Background/Purpose: Knowledge of co-occurring disease processes (comorbidities) is important for understanding disease pathogenesis, refining disease classifications, developing appropriate screening and prevention strategies, and determining overall burden of disease. We analyze prevalence and patterns of comorbidities in children with JIA and SLE in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry.

Methods: We analyzed cross-sectional data from the CARRA Registry across 60 sites between 2010 and 2013. Patterns of comorbidities in children with JIA and SLE were assessed. Network analysis of disease co-occurrence was carried out to identify clusters of comorbidities that are likely to present in the same individual.

Comorbidities were present and captured (1 or more) for subjects in the registry according to disease type:

JIA: uveitis, fibromyalgia, sarcoid, autism, autoimmune hepatitis, autoimmune thyroiditis, celiac disease, cerebral palsy, congenital heart disease, cystic fibrosis, type 1 diabetes (T1D), type 2 diabetes (T2D), malignancy, immunodeficiency, asthma, Down syndrome, demyelinating disorder, seizure, pulmonary hypertension, psoriasis, IBD, interstitial lung disease.

SLE: chronic vasculitis, autoinflammatory disease, uveitis, fibromyalgia, sarcoid, autism, autoimmune hepatitis, autoimmune thyroiditis, celiac disease, cerebral palsy, congenital heart disease, cystic fibrosis, T1D, T2D, malignancy, immunodeficiency, asthma, Down syndrome, demyelinating disorder, coronary heart disease.

Results: 6,150 children with JIA and 1,067 children with SLE were enrolled in the CARRA registry. 20.5% of JIA and 11.5% of SLE patients had at least 1 comorbidity (Table 1). Multiple co-occurring co-morbidities were individually much less common (0.6% or less), including in known associations (e.g. JIA with Down’s syndrome and hypothyroidism or congenital heart disease).

Conclusion: The prevalence of comorbidities in children with JIA and SLE in the CARRA Registry is in accordance with ranges reported in other studies. We analyzed multiple co-occurring co-morbidities, including a network clustering analyses, which showed established and potential associations that are candidates for further evaluation in larger, population-based data sets.