Methods In this single centre clinical service evaluation, a retrospective case note review of 105 consecutive JIA patients attending outpatient follow up clinics in 2011–12 was undertaken. The completion rate of a standardised ACR COV proforma (present in all case notes) was determined, and JADAS was calculated from additional clinical data. Pearson’s correlation and logistic regression were used to assess the impact of individual items on changes in JADAS.

Results 105 children with JIA had 193 clinic visits but complete ACR COV data were found in only 68/193 records (35%). Of the 6 items comprising the ACR COV, the ESR accounted for the majority of missing data. Sufficient data was available to calculate JADAS scores in 22 children with two consecutive outpatient visits. The table indicates that changes in JADAS were most dependent on the physician’s global assessment, and least dependent on ESR (Pearson correlation). Stepwise regression showed that the physician’s global assessment alone predicted 87.5% of JADAS change and the ESR contributed an additional 3.3%.

Conclusion In a routine clinical setting, frequent missing data reduced the potential clinical utility of the ACR COV and JADAS composite scores. We speculate that a composite clinical score which does not rely on recording the ESR may improve completion rates without diminishing clinical utility.