Statistical analysis
Summary statistics were presented as median with interquartile range
(IQR) or percentage where appropriate. Wilcoxon rank sum test or Pearson
Chi-square test were used to compare variables between wheezing and
non-wheezing patients. The 14 tests were Bonferroni-corrected and
interpreted based on Type I error rate of α=0.0036. To assess the
association between demographic and clinical characteristics, and
viruses with the primary outcome of interest (i.e., wheezing),
multivariable logistic regression was used to determine the odds ratios
(OR) for wheezing across a-priori selected predictors of
interest, including: age, sex, smoke exposure, breastfeeding history,
birthweight, type of delivery, prematurity, personal, mother and family
history of asthma, and virus type. All analyses had a significance level
of 0.05 (two-tailed) and were performed using StataCorp, College
Station, TX software (version 15.1).