent symptoms reported across diagnostic groups.
• The RHEUMA group had significantly higher scores in selected symptom domains and an immunoserological laboratory index (p < 0.05).
• The CHAID-based decision model outperformed logistic regression in classification accuracy (81.5% vs. 76.3%).

Clinical Implications

The decision tree model provides a structured framework for risk stratification in patients with unclear systemic complaints.

Conclusion

The proposed decision tree model requires further validation in independent cohorts.

Related Resources & Content

1. Clinical Rheumatology, 2025 -- Leveraging Pre-Consultation Data from Electronic Health Records to Develop a Predictive Model for Diagnosing Chronic Rheumatic Disorders in Children
2. Journal of Medical Internet Research (JMIR), 2026 -- Rheumatic Digital Twin: Proposed Machine Learning–Based Multimodal Framework to Inform Clinical Decision-Making
3. Knee Surgery, Sports Traumatology, Arthroscopy, 2020 -- Enhanced Diagnostic Precision for Knee Septic Arthritis Using a Machine Learning Approach
4. ERS/EULAR clinical practice guidelines for connective tissue disease-associated interstitial lung disease, 2026
5. Clinical Rheumatology — Application of machine learning methods in creating and enhancing a predictive model for the early detection of ankylosing spondylitis
6. EULAR recommendations for the management of Behçet's syndrome: 2025 update
7. Identification of red flags for IgG4-related disease
8. ERS/EULAR clinical practice guidelines for connective tissue disease-associated interstitial lung disease developed by the task force for connective tissue disease-associated interstitial lung disease of the European Respiratory Society (ERS) and the European Alliance of Associations for Rheumatology (EULAR) Endorsed by the European Reference Network on rare respiratory diseases (ERN-LUNG)
9. The EULAR points to consider regarding the development of criteria for the assessment of the disease activity in adult-onset Still’s disease - ScienceDirect
10. Rheumatoid Arthritis referral criteria: systematic review of the literature | BMC Rheumatology | Springer Nature Link
11. Accuracy of Diagnostic Codes and Algorithms Used to Identify Rheumatoid Arthritis and Juvenile Idiopathic Arthritis in Administrative Claims and Electronic Health Records: Systematic Review and Meta‐Analysis - Saka‐Herrán - 2026 - Arthritis Care & Research - Wiley Online Library
12. The utility of machine learning-based decision support system in referral of suspected rheumatic disease
13. Criteria | American College of Rheumatology