Objectives To define distinct disease course groups among children with JIA based on observed changes in quality of life, pain, medication requirements, medication side-effects and active joint counts during the first 5 years of the disease. These variables were prioritized by patients, parents and clinicians in a previous study.

Methods We used data from the Research in Arthritis in Canadian Children emphasizing Outcomes (ReACCh-Out) prospective inception cohort. We included 618 children newly diagnosed with JIA between 2005 and 2010 with information for at least 6 study visits within the first 5 years after diagnosis. Health-related quality of life and pain were measured with validated 10 cm visual analogue scales. Each current medication was given a weight and weights were added to obtain a medication requirements score. Medication side effects were reported using the Juvenile Arthritis Quality of Life Questionnaire and were weighted according to how frequently they were experienced by the child to obtain a side-effects score. Active joint counts (AJC) were reported by the attending pediatric rheumatologist. We grouped patients by course using multivariable cluster analysis and K-means clustering. Silhouette coefficients, R-Square statistics and clinical judgment were used to select the ideal number of clusters. The frequency of each disease course was described by JIA category.

Results Four clusters provided the best discrimination: 1) Mild (33.7% of children)-almost normal quality of life throughout the disease course with initial mild pain and low AJC requiring minimal treatment, followed by normalization; 2) Moderate (43.0%)-moderate initial impact on quality of life and mild to moderate pain and AJC, followed by normalization; 3) Severe Controlled (11.6%)-moderate initial impact on quality of life and moderate pain levels with high AJC requiring aggressive treatment, followed by normalization; 4) Severe Persisting (11.6%)-persisting mayor impact on quality of life and moderate pain with moderate decreasing AJC, but ongoing treatment needs and side-effects. Children with oligoarthritis most often followed a Moderate course. Almost half the children with RF negative polyarthritis, systemic and psoriatic JIA followed a Mild course. Children with RF positive polyarthritis most often followed a Severe Controlled course. The Severe Persisting course was observed in all JIA categories but it was infrequent in systemic JIA and oligoarthritis.

Conclusions Using patient-relevant variables the course of JIA can be described by four disease course groups with two of them reflecting a severe disease course, one that responded to treatment despite severe initial presentation and one with persisting impact on quality of life and pain despite moderate decreasing AJC.

Disclosure of Interest None declared