Version 3.3 of the Study Data Tabulation Model Implementation Guide for Human Clinical Trials (SDTMIG) is available in PDF or HTML format.
Enhancements, Key Additions and Revisions
SDTMIG v3.3 provides the following enhancements, key additions and revisions to prior versions:
- Revised Disposition (DS) assumptions to facilitate greater clarity
- Introduction of morphology/physiology domains to support specific body systems
- New Functional Tests (FT) domain supporting QRS supplements
- Expanded Disease Response and Clin Classification (RS) domain to include QRS clinical classification content
- New class of Study References (e.g., Non-host Organism Identifiers)
- Many new examples, revisions, and corrections across domains
- New variables and assumptions for Demographics (DM)
- Associations to supplemental implementation guides (e.g., SDTMIG-PGx, SDTMIG-MD)
- Significant revisions to microbiology domains
- Additional new supporting domains (e.g., ML, TM, SM)
- Greater guidance on managing text strings that are more than 200 characters
CDISC posts Public Review comments and resolutions to ensure transparency and show implementers how comments were addressed in the standard development process.
Files
Please Sign in or Sign up to access the file(s) for this standard
HTML Document
Please Sign in or Sign up to access the HTML Document for this standard.
Errata
These corrections have been incorporated into the SDTMIG v3.4.
Section 3, Submitting Data in Standard Format
- Section 3.2.1, Dataset-Level Metadata: In the row for TI, in the value for Structure, "crierion" should be changed to "criterion".
- Section 3.2.1, Dataset-Level Metadata UR metadata row, in the column “Structure”, "per per" should be changed to "per".
Section 4, Assumptions for Domain Models
- PDF version only: The title of section 4.1.7.1 should be changed from "Trial Inclusion/Exclusion Criteria" to "Example of Splitting Questionnaires".
Section 5.2, Demographics
- DM – Specification: The value in the "Controlled Terms, Codelist or Format" column for COUNTRY should be changed from "ISO 3166-1 Alpha-3" to "ISO 3166-1 alpha-3".
Section 5.3, Subject Elements
- SE – Examples, Example 1, se.xpt: In row 10, the value of SEENDTC should be changed from "2006-06-7" to "2006-06-07".
- Example 1, dm.xpt, Variable BIRTHDTC should be BRTHDTC.
- Example 2, dm.xpt, Variable BIRTHDTC should be BRTHDTC.
Section 6.3.10, Urinary System Findings
- UR – Specification: In the table title, "per per" should be changed to "per".
Section 6.3.10.7, Respiratory System Findings
- RE – Examples, Example 1, di.xpt: "SPSEQ" should be changed to "DISEQ".
Section 7.4.1, Trial Inclusion/Exclusion Criteria
- TI – Specification: In the table title, "crierion" should be changed to "criterion".
Section 7.4.2, Trial Summary
- TS – Examples, Example 1: The value for TSVCDREF in row 27 should be changed from "EUDRAC" to "EUDRACT".
Section 8.4.3, Trial Summary
- --SUPP Examples, Example 1, suppae.xpt: The value for QEVAL in Row 2 should be changed from "SPONSOR" to null.
Appendix E, Revision History
- In the "Change" column for Section 6.3.16, Tumor/Lesion Domains, "TU and TI" should be changed to "TU and TR".
Errors that Affect Conformance
| Short Name |
LBSTREFC Core Attribute Should Not Have Been "Expected" in SDTMIG v3.3
|
|---|---|
| Affected Standard |
SDTMIG v3.3
|
| Description of Error |
LBSTREFC should not have been expected in SDTMIG v3.3. No use cases have been identified of laboratory tests which have a reference value rather than a normal range. Furthermore, a --STREFC value would be derived from an --ORREF value and LBORREF is not included in the domain specification for the LB domain. |
| Efforts to Correct Error |
LBSTREFC was removed from the LB domain specification in SDTMIG v3.4. |
| Concerned Published Element | Concerned Published Attribute | Published Attribute Value | Revised Attribute Value |
|---|---|---|---|
|
LBSTREFC
|
Core
|
Expected
|
(LBSTREFC is not in the LB domain specification)
|
| Impact of Issue | Coping Strategy |
|---|---|
|
LBSTREFC is expected, even though it is almost certainly null for tests in the LB domain. |
Omit LBSTREFC in the LB dataset(s). Explain in the reviewer's guide why LBSTREFC was omitted, even though the omission caused validation errors, The explanation would be that LBSTREFC was omitted because it is uniformly null and was erroneously identified as an expected variable in SDTMIG v3.3. |
|
LBSTREFC is expected, even though it is almost certainly null for tests in the LB domain. |
Include LBSTREFC in the LB dataset(s). Explain in the reviewers guide that LBSTREFC was included because it identified as an expected variable in SDTMIG v3.3, even though the core value was erroneous. |
| Short Name |
DSDY should be permissible, not expected
|
|---|---|
| Affected Standard |
SDTMIG v3.3
|
| Description of Error |
In the DS domain specification, the core value of DSDTC is "Perm" but the core value of the corresponding study day variable, DSDY, is "Exp". The study day variable corresponding to a date/time variable that is "Perm" cannot be "Exp". Note that DSDY was not included in the DS doman specification in previous versions of the SDTMIG. |
| Efforts to Correct Error |
The core value for DSDY will be changed to be "Perm" in the next version of the SDTMIG. |
| JIRA Issue |
| Concerned Published Element | Concerned Published Attribute | Published Attribute Value | Revised Attribute Value |
|---|---|---|---|
|
DS.DSDY
|
Core
|
Exp
|
Perm
|
| Impact of Issue | Coping Strategy |
|---|---|
|
If DSDTC was not collected, so that DSDY is always null, failure to include DSDY in the DS dataset may result in errors or warnings. |
Explain any validation errors or warnings in the Clinical Study Data Reviewers Guide (cSDRG). |
| Short Name |
Core Values for MIBLFL, MILOBXFL, MOBLFL, and MOLOBXFL
|
|---|---|
| Affected Standard |
SDTMIG v3.3
|
| Description of Error |
The --LOBXFL (Last Observation before Exposure) variable was introduced as a Findings Domain qualifier with the intent of eventually replacing --BLFL (Baseline Flag). In findings domains in which --BLFL was expected, its core value was to be changed to permissible, while --LOBXFL became expected. This was not carried out as intended for the MI (Microscopic) and MO (Morphology) domains. In the MI domain, MIBLFL is expected but should be permissible and MILOBXFL is permissible but should be expected. In the MO domain, MOBLFL is expected but should be permissible and MOLOBXFL is permissible but should be expected. |
| Efforts to Correct Error |
The core values will be corrected in the next version of the SDTMIG. |
| Concerned Published Element | Concerned Published Attribute | Published Attribute Value | Revised Attribute Value |
|---|---|---|---|
|
MI.MIBLFL
|
Core
|
Exp
|
Perm
|
|
MI.MILOBXFL
|
Core
|
Perm
|
Exp
|
|
MO.MOBLFL
|
Core
|
Exp.
|
Perm
|
|
MO.MOLOBXFL
|
Core
|
Perm
|
Exp
|
| Impact of Issue | Coping Strategy |
|---|---|
|
An MI dataset which does not include MIBLFL variable may result in a validation error or warning. |
If MIBLFL is not included in an MI datasets, explain any resulting validation errors or warnings in the Clinical Study Data Reviewers Guide (cSDRG). |
|
An MI dataset which does not include the MILOBXFL will flag baseline values in a way that is different from other findings domains, creating confusion for reviewers. |
Include MILOBXFL in MI datasets. |
|
An MO dataset which does not include the MOBLFL variable may result in a validation error or warning. |
If MOBLFL is not included in an MO dataset, explain any resulting validation errors or warnings in the Clinical Study Data Reviewers Guide (cSDRG). |
|
An MO dataset which does not include the MOLOBXFL will flag baseline values in a way that is different from other findings domains, creating confusion for reviewers. |
Include MOLOBXFL in MO datasets. |