2.4 Statistical analysis
Descriptive analysis of the baseline characteristics, disease phenotypes and covariates of the study population was performed, stratifying by presence of FA. Continuous and categorial variables were tested using one-way ANOVA and Chi2-test, respectively. To investigate the associations of ever eczema and eczema phenotypes with FA presence, multinomial logistic regression analysis with and without adjustment for all covariates (excluding age at baseline) was conducted. Association of ever eczema and eczema phenotypes with presence of asthma and hay fever were analyzed using logistic regression analysis with and without adjustments for all covariates (excluding age at follow-up). The trajectories of the different allergic diseases in the study population, stratified by FA presence, was based on the prevalence of eczema, asthma and hay fever within each age bracket. Differences in prevalence of each allergic disease between the FA groups were tested using a Chi2-test with post-hoc group-wise comparison within each age bracket.
Data management tasks and statistical analysis were carried out using R Studio (2022.02.0443, based on R version 4.1.2, RStudio Team (2022), RStudio: Integrated Development Environment for R. RStudio, PBC, Boston, MA URL http://www.rstudio.com/). Relevant packages are listed in the supplement. The significance level for all tests conducted was a p-value of 0.05.