AE Bayesian Hierarchical Model

This process fits the multi-level Bayesian hierarchical models of Berry and Berry (2004) ¹ and Xia, Ma, and Carlin (2011) ² . Adverse events are modeled taking into account a grouping variable , such as system organ class .

What do I need?

This process requires several demographic- and adverse event-related variables. These include:

•

USUBJID ,

•	A treatment variable (the actual treatment received by each subject ( TRT01A )), the planned treatment (intent-to-treat) for each subject ( TRT01P ), or the description of the planned treatmen arm ( ARM ),

•	Term- ( AEDECOD ) and grouping-level ( AEBODSYS ) variables from the AE domain,

•	AESTDTC from the AE domain, and

•	RFSTDTC and RFENDTC from DM or ADSL .

Variables can be taken from the AE domain (or ADAE ) and either from the subject level analysis data set ( ADSL ) included in the Analysis Data Model ( ADaM ) folder or from the DM and AE domains in SDTM . Refer to Localization-Specific Value Specification for more information about these data sets.

Output/Results

The output generated by this process is summarized in a tabbed report. Refer to the AE Bayesian Hierarchical Model output documentation for detailed descriptions and guides to interpreting your results.

---

1

Berry, S.M., and Berry, D.A. (2004). Accounting for Multiplicities in Assessing Drug Safety:

A Three-Level Hierarchical Mixture Model. Biometrics 60 , 418-426.

2

Xia, H.A., Ma, H., and Carlin, B.P. (2011). Bayesian Hierarchical Modeling for Detecting Safety Signals in Clinical Trials. Journal of Biopharmaceutical Statistics 21 , 1006-1029.