Regency EF
Model-Based Standardization Using an Outcome Model with Random Effects (304058)
Adel A. Alrwisan, University of Florida
Babette A. Brumback, University of Florida
*Zhongkai Wang, University of Florida
Almut G. Winterstein, University of Florida
Keywords: Causal Inference, Confounding, Generalized Linear Mixed Models, Marginal Effect, Model-based Standardization
Model-based standardization uses a statistical outcome model or exposure model to estimate a population-average association that is unconfounded by selected covariates. We develop an approach based on an outcome model, in which the mean of the outcome is modeled conditional on the exposure and the confounders. In our approach, there is a confounder that clusters the observations into a large number of categories. We treat the parameters for the clusters as random effects. We use a between-within model to account for the association of the random effects not only with the exposure but also with the cluster sizes. We review alternative approaches in the literature, and we compare our approach to the exposure-modeling approaches incorporating random effects. We illustrate our method with simulation studies and apply it to real-world data application in medical research. We use 2014 Truven Health MarketScan Research data to compare proportions of Acute Respiratory Tract Infection diagnoses with an antibiotic prescription for emergency department versus outpatient visit, adjusting for confounding by unmeasured patient-level (cluster) and measured diagnosis-level (observation) variables.