ADaM Grouping

When designing ADaM analysis datasets, you may have various needs for data categorization. A primary need to categorize data is to group similar values for your analysis.

ADaM standard allows you to create variables that categorize the information in your dataset. These grouping variables are to be used during the analysis.

New PARAM and PARAMCD iteration can be created if the standard mapping rules give you limitation when creating grouping variables. In that way you will avoid mapping rule violation.

PARCAT PARAMETER CATEGORIZATION PARCATy

This kind of grouping variable is intended to group PARAM values into categories. PARAM to PARCATy mapping has to obey the many-to-one mapping rule. That means that one PARAM may be associated with at most one level of PARCATy. Often times, people try to use PARCATy to subdivide or qualify PARAM values further. The many-to-one mapping rule does not allow that.

GR GROUPING VARIABLES

*GRy and *GRyN types of variable are general grouping variables. Variables whose names end in GRy, Gy, or CATy are grouping variables. When creating this kind of grouping variable it is recommended that the grouping or categorization variables begin with the name of the variable being grouped and end in GRy ( For example: Variable ABCGRy is a character description of a grouping or categorization of the values from the ABC variable for analysis purposes).

The ADaM implementation guide mentions some grouping variables that are usually used in studies. Here is some examples:

• SITEGRy
• RACEGRy
• AGEGRy
• TRxxPGy/TRxxAGy/TRTPGy/TRTAGy

Note : For variables that starts with TRT the R in GR is dropped here due to variable length issues.

CAT ANALYSIS VARIABLE CATEGORIZATION VARIABLES

CAT categorization variables are generally used to categorize the BDS AVAL/BASE/CHG/PCHG values. Here is some examples of such a variable:

• AVALCATy
• BASECATy
• CHGCATy
• PCHGCATy

Variables which name contains CAT are used as a categorization of the analysis variable (e.g., AVAL/AVALC) within onelevel of PARAM.

(M)CRITY SELECTION VARIABLES

(M)CRITY are used to categorize records to a specific criterion. The original intent behind the creation of (M)CRITy was to select subgroups of records that met a given criteria. MCRITy is populated with a text description identifying the criterion being evaluated.

A flag variable (Named CRITyFL) can be associated with (M)CRITY. These flag variables indicate when the criterion defined in (M)CRITy is met within the record. A complex criterion which draws from multiple rows (different parameters or multiple rows for a single parameter) will require a new PARAM be created.