OBJECTIVE: The aim of this study was to describe a model for predicting well-being in children with juvenile chronic arthritis (JCA). METHODS: 125 children (43 boys) (median age 14.1 yrs; range 10.3-17.8) rated disability and discomfort (Childhood Health Assessment Questionnaire). Pain control, pain reduction and fatigue were evaluated (visual analogue scales). In addition, variation of pain intensity was rated by a pain intensity scale. Analysis by the stepwise regression technique was used to explain the variability in well-being. Eight independent variables were included as possible predictors in the model (p < 0.1). RESULTS: The analyses indicated that well-being in children with JCA is related to three clusters of variables; pain "as it normally is", number of pain-free days and attending physical education classes. The analysis explained a substantial portion of the total variance in the children's well-being (55.1%). CONCLUSION: Pain is a robust predictor of well-being in children with JCA. This supports the concept of the benefits of reducing chronic joint pain as a major goal in caring of these children.