Evaluate the association between dexamethasone dosing and outcomes for children hospitalized with croup.
This study was nested within a multisite prospective cohort study of children aged 6 months to 6 years admitted to 1 of 5 US children’s hospitals between July 2014 and June /2016. Multivariable linear and logistic mixed-effects regression models were used to examine the association between the number of dexamethasone doses (1 vs >1) and outcomes (length of stay [LOS], cost, and 30-day same-cause reuse). All multivariable analyses included a site-specific random effect to account for clustering within hospital and were adjusted for age, sex, race and ethnicity, presenting severity, medical complexity, insurance, caregiver education, and hospital. In cost analyses, we controlled for LOS.
Among 234 children hospitalized with croup, patient characteristics did not differ by number of doses. The proportion receiving >1 dose varied by hospital (range 27.9%–57.1%). In adjusted analyses, >1 dose was not associated with same-cause reuse (odds ratio 0.87 [95% confidence interval (CI): 0.26 to 2.95]) but was associated with 45% longer LOS (relative risk = 1.45 [95% CI: 1.30 to 1.62]). When we controlled for LOS, >1 dose was not associated with differential cost ($−31.2 [95% CI $−424.4 to $362.0]). Eighty-two (35%) children received dexamethasone before presentation.
We found significant interhospital variation in dexamethasone dosing and LOS. When we controlled for severity on presentation, >1 dexamethasone dose was associated with longer LOS but not reuse. Although incomplete adjustment for severity is one possible explanation, some providers may routinely keep children hospitalized to administer multiple dexamethasone doses.