Objective To explore the application of quantile regression models in public health and its SAS implementation, so as to provide a reference for the promotion of this method.
Methods Based on the introduction of the basic concept of quantile regression model, the application situation and advantages of quantile regression model are introduced. The application of quantile regression model and its SAS implementation are discussed by performing simple linear regression model as a contrast when data of serum Clara cell protein levels and FEV1/FVC changes were used as an example.
Results Compared with the simple linear regression model, the quantile regression model could not only analyze the impact of serum Clara cell protein levels on the mean FEV1/FVC, but also analyze the effects of serum Clara cell protein levels on different quantiles of FEV1/FVC to obtain more comprehensive information, which could be easily achieved by statistical software SAS.
Conclusion Quantile regression models can compensate for the deficiency of simple linear regression models that only focuses on the mean value of the dependent variables but not complete distribution characteristics, and software SAS provides relatively mature analytic expressions, which is worth promotion.