Objectives
Assess reliability and validity of Patient-Reported Outcomes Measurement Information System (PROMIS) pediatric self-report and parent-proxy report fixed short forms in juvenile myositis (JM).
Methods
Children with JM (8–17yo) and parents of 5–17 yo JM patients completed PROMIS measures (Physical Function, Pain Interference, Fatigue, Emotional Distress), PedsQL Generic Core scales and Rheumatology Module (PedsQL-GC/-RM). Internal consistency reliability was assessed via Cronbach's alpha. Patient-parent agreement was assessed via intraclass correlations (ICC). Concurrent and construct validity were assessed via Spearman's correlations between PROMIS versus PedsQL-GC/-RM and clinical/lab data respectively. Known-groups validity was assessed by comparing PROMIS T-scores between clinically distinct JM patients.
Results
We enrolled 75 JM participants, with 57 administered self-report and all 75 administered parent-proxy report measures per participant age. PROMIS measures were feasible (>96% completion), with high internal consistency reliability (Cronbach's alpha >0.8). Patient-parent assessments demonstrated moderate agreement (ICC >0.5) for Mobility, Upper Extremity, and Fatigue domains, and smaller correlations (ICC 0.41–0.47) as expected for Pain Interference, Depressive Symptoms, and Anxiety. Concurrent validity was demonstrated by moderate correlation (Spearman's rho >0.5) for all but 1 hypothesized relationships of PROMIS and PedsQL-GC/-RM domains. Although low disease activity and small sample size limited statistical power, construct validity and known-groups validity were demonstrable for multiple PROMIS pediatric self-report and parent-proxy report measures.
Conclusion
PROMIS measures show evidence of reliability and validity in JM. Child and parent reports differ sufficiently to suggest both should be collected. PROMIS measures can be considered for clinical and research use in JM.