How to Extend Metadata

If data are in scope for a domain defined in this guide, but not all data can be collected using the domain's collection fields, then fields may be added from the CDASH Model or created using guidance below. Prior to adding fields from the CDASH Model or considering a new field, confirm that none of the fields in the domain will fit the need. Fields may be added from the CDASH Model or created only when data are different in nature and are not in scope for existing domain fields.

If a logically related grouping of data is in scope for a domain defined in this guide, but not all data can be represented using the domain variables, then data may be represented by adding variables from the SDTM to the domain or by using a Supplemental Qualifier dataset related to the domain. Prior to adding variables from the SDTM or considering a Supplemental Qualifier dataset, confirm that none of the existing domain variables will fit the need. Variables may be added to a domain and Supplemental Qualifier datasets may only used when data are different in nature and are not in scope for existing domain variables.

Once confirmed, determine whether adding a variable from the SDTM will fit the need. If adding a variable from the SDTM meets the need, then a Supplemental Qualifier dataset will not be implemented. Supplemental Qualifier datasets will only be implemented for domains when data are not in scope for a variable defined in the SDTM.

The overall process for extending a domain is as follows:

Select variables for the domain from the SDTM. Selection of variables must align with SDTM Usage Restrictions. The general approach for adding variables to a domain from the SDTM is to:
1. Include applicable Identifier variables.
2. Include applicable Qualifier variables in alignment with the SDTM General Observation Classes.
3. Include applicable Timing variables.
4. Apply the domain code to the variables as appropriate by replacing all variable prefixes (shown in the SDTM as “--“) with the domain code.
5. Set the order of variables consistent with the order defined in the SDTM.
6. Adjust the labels of the variables only as appropriate to convey their meaning in the context of the data in domain. Use title case for all labels.
7. Ensure that appropriate standard variables are being properly applied by comparing their use in the domain to their use in related TIG domains.
Place any non-SDTM variables in a Supplemental Qualifier dataset when variables cannot be selected from the SDTM.

Adding Fields from the CDASH Model

The overall process for extending a domain is as follows:

During the development of conformant CRFs the tabulation domain to which the collected data will be mapped must be determined. The choice of the domain will be based on to use does not depend upon the mode of transmission, the methodology used to generate the data, the medium used to store the data, the person who recorded the data, or the subject described by the data. The SDTMIG domain to be used affects what CDASH variable names, question texts, prompts, controlled terminology, and so on, to use. CDASH suggests a format to be presented to those entering the data, but it does not dictate any data structure in which to store the collected data (often referred to as a data management operational database).

Creating New Fields

When considering a new data collection field, determine whether the data will be used for an operational use case, such as data cleaning, or are actual data to be represented in a tabulation dataset. In general, new data collection fields (not already defined in the CDASH Model) will fall in one of following categories:

A field used for operational, Data Cleaning purposes only.
A field used to collect data that have Direct Mapping to a target variable in the tabulation dataset.
A field used to collect data that have No Direct Mapping to a target variable in the tabulation dataset.

Prior to adding a custom field to a CRF, confirm that none of the collection fields in this guide or in the CDASH Model fit the need.

Once confirmed, the following guidance will be used to add a new field to a CRF.

Num	Field Category	Target Tabulation Variable	Implementation
1	Data Cleaning	NA	The field --YN with Question Text "Were there any [interventions/events/findings]?" can be used for this purpose. Replace the two dashes (--) with the two-character domain code and create the Question Text or Prompt using generic Question Text or Prompt from the CDASH Model as a base. Always create custom data-cleaning/operational variables using consistent naming conventions.
2	Direct Mapping	Yes	If a value can be collected exactly as it will be reported in the tabulation dataset (i.e., same value, same data type, same meaning, same controlled terminology), the tabulation variable name will be used as the data collection variable name in the operational database to streamline the mapping process. Extensions may be appended if needed to create a unique variable name in the collection database. Any collection variable whose meaning is the same as tabulation variable will align with tabulation variable and the meaning will not be modified for data collection.
3	No Direct Mapping	Yes	If a value cannot be collected in alignment with the tabulation dataset variable (e.g., collected data type is different from the data type in the corresponding tabulation variable) or if the tabulation variable is derived from the collected value, then the operational database should use a collection variable with a different name from tabulation variable into which it will be mapped.
3	No Direct Mapping	No	If a field does not align with a tabulation variable, a unique name should be assigned based on applicant business rules using CDASH naming fragments (e.g., --DAT, --TIM) as appropriate and CDISC variable naming fragments, found in Section x.x, CDISC Variable-naming Fragments, where possible.

Page tree

How to Extend Metadata