JMIR Publications

ABSTRACT

Background:

Inflammatory Rheumatic Diseases (IRDs) affect 5% of the general population, while 35% of the population experiences Musculoskeletal Complaints (MSCs). IRDs cause early disability, reduced life-expectancy and considerable healthcare costs. Early diagnosis is essential to prevent long-term damage. Similarly important is the early identification of patients with MSCs without IRDs to prevent unnecessary healthcare expenses. Of the population referred to the rheumatologist, 60% have non-inflammatory MSCs while only 20% of patients with an IRD see a rheumatologist within three months of symptom onset. The need for digital predictive (triage) tools for Rheumatic and Musculoskeletal Diseases led to the development of the Frysian Questionnaire for differentiation of MSK complaints (FRYQ).

Objective:

To assess if FRYQ questionnaire can distinguish inflammatory rheumatic disease (IRD) from non-IRDs in general, and Rheumatoid arthritis and Fibromyalgia specifically, in newly referred patients.

Methods:

The FRYQ is an 87-item tool (20 open, 67 closed-ended questions) used to triage new rheumatology patients at Frisius Medical Center in the Netherlands. We analyzed data from two sources: Dataset A with 728 outpatient clinic patients and Dataset B with 373 patients from the JPAST study. We built a classifier using eXtreme Gradient Boosting to distinguish inflammatory from non-inflammatory conditions based on closed-ended questions. Using Elastic Net regularization optimized with Bayesian techniques through 1000 iterations, we identified the most informative questions. Data was split 80/20 for model construction and performance estimation. We evaluated classification using ROC-curve analysis and assessed feature importance through SHAP analysis. To test generalizability, we replicated our analysis on Dataset B. Finally, we examined whether the questions of the FRYQ could be used to identify specific conditions beyond the general categories of IRD and non-IRD, specifically for detecting Fibromyalgia and Rheumatoid arthritis.

Results:

Feature selection reduced the questionnaire from 67 to 28 items while maintaining discriminative power. After initial development, the model showed stable performance when applied to the external validation cohort, achieving an AUC-ROC of 0.72 (95% CI: 0.67-0.78) for distinguishing inflammatory from non-inflammatory conditions. A clinical threshold of 0.30 provided a sensitivity of 71% and specificity of 56% in the external validation. The FRYQ demonstrated even stronger performance in identifying fibromyalgia specifically (AUC-ROC 0.78) and moderate ability to detect rheumatoid arthritis (AUC-ROC 0.74) in the external validation. Key discriminating features included symptom duration, pain response to movement, anti-inflammatory medication effectiveness, and presence of specific comorbidities.

Conclusions:

The FRYQ questionnaire effectively distinguishes inflammatory from non-inflammatory rheumatic conditions prior to specialist consultation and shows particular strength in identifying fibromyalgia. This tool could potentially improve rheumatology triage, reducing delays for patients requiring urgent specialist care while directing others to more appropriate resources. Prospective studies are needed to determine its impact on clinical outcomes and healthcare efficiency.