Statistical analysis.
Data were described with frequencies for categorical variables and means
(SD) or medians (interquartile range [IQR]) for continuous variables
(depending on normal or non-normal distribution), both in the total
population and stratified by type of pneumonia (Table 1).
χ2 or Fisher’s tests were applied to assess
differences across groups for categorical variables, and Student’s t
test or the Mann–Whitney U test was used for continuous variables.
Two-tailed p<0.05 was considered statistically significant.
Univariable comparisons were segmented by the presence (Yes) or absence
(No) of patients’ features among children with SARS-CoV-2 CAP and other
viral CAP. We also performed univariate comparisons for different
outcome endpoints (admission to PICU, complications, etc., to test for
differences between patients with COVID-19 CAP or other virus-associated
CAP. The latter analysis consisted of stepwise multivariable binary
logistic regression, with the endpoint being PICU admission (Table 2).
The multivariable model was adjusted by type of pneumonia, sex, age
(years), asthma, respiratory rate, oxygen saturation, wheezing,
shortness of breath or work breathing, radiological image
interpretation, leukocytes, C-reactive protein (CRP), neutrophils,
lymphocytes, sodium, albumin, procalcitonin and hemoglobin. The optimum
model was selected according to Akaike Information Criteria (AIC).
REDCap data were exported to the R language (4.0.3)15for analysis. R packages were used for specific analysis, such as
compare Groups (4.4.6) for comparisons or MASS
(7.3.53)16 for stepwise logistic regression.