2124: Reliability and Validation of the Physician's Global Assessment of Lung Disease (PGALD) in Systemic Juvenile Idiopathic Arthritis -Associated Lung Disease (SJIA-LD)

Background/Purpose: The physician global assessment of lung disease (PGALD) is a recently proposed disease activity measure for patients with systemic juvenile idiopathic arthritis-associated lung disease (SJIA-LD) – a significant cause of morbidity and mortality in SJIA. This study evaluates the reliability and construct validity of the PGALD.

Methods: 20 SJIA-LD clinical vignettes were created, informed by historical patient encounters as templates. Vignettes included brief patient medical histories, physical examination findings, a list of prescribed medications, laboratory results, diagnostic imaging, pulmonary function test results, and biopsy results – when available. Fifty-seven pediatric rheumatologists and pulmonologists with experience caring for children with SJIA-LD were invited to rate vignettes utilizing the PGALD, a 10-point Likert scale (0 = inactive SJIA-LD; 10 = highly active SJIA-LD). Raters who completed the initial round of 20 vignettes were asked three days later to rate a subset of 8 repeat vignettes. Inter- and intra-rater reliability were assessed using intraclass correlation coefficients (ICC). Based on the 95% confidence interval of ICC estimates, values less than 0.5 indicated poor reliability, between 0.5-0.75 moderate-, between 0.76-0.9 good-, and >0.9 excellent reliability. SJIA-LD features influencing PGALD ratings were assessed in univariate analysis.

Results: The ICC for all raters was 0.678 (0.543, 0.819), indicating moderate to good inter-rater reliability. The intra-rater ICC was 0.863 (0.823, 0.893) indicating good intra-rater reliability. Factors associated with higher mean PGALD scores included the presence of crackles on auscultation (5.7 vs. 2.9; p=0.001), hypoxemia on pulse oximeter (5.6 vs. 3.2; p=0.01), current oxygen requirement (6.3 vs. 3.5; p=0.04), suggestive diagnostic imaging features (p≤0.01), and three or more prescribed medications for SJIA-LD (p≤0.01). Pulmonary function measures demonstrated significant negative correlations with mean PGALD scores, including forced vital capacity (r=−0.71; p=0.01), total lung capacity (r=−0.92; p=0.01), and diffusion capacity (r=−0.97; p=0.0002). Only the CRP was weak-to-moderately correlated with PGALD scores (r=0.39; p=0.09), while other laboratory values including ferritin, IL-18, CXCL9, and sIL2R were not significantly correlated.

Conclusion: The PGALD is a novel measure of SJIA-associated LD activity. Its initial validation suggests acceptable construct validity and reliability. Additional studies are needed to assess its responsiveness to change over time.