Predicting Psoriatic Arthritis in Psoriasis Patients - A Swiss Registry Study

Area of Science:

• Dermatology
• Rheumatology
• Medical Informatics

Background:

• Psoriatic arthritis (PsA) is a common comorbidity in psoriasis patients.
• PsA diagnosis often occurs years after psoriasis onset, increasing disease burden.

Purpose of the Study:

• To determine the predictability of PsA in psoriasis patients.
• To identify key predictors for PsA development.

Main Methods:

• Utilized data from the Swiss Dermatology Network on Targeted Therapies (SDNTT).
• Employed gradient-boosted decision trees and mixed models for patient classification.
• Identified high-predictive variables and visualized time to PsA diagnosis using Kaplan-Meier method.

Main Results:

• The predictive model achieved an AUROC of up to 73.7%.
• Significant predictors included age, Psoriasis Area and Severity Index (PASI), physical well-being, and nail psoriasis severity.
• PASI score showed a significant correlation with PsA burden/severity.

Conclusions:

• Clinical characteristics can effectively distinguish between patients with and without PsA.
• Predicting future PsA diagnoses is achievable, allowing for early identification of at-risk patients.