Characteristics Associated with Variation in Corticosteroid Exposure in Children with Steroid-Sensitive Nephrotic...

Variation in dose and duration of corticosteroids for childhood-onset steroid-sensitive nephrotic syndrome occurs worldwide, likely reflecting the evolving evidence on optimal dosing and variable severity of the disease observed between patients. We conducted a study to determine the associations between site, physician, and patient factors, and average daily corticosteroid dose and duration of therapy.

Data were derived from the Canadian Childhood Nephrotic Syndrome (CHILDNEPH) Project, an observational longitudinal study from 2013 to 2019 of children with nephrotic syndrome involving pediatric nephrologists in 11 sites across Canada. The primary outcome was average daily corticosteroid dose prescribed per episode of proteinuria, reported as mg/m² prednisone equivalents. Secondary outcome was duration of treatment for each episode of proteinuria in days. Exposure variables were categorized into site-, physician-, and patient-level variables.

In total, 328 children, median age at enrollment of 4.3 years old (interquartile range [IQR], 3.6), participated and were followed for a median time of 2.62 years (IQR, 2.6). The observed variability in average daily corticosteroid dose and in duration of therapy was mostly attributed to the site where the patient was treated. Accounting for between patient, physician, and site differences, average daily corticosteroid dose decreased with increasing age (beta coefficient, -0.07; 95% confidence interval [95% CI], -0.09 to -0.05], P<0.001). African and Indigenous ethnicity was associated with longer treatment duration compared with White patients (beta coefficient: African, 42.29, 95% CI, 7.85 to 76.73, P=0.02; Indigenous, 29.65, 95% CI, 2.79 to 56.52, P=0.03).

We found practice variation with respect to corticosteroid prescriptions across 11 Canadian sites, and that variation is mostly explained at the site level. Age and ethnicity are important factors to be considered, because they are significantly associated with the average corticosteroid dose and duration of therapy.