Statistics
Data were examined for calculation and distribution for statistical
analysis SAS v9.4 (SAS institute, Cary, NC). A power analysis performed
prior to the study showed that with a sample size of 44, an alpha level
of 0.05 and power ≥ 0.80, we could detect a significant association
between the clinical variables with iPFT z-scores when the correlation
was as low as 0.41. Statistical significance was defined as p ≤ 0.05.
For categorical variables, frequency and percentage analysis were
performed. For continuous iPFT variables, mean and 95% confidence
intervals were calculated. Chi-square analysis was used to evaluate the
proportion of infants in each group with abnormal FRC vs. forced
expiratory volume in 0.5 seconds (FEV0.5) parameters. To
determine the association between WFA/WFL z-score and lung function,
Pearson correlation coefficients were calculated. Associations between
lung function and categorical clinical risk factors were calculated
using Student’s t-test or Mann-Whitney U test as appropriate.
Correlation between FRC and respiratory rate (RR) was determined using
Spearman rank-sum correlation coefficients. We conducted additional
analyses stratifying infants into normal versus obstruction or
hyperinflation. For these analyses, we defined obstruction as an
FEV0.5 z-score < -2 and hyperinflation as FRC
z-score > 2. To identify the respiratory rate cutoff that
yielded the best combination of sensitivity and specificity in
predicting hyperinflation, we used logistic regression with Youden’s
Index.