Skip navigation

Predictors of pain in JIA

Clinical bottom line

Disease related factors only account for a small proportion (22%) of the variation in pain scores in children with JIA.


Background

Pain is an importnat symptom in children with JIA, but factors that might predict pain are nort well explored.

Reference

PN Malleson et al. Predictors of pain in children with established juvenile rheumatoid arthritis. Arthritis & Rhemuatism 2004 51: 222-227.


Study

Children with JIA were diagnosed according to ACR criteria, and further subcalssified into rheumatoid factor (RF) psoitive (RF+) or negative (RF-). Information was collected as part of a long term outcome study, in which onset of disease was at leats five years before the study, and a minimum age of eight years.

There were 388 children who had full data sets, including a visual analogue pain score, which was the outcome measure. Explanatory variables were sought from age, sedx, ethnicity, disease subtype, active disease duration, joint count, physician global assessment, and morning stiffness. Univariate analysis and mulivariate analyses were used to assess the explanatory variables, and analyses were further stratified by age 8-15 years (n=183), and 16 years or older (n=205).

Results

Children in the study had a mix of different types of juvenile arthritis, with active disease duration of about six tyears on average, with wide variation between a few months and 22 years). There was a reasonable variation in disease related factors in both groups.

In all the children, univariate analysis showed significant association between pain and active disease duration, joint count, physician global assessment (Figure 1, in which 0 = inactive, 1 = mild activity, 2 = moderate activity, 3 = severe activity), and onset subtype. For the latter, pain scores were higher with polyarticular disease and RF+.

Figure 1: Physician global assessment and median pain score

In all the children, multivariate analysis showed significant association between active disease duration, physician global classification and age at study entry. In children aged 8-15 years, significant associations were with active joint count and age at entry. In children aged 16 years or older, significant associations were with active disease duration and physician global assessment of moderate or severe disease.

The independent predictors only accounted for a small amount of the variation in pain scores, about 22%.

Comment

Disease related factors only account for a small proportion of the variation in pain scores in children with JIA. It makes some sense that in younger children active joint count is important, while in older children it is disease severity and duration of active disease that are important. But what this shows in that other factors are important in pain in JIA.

The strength of the study is its large size following children over a long period. It seems to confirm smaller studies that also found little association between disease-related factors and pain.