Metadata tables guide implementation of the CDISC CDASH model Collection metadata in this guide are instructions for implementing the CDASH Model to develop CRFs and the underlying database or other data collection structure. Domain specifications Collection metadata for the TIG are provided in the following sections:
Section 2.7.6, Metadata for Individual Health. Guidance in this section describes how to read collection metadata provided in Section 2.7.6. Implementers should refer to Section 2.7.6 when reading this guidance.
Collection metadata will be used with guidance in this section and are organized with 1 set of metadata per domain.All collection metadata begin with a description of the domain. A metadata table follows to describe data collection fields and their attributes. Assumptions for the domain are also provided to further guide implementation. Metadata tables are structured to present one row 1 or more rows for each data collection field with columns to describe expected attributes of the field and mapping of collected data in resulting datasets. The order of variables in a domain specification table reflects the expected order of variables in the resulting dataset.
CDASHIG metadata table attributes provide building blocks for the development of a CRF and the underlying database or other data-collection structure.
CRF and Data Management System Design Metadata
...
for each field. More than 1 row will be present in a metadata table when data collection for a field can be performed using different scenarios/options. Multiple data collection options may also be provided in single cell within a row for a field with each option separated by a semicolon.
Each column in a metadata table provides domain specific guidance for data collection, including guidance for:
- CRF and System Design, with attributes to assist the development of CRF(s) and the underlying database structure
...
- in conformance with the standard
...
- Tabulation Dataset Generation, with attributes to assist the electronic generation of tabulation datasets from compliant data (e.g., via programming)
- CRF Completion Instructions, to assist the creation of applicant documentation for instructing how to complete CRF fields
The columns present in each metadata table are described below with the column name, guidance, and purpose:
| Metadataspec |
|---|
| Num | Metadata Table Column Name | Guidance | Purpose of Column Content | | 1 | Observation Class | Tabulation Dataset Generation |
|
...
- Question Text (full sentence/question forms to prompt for data) OR Prompts (short phrases, often suitable as column headers, to prompt for data)
- CDISC Controlled Terminology lists and subsets of list values, when applicable
- DRAFT CDASH Definition (to assist in understanding the purpose of each variable, which ensures proper usage and simplifies subsequent pooled data analyses)
- CDASHIG Core Designations and Implementation Notes (which, when used together, can assist a designer in determining the complete set of data to be collected on a form)
SDTMIG Programming Metadata
Columns in the CDASHIG metadata tables that will assist in developing programs to generate SDTM domain datasets from CDASHIG-compliant data include:
- Domain
- CDASHIG Variable
- Data Type
- SDTMIG Target
- Mapping Instructions
- Controlled Terminology Codelist Name
- Subset Controlled Terminology/CDASH Codelist Name
- Implementation Notes
Additional Metadata
Clear and consistent completion instructions for sites help to ensure collection of quality, reliable data, a critical factor in the development of high-quality pooled/submission data. The CDASHIG metadata tables include the Case Report Form Completion Instructions column to assist authors in creating this study-level documentation for instructing sites how to complete CRF fields.
The columns in a metadata table are:
...
| Specifies the SDTM class for the domain |
|
...
(see SDTM v2.1 Sections 2.1 and 3.1 for more information on Observation Classes). | | 2 | Domain | Tabulation Dataset Generation | Specifies the 2-letter domain code |
|
...
| | 3 | Data Collection Scenario |
|
...
| CRF and System Design | Specifies when there are different data collection contexts for collection |
|
...
...
...
(e.g., central or local processing for collection of laboratory test results). Values provide context for values in the TIG Core column to determine when a field should be present on the CRF |
|
...
. Values in this column are: - A short description of the data collection context
- N/A for when only 1 context is provided for the domain
|
|
...
Implementation Options
This column can be used to filter the metadata table for specific data collection scenarios when developing CRFs. | 4 | Implementation Options | CRF and System Design | Specifies when a sampling of collection |
|
...
metadata is provided as a template for the metadata of the CRF in a denormalized structure. Values in this column are: - NA to indicate data are collected in a normalized structure using a general variable for the name of the test and 1 variable for the collected results, resulting in a vertical dataset structure in which there is 1 record for each test/result
- Horizontal-generic to indicate data are collected using a unique variable name for each test, resulting in a wide, horizontal dataset with multiple test results in each record
| | 5 | Order Number | CRF and System Design | Provides a suggested order of |
|
...
| fields to be displayed on a CRF |
|
...
...
| Collection Variable | Tabulation Dataset Generation | Provides the |
|
...
data collection variable name When the value in the Observation Class column is Findings, controlled terms must be incorporated into the data collection variable name. | | 7 | Collection Variable Label | CRF and System Design | Provides the data collection variable label | | 8 | DRAFT Collection Definition | CRF and System Design | Provides the draft definition for the data collection variable to support understanding and intended usage. Values in this column are further described in Section 1.5, Known Issues. | | 9 | Question Text | CRF and System Design | Provides full sentence/question forms to prompt for data |
|
...
. Values in this column are |
|
...
: - Without text in brackets or parentheses
|
|
...
- , to be used as-is
- With text in brackets
|
|
...
- , indicating an option on the verb tense of the question or text that can be replaced with protocol-specified verbiage
|
|
...
- or controlled terminology
- With text in parentheses, indicating
|
|
...
- options for text (e.g., singular/plural) or optional text that may be omitted
- With text separated with a forward slash
|
|
...
Prompt
...
- , indicating options for wording from which an implementer may choose.
The CDASH Model can be referenced to create conformant alternative question text. | 10 | Prompt | CRF and System Design | Provides short phrases to prompt for data. Values in this column are phrases: - Without text in brackets or parentheses, to be used as-is
- With text in brackets, to be replaced with protocol-specified verbiage
|
|
...
- or controlled terminology
- With text in parentheses, indicating optional text
- With text with NULL indicating the prompt is not required on a CRF screen/page if not needed
The CDASH Model can be referenced to create conformant alternative prompt text. | | 11 | Data Type | Tabulation Dataset Generation | Specifies the data type of the variable. Values for in this column are: |
|
...
- for character or alphanumeric
|
|
...
...
...
CDASHIG Core
Collection Core | CRF and System Design |
|
...
| Specifies expectations for inclusion of |
|
...
the collection field on the CRF. Values are used with values in the Data Collection Scenario column to determine when a field will be present on the CRF. Values in this column are: |
|
- Req for variables which are Required and must be included in the resulting dataset and cannot be null for any record. Such variables are basic to the identification of a data record or are necessary to make the record meaningful.
- Exp for variables which are Expected to be included in the resulting dataset, even if all values are null. Such variables are considered necessary to make the data record useful in the context of the domain.
- Perm for variables for which it is Permissible to include or exclude the variable from the resulting dataset. Permissible variables must be included in the resulting dataset when data appropriate for the variable have been collected or derived, even if all values are null.
- HR, for fields that are Highly Recommended and should always be included on the CRF
- R/C, for fields that are Recommended/Conditional and should be included on the CRF per certain conditions
- O, for fields that are Optional and should be included on the CRF when needed to support data collection
When the value in this column is R/C, conditions are described in the Implementation Notes column. | | 13 | Case Report Form Completion Instructions | CRF Completion Instructions | Provides recommended example instructions for the |
|
...
data collection field to support authoring of study-level instructions for how to enter collected information on the CRF |
|
...
| | 14 | Tabulation Target | Tabulation Dataset Generation | This column provides the suggested mapping to the tabulation dataset variable. Values in this column are: - The name of the tabulation dataset variable to which the value of the data collection variable will be mapped
- N/A for when no direct mapping to a tabulation dataset variable is available
|
|
...
- Text with the prefix SUPP for when the value of the data collection variable
|
|
...
...
...
| | 15 | Mapping Instructions | Tabulation Dataset Generation | Provides information on the suggested mapping of the data collection variable to the tabulation variable. Additional guidance for mapping values may be described in the Implementation Notes column. | | 16 | Controlled Terminology Codelist Name | CRF and System Design |
|
...
Tabulation Dataset Generation | Specifies the name or names of the CDISC Controlled Terminology codelist or codelists associated with the data collection field and variable. Values in this column are: |
|
In cases where a CDASH/CDASHIG variable has associated CT, the codelist is referenced in the Controlled Terminology column in the CDASH Model and CDASHIG metadata table in this format: (codelist name).
- A single, required controlled term for the variable represented by the controlled term in double quotes. This value indicates the controlled term must be populated in the variable for all records in the resulting dataset.
- Names of controlled terminology codelists with allowable terms to populate the variable. Codelist names are specified in paratheses. Multiple codelist names indicate the variable is subject to one or more of the codelists.
- Names of formats to be applied to values in the variable.
- The codelist name in parentheses, when 1 codelist is associated
- Codelist names in parentheses separated by commas, when more than 1 codelist is associated
- N/A for when controlled terminology is not associated with the data collection field and variable
| | 17 | Subset Controlled Terminology/CDASH Codelist Name | CRF and System Design Tabulation Dataset Generation | Specifies CDISC CT terms or the name of the CDASH subset codelist associated with the data collection field and variable.Values in this column are: - A controlled terminology term or a subset of terms from the codelist referenced in the Controlled Terminology Codelist Name column
- The name of the CDASH subset codelist in parentheses
- N/A for when specific controlled terms or CDASH subset terminology are not applicable
| 18 | Implementation Notes | CRF and System Design Tabulation Dataset Generation | Provides further context for implementation of data collection fields and variables. Values in this column may include: - Rationale for implementation guidance
- Additional instructions for implementing fields and variables on the CRF and in the underlying database structure
- Example values which could appear in the in the data collection field and variable
- Additional guidance for mapping values from data collection to tabulation variables
Example values provided are intended to support understanding and are not intended to influence decisions regarding what data to collect. For guidance on the selection of data to collect, please refer to the appropriate regulatory authority. |
|
...
Implementation Notes
...