OBJECTIVE: To quantify GI symptoms of children with JIA using the Gastrointestinal Symptom Scale for Pediatrics (GIPS), to verify for the GIPS reliability, construct validity and quality of parent proxy reporting, and to evaluate the relationship of GI symptoms severity and HRQL in JIA. METHODS: A convenience sample of 54 families of JIA patients was interviewed twice (patients (pts) > 7 yrs,1 parent per family). GI symptom severity was measured by the GIPS (7 item scale with yes/no answers; score 0 - 7; 0=no GI symptoms) and a visual analog scale (VAS-GI). Information on other outcomes was obtained, incl. the Childhood Health Assessment Questionnaire (CHAQ); the Pediatric Quality of Life Inventory Rheumatology Module (PedsQL-R) and Generic Core Scale (PedsQL-G). RESULTS: Forty-six percent of the pts (mean age: 10.3 yrs) had some GI symptoms. Treatments included NSAIDs (n=45), methotrexate (MTX; n=33) and GI protectants (n=21). Intrarater reliability and internal consistency of the GIPS (parent report: weighted kappa=.7; Crohnbachs-á=1) were high; the quality of parent proxy reports was very good (intraclass corr. coeff.=.7). Scores of the GIPS, GI-VAS, PedsQL-R and PedsQl-G (r=.5-.8) were strongly correlated. The mean GIPS score of pts having GI symptoms was 2, with nausea and epigastric pain being most common. Use of MTX (p<.003) and NSAIDs (p<.03) led to signficantly higher and GI protectants to signficantly lower GIPS scores (p<.008) in univariate analysis. Corrected for the disease severity and activity, children with moderate/severe GI symptoms (GIPS>2; n=16) had significantly lower HRQL (PedsQL-R: p<.005; PedsQL-G: p<.04) and more disability (CHAQ; p<.005) compared to patients without GI symptoms. CONCLUSION: The GIPS is a reliable and valid measure of GI symptom severity. GI symptoms are frequent among children with JIA and, if moderate or severe, have a significant negative impact on the HRQL. GI side effects require special consideration for patient management and medication choices in JIA.