Statistical methods
The results obtained from the questionnaires were analyzed using classical descriptive statistics for the study population as a whole, and by study group to compare patients who attended the gym (GYM group) with those who did not (No-GYM group). The scores were calculated with the PedsQL™ 4.0 scoring algorithm, and the medians and interquartile ranges (IQR) were reported for each item and each dimension, and for total scores (if any), for each PedsQL questionnaire.
Differences between the two groups in terms of patients’ characteristics, and their scores in the questionnaires (by single dimension and total scores) were assessed with Fisher’s exact test or the Wilcoxon test. Possible correlations between total scores (for physical health, psychosocial health, and fatigue) were computed using Pearson’s correlation index.
The statistical analyses were performed with SAS™ (SAS Institute, Cary, NC) and R software (R Foundation for Statistical Computing, Vienna, Austria).