WITHDRAWN: Hierarchical Poisson Regression Model with Application to Efficacy Data of ZOSTAVAXTM for Evaluating Vaccine Waning Effect

• Author(s): Shu-Chih Su
• Keywords: Bayesian ; Poisson regression

Abstract

In the work, we consider using Bayesian approach at the Efficacy Data of ZOSTAVAXTM Shingles Prevention Study (SPS) and Short-Term Persistence Study (STPS)) to investigate whether there is waning effect of the vaccine efficacy in terms of the occurrence of Herpes Zoster (HZ)cases. Bayesian method can be useful when the data set may be large but only a few data points contributed in the analysis. This is true for Efficacy Data of ZOSTAVAXTM because the HZ count is relative rare as compared to the total number of the subjects in the study. In such situations standard frequentist estimators might be inappropriate since the likelihood may not be well-approximated by a normal distribution. Poisson regression has been conducted utilizing the frequentist method in pervious work to examine the waning effect and found that the effect of time since vaccination on vaccine efficacy (waning effect