AyIND, ByIND, and CRITy vs. MCRITy vs. AVALCATy

Introduction

The ADaMIG provides multiple ways to categorize and flag values to be analyzed. These ways may seem overlapping, and the most appropriate set of variables for a particular analysis may not be obvious to the user. This article is intended to provide some clarity on how to use appropriate variables and how to select the appropriate analysis variables.

Advice/Instructions/Guidance

This article covers 4 groups of methods/variables:

• AyIND/ByIND range indicator and associated variables
• CRITy and associated variables
• MCRITy and associated variables
• AVALCATy variables

AyIND/ByIND range indicator and associated variables are used to indicate if the analysis value is within a prespecified range. AyLO and AyHI are usually but not necessarily constants, and can vary within a subject and/or parameter. Note that AyIND/ByIND values are based upon AVAL and corresponding ranges only and do not include change from baseline or other values. AyLO/AyHI are required if analysis contains ratio of AVAL to ranges (R2AyLO/R2AyHI).

CRITy and associated variables are used to identify whether a row meets prespecified binary criteria CRITy within a parameter. CRITyFL can take on values of Y/N/null or Y/null, depending on whether it is important for analysis to distinguish between negative and missing values. The criteria can be based on a single analysis variable AVAL, or on a combination of AVAL and other variable values present on the same record (e.g., demographics, baseline or change from baseline). CRITy for one parameter can be different than CRITy for a different parameter in the same dataset. 

MCRITy and associated variables, like CRITy, are prespecified within a parameter, but have multiple possible values instead of a binary outcome. As with CRITy, the criteria can be based on the values of any of the variables present on the record. Since there are more than 2 possible outcomes, unlike CRITyFL there is no MCRITy flag variable; instead, outcomes are stored in MCRITyML and corresponding numeric values in MCRITyMN.

AVALCATy variables are simple categorizations of AVAL or AVALC within a parameter, and do not depend on the values of other variables on the record. For example, pain score values ranging from 0 to 10 could be categorized from No Pain (0) or Light Pain (1-2) to Worst Pain (9-10). Unlike the AyIND variables, where AVAL or AVALC are compared to a specified analysis range, AVALCATy can only be assigned based on values of AVAL or AVALC alone. Within a parameter, each distinct AVAL/AVALC value is associated with a single AVALCATy value, whereas a given AVALCATy value may correspond to multiple AVAL/AVALC values.

CRITy and MCRITy can use any variable(s) located on the row. For all of these methods/variables except AyIND/ByIND, the definition must stay constant across all rows within the same value of PARAM.

Example

This example uses a single lab test to show how all the 4 groups of variables are used for analysis. In this example, study treatment is expected to have an effect on calcium values. Hence, the lower and upper ranges for calcium are expected to change based on subject medical history and treatment duration; calcium A1LO and A1HI values change over time within a subject, and A1IND contains the indicator values based on these changing ranges. If the analysis needs to identify records where calcium >10, then CRITy variables can suffice. In this example, CRIT1 defines a single binary criterion for calcium, AVAL > 10; CRIT1FL is assigned a value of "Y" if the criterion is met, and a value of "N" otherwise. MCRIT1 and MCRIT1FL are used to represent the amount that the toxicity grade increased from baseline. Because multiple outcomes (e.g., no grade increase, increase of 1 grade) are possible, MCRIT1ML represents more than 2 possible values. 

References and Resources

ADaMIG v1.3, Section 3.3.7, 4.7.1, 4.7.2

Watson R, Miller K, Slagle P. Proper Parenting: A Guide in Using ADaM Flag/Criterion Variables and When to Create a Child Dataset. 2015; accessed August 19, 2026.

Shostak J. ADaM Grouping: Groups, Categories, and Criteria. Which Way Should I Go? PharmaSUG 2017 Paper DS17. Accessed August 19, 2026.