Statistical analysis
Descriptive statistics were presented for all outcome variables and covariates. Analyses of progression of AR, progression or occurrence of asthma were carried out using a general linear regression model, with the ratio of annual number of prescriptions in the follow-up period vs the pre‐index period used as the outcome variable. Analysis of AR and asthma medication intake was achieved by logistic regression with a stepwise selection of significant covariates. The individual length of the analytical time span was included as a covariate. Time to asthma medication intake was investigated using survival analysis. The proportion of patients with any level of treatment between the AIT and non‐AIT control groups was also analysed by logistic regression. Statistical analyses were based on two‐way testing without exception. For all statistical tests, significance level was set to 5% (P < 0.05). Analyses were performed using SAS 9.4 software SAS Institute, Inc., Cary, NC, USA.
Results