Statistical Analysis
The Pre-E cohort was evaluated as a single group, as well as divided
into 2 groups, non-severe Pre-E and severe Pre-E, using the American
College of Obstetrics and Gynecology criteria for Pre-E with severe
features [18]. Basic clinical and demographic comparisons were
performed using Chi-Square tests for categorical variables and Students
t-tests and Wilcoxon rank-sum tests for continuous variables, depending
on the linearity of the data. Pairwise comparisons were made using a
Bonferroni correction. To analyze wheezing outcomes, a participant level
variable of ever/never wheezed was created for participants who had at
least one survey response in the the first six months of surveys, as
well as after 6 months. We then analyzed all the survey data from
infants meeting this criterion. For each of the main outcomes, simple
bivariate analyses were first performed and those variables with
significance of P≤0.10 were included in a multivariable model. In
addition to these variables, we included clinically relevant ones, e.g.
sex, race, etc. For the outcome of wheezing, using the dichotomous
wheezing outcome as described above, logistic regression analysis was
performed using multivariable model including Pre-E group, sex, GA,
mother’s smoking history, family history of asthma, and antenatal
steroid use. IPFT outcomes were analyzed with raw data adjusting for
race, sex, and body length at testing in multivariable models. Simple
t-tests were used for the bivariate analyses and ANCOVA models were
performed for the adjusted models, with the covariates sex, race,
length, and GA. Fetal growth restriction, antenatal steroid use, and
family history of asthma were also included as covariates based on the
results of the bivariate analysis. All analytic assumptions were
verified, collinearity was assessed for multivariable models, and
analyses were performed using SAS v9.4 (SAS Institute, Cary, NC).