Background Juvenile chronic/idiopathic arthritis (JCA/JIA) is a heterogeneous disease. Some long-term studies are published, but few are population based and predictions of long-term outcome are incompletely defined.

Objectives The aim of this study was to investigate disease course, outcome and predictors of outcome in an unselected population based cohort diagnosed with JCA followed for 17 years.

Methods This cohort consists of 132 incidence patients identified 1984 to 1986 in southwestern Sweden according to EULAR criteria for JCA. At 5-year follow-up 129 individuals were examined with joint assessment, laboratory measurements, radiographic examinations, as well as medication and functional assessment. At 17-year follow-up the individuals were examined with joint assessment, laboratory measurements, medication, Health Assessment Questionnaire (HAQ), Keitel functional test (KFT) and short form 36 health survey (SF-36).

Results At 17-year follow-up 86 out of 132 individuals (65%) were examined. The individuals lost to follow-up did not significantly differ with respect to baseline and 5-year data from the individuals examined at 17 years. Forty percent of the individuals were in remission. The median HAQ score was 0.0 (0.0 – 1.5) and the median KFT 100 (54 – 100). Fifty one (59%) individuals had HAQ score >0 and/or KFT <100. SF-36 was significantly lower in all domains compared to a reference group. During the 17 years 44% changed subgroups. Thirty-nine percent of the individuals in remission at 5-year follow-up were not in remission at 17-year follow-up. In multivariate analyses, remission at 17-year follow-up was predicted by remission at 5-year follow-up (OR 4.8, 95% CI 1.8 – 12.5). At 17-year follow-up disability defined by HAQ >0 was predicted by RF-positivity at 5-year follow-up (OR 3.6, 95% CI 1.0 – 13.3) and disability defined by KFT<100 was predicted by non-remission at 5-year follow-up (OR 11.3, 95% CI 2.7 – 47.9) and RF-positivity at 5-year follow-up (OR 5.6, 95% CI 1.0 – 30.6). At 17-year follow-up the physical summation score in SF-36 better than the average of the reference group was predicted by remission at 5-year follow-up (OR 5.8, 95% CI 2.2 – 15.4). Long-term outcome was not predicted by any baseline variables in multiple logistic regression.

Conclusions In a population based cohort of patients with JCA, 40% were in remission but more than half of the individuals had some physical disability and the cohort had lower health related quality of life compared to a reference population. Many individuals changed subgroups and disease activity. Long-term outcome was best predicted by and associated with characteristics at 5-year follow-up and not by onset variables.

Disclosure of Interest None Declared