Reliability of exercise testing and functional activity questionnaires in children with juvenile arthritis


  • identifier: NTC00213187.



To determine the reliability of formal exercise testing and the reliability of functional and activity questionnaires in children with juvenile idiopathic arthritis (JIA).


Children with JIA of any subtype ages 8–16 years who were recruited to a randomized trial comparing different exercise therapies participated in 2 preintervention sessions of exercise testing 2–6 weeks apart. Exercise testing included 1) submaximal oxygen uptake (VO2submax), 2) peak VO2 (VO2peak), and 3) anaerobic power using modified Wingate tests (Want). Two physical function questionnaires (the Childhood Health Assessment Questionnaire [C-HAQ] and Revised Activity Scale for Kids [ASK]) and 1 daily physical activity questionnaire (the Habitual Activity Estimation Scale [HAES]) were also completed at these times. Test–retest reliability was assessed using type 3, intrarater intraclass correlation coefficient (ICC3,1) and Bland and Altman plots were used to determine limits of agreement.


Data were available for 74 patients (58 girls). VO2submax, VO2peak, and Want demonstrated high reliability (ICC3,1 0.82, 0.91, and 0.94, respectively). C-HAQ and ASK questionnaires also had very high reliability (ICC3,1 0.82 and 0.91, respectively). The HAES demonstrated low reliability for total activity score (ICC3,1 0.15) and moderate reliability when the number of very active hours was analyzed separately (ICC3,1 0.59).


Results of this investigation suggest that exercise testing and functional questionnaires in children with JIA are consistent and reliable. Reliability of the HAES total score was poor, but moderate when the very active hours subscale score was used.