Predictive Utility of Haptoglobin and Erythrocyte Sedimentation Rate for Joint Disease Activity Score in Pediatric Juvenile Idiopathic Arthritis: A Cross-Sectional Statistical Modeling Study

Juvenile idiopathic arthritis (JIA) requires clinically interpretable indices that integrate patient status with objective inflammatory activity. This study reformats and expands a statistical consulting analysis that evaluated whether haptoglobin (HPTG), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) predict the Joint Disease Activity Score (JADAS) in pediatric patients. A dataset of 269 observations was screened; 266 complete observations were analyzed after exclusion of three records with missing CRP values. Exploratory analysis showed positive correlations between JADAS and ESR (r = 0.474), CRP (r = 0.398), HPTG (r = 0.471), and CalcAlpha2 (r = 0.332), whereas AgeTest was not correlated with disease activity. Because raw JADAS exhibited right-skewness and heteroskedastic residuals, regression modeling was conducted using log(JADAS + 1). Stepwise selection identified HPTG and ESR as the most parsimonious independent predictor set, explaining approximately 30% of the variance in log-transformed JADAS. CRP, CalcAlpha2, and AgeTest did not provide additional marginal predictive value after HPTG and ESR were included. Gender was examined as an effect modifier and confounder; neither ESR-by-gender nor HPTG-by-gender interactions were statistically significant, and adjustment for gender changed the primary coefficients by less than 2%. The final model was log(JADAS + 1) = 1.61592 + 0.00769 x ESR + 0.00215 x HPTG. These findings support HPTG as a candidate complementary inflammatory biomarker alongside ESR, while highlighting the need for external validation, explicit specification of the JADAS variant, and longitudinal replication before clinical implementation.