CDISC Technical Roadmap

Quick Links to Models:

Operational Data Modeling Team (ODM)

ODM Team Publishes V1.3 Final for Implementation
CDISC's Operational Data Modeling Team (ODM) has released ODM Final Version 1.3 for implementation. This version includes numerous enhancements and new features in support of regulatory data submission and eCRF application support.

Significant changes are the addition of support for new data formats including partial dates, booleans, URLs, hex and base 64 binary, hex and base 64 float data and the introduction of support for typed clinical data transmission. Typed data transmission will allow for more efficient, higher quality clinical data interchange as adopters incorporate this functionality into their ODM enabled applications. Other important changes include the incorporation of some Define.XML extensions back into the core ODM: Description element that can be used at any ODM Metadata level and the Rank attribute for the ODM CodeList element.

A more complete introduction to the new features in ODM 1.3 can be found in section 2.5 of the ODM 1.3 specification.

Please submit comments to CDISC’s Public Discussion Forum.  Click here for the ODM V1.3 Posting Package.

Submissions Data Standards Team (SDS)

The SDTM Implementation Guide, Version 3.1.2
The CDISC Submission Data Standards (SDS) Team has released a draft version of the SDTM Implementation Guide (SDTMIG) for Human Clinical Trials, Version 3.1.2, for public review and comment. Reviewers are asked to submit comments by the close of business on Friday, September 28, 2007.  A final version is expected to be available in early 2008.

Changes in SDTMIG Version 3.1.2 from Version 3.1.1 have been made based upon industry feedback and previously identified needs. These include recommended solutions for signs and symptoms, findings about events or interventions, questions with multiple answers, and questions such as “other, specify. Version 3.1.2 contains new domains such as those for PK and Microbiology and includes more examples for previously published domains.

Along with the SDTMIG V3.1.2, the review package (SDTMIG v3.1.2_SDTM v1.2 Review Package.zip) includes Notes for Reviews and an Excel spreadsheet for submitting comments. The latter highlights some of the most important changes in Version 3.1.2 from the previous version and request specific feedback on several important questions and issues that are still under consideration by the SDS Team.

The SDTM Version 1.2
The SDTM Version 1.2 has been released by the CDISC Submission Data Standards (SDS) Team for public review and comment. This version of the SDTM has undergone only minor changes (corrections and the addition of several new allowable variables) since Version 3.1, and is still the foundation for the SDTM Implementation Guide. Reviewers are asked to submit comments using the spreadsheet in the review package (SDTMIG v3.1.2_SDTM v1.2 Review Package.zip) by the close of business on Friday, September 28, 2007.

To download the SDTM IG V3.1.2 Draft and SDTM version 1.2, click here.

Metadata Submission Guidelines, Appendix to the Study Data Tabulation Model Implementation Guide 3.1.1
The CDISC Submission Data Standards (SDS) Metadata Team has released a draft version of the Metadata Submission Guidelines, Appendix to the Study Data Tabulation Model Implementation Guide 3.1.1 for review. Included with the IG Appendix is a sample electronic submission that contains examples of CRF annotations, metadata associated with the submission domains, SDTM domains, and an example of a define.xml. The Appendix describes the sample submission and the steps that need to be taken to prepare the data for submission.

Please note that this Appendix refers to the SDTMIG version 3.1.1.  The Appendix will be published as final and updated to incorporate the review comments and reflect the features of the SDTMIG 3.1.2 after the comment period for both documents.

Reviewers are asked to submit comments to Sheila Camhi at scamhi@cdisc.org using the spreadsheet in the review package (Metadata IG Review Package.zip) by the close of business on Friday, September 28, 2007.

To download the Metadata Submission Guidelines, Appendix to the SDTM IG V3.1.1, click here.

SDTM V1.1 Final Published for Use
Version 1.1 (Final) of the CDISC Study Data Tabulation Model (SDTM), prepared by the CDISC Submission Data Standards (SDS) Group, is now available for use. Version 1.1 includes seven new variables that have been included for non-clinical and clinical data applications and the new Trial Summary dataset (that is part of the Trial Design Model). This final version includes several minor corrections and clarifications (listed in Section 6.3) to address issues raised during the comment period, but no major changes. Sponsors who wish to contribute comments for consideration for future versions of the SDTM may do so through the CDISC Discussion Board. Click here for SDTM V1.1 Final.

SDS Team Releases SDTM Implementation Guide V3.1.1 Final
Version 3.1.1 (V3.1.1) of the CDISC Study Data Tabulation Model Implementation Guide: Human Clinical Trials has been released by the CDISC Submission Data Standards (SDS) Team. V3.1.1 is based upon Version 1.1 of the SDTM, which was posted in May, 2005. This newest version includes new features, corrections, clarifications, and examples, all of which are described in detail in Section 10.6 of the document. V3.1.1 also includes a new domain model for Protocol Deviations, which is posted separately. Although V3.1.1 represents the most current, approved version of the SDTMIG, the SDS Team will continue to work on providing additional examples to represent more complex data domains such as signs and symptoms, pharmacokinetic concentrations and parameters, and microbiology. In addition, the team plans to continue work on developing additional models and examples for intervention and event data that include both spontaneous and pre-specified treatments and terms. Comments on this released version can be posted through the CDISC Discussion Board at http://www.cdisc.org/discussions/discussions.html. Click here to download the SDTMIG V3.1.1.

Status of Study Design
CDISC work on Study Design began as part of the SDS team, and resulted in the Trial Design datasets first introduced in SDTM V1.0 and SDTMIG 3.0.  Study Design has also been in the scope of the Protocol Representation Group since its inception. The TDM (Trial Design Model) team continues to develop concepts in this area, and coordinates study design work occuring in several CDISC and CDISC-HL7 teams.

In October 2005, a proposal for TDM Part 2 was published for public comment.  It included proposals for several SDTM-style datasets which would be used to represent the schedule of activities (planned assessments, interventions, administrative activities, and encounters).  Consideration of the comments received and further thought about the challenges presented by complex scheduling examples led the team to abandon this approach and to concentrate on creating a UML model for study design. That model was harmonized with BRIDG, and study design concepts were included in BRIDG Release 1.1.

Two xml implementations of study design are currently under development, an extension to ODM, and an HL7 message.  The first builds on the representation of study data collection already in ODM, adding information on the timing and conditions for data collection, as well as study summary and high-level experimental design information.  The HL7 Study Design message builds on existing HL7 messages and methodolgy for describing workflow, while adding representation of experimental design.

Work on study design within the SDTM framework has been incorporated into SDTM 1.2 and SDTM 3.1.2.  The SEND pilot is utilizing a proposed Trial Sets dataset, and the TDM team is working on generalizing the Trial Visits dataset to other kinds of encounters (e.g., telephone contacts, diary entries) and developing datasets to describe planned timepoints.

See SDTM Trial Design datasets.
See BRIDG Study Design diagram.

CDISC SDS PK Working Group Update
In February, the CDISC Submission Data Standards (SDS) Pharmacokinetic (PK) Working Group posted two new domain models for representing PK data for individual subjects, based on the CDISC Study Data Tabulation Model (SDTM). The posting package contained a domain model for drug concentration data used for pharmacokinetic analysis (PC.XPT), a model for pharmacokinetic parameter data (PP.XPT), a discussion of methods to relate data observations in both domains, and examples using sample data. The comment period closed on March 15. Received comments will be addressed, disposition decisions posted, and the models finalized by June 1, 2006. Click here for the draft PK Posting Package.

Drug Accountability (DA) Domain Status
The draft Drug Accountability (DA) Domain was posted for comment in October 2005, and has been finalized. It will be incorporated into the next version of the SDTMIG. Click here for the October 2005 DA domain draft.

Microbiology (MB) Domains Update
The CDISC Submission Data Standards (SDS) Microbiology (MB) Working Group is posting for comment draft versions of two SDTM microbiology domains, one for storage of microbiology specimen findings (mb.xpt) and the other for microbiology susceptibility test results or other organism-related findings (ms.xpt). The posting package includes the domain models, examples using sample data, and a spreadsheet to record comments. These documents have been shared with reviewers in the Division of Anti-Infective Drug Products, the Division of Special Pathogens and Immunologic Drug Products, and the Division of Anti-Viral Drug Products within the Center for Drug Evaluation and Research (CDER) of the Food and Drug Administration.

The MB domain is designed to store microbiology specimen findings for clinical trial subjects, including organisms found.

The MS domain is designed to store any findings related to the organisms found and submitted in MB. This will usually consist of susceptibility testing results, but can also be other organism-related findings such as extent of growth of an organism. Examples of how the SDTM RELREC dataset is used to relate MB and MS are also included.

The CDISC Controlled Terminology Team will be preparing recommendations on controlled terminology for anti-microbial abbreviations and organism names in the future based on external sources such as the CDC (Centers for Disease Control), WHO (World Health Organization) and ATCC (American Type Culture Collection).

The creation of analysis datasets for microbiology analyses is not covered by these SDTM domain models.

Comments on the domain models and examples are being requested. Please submit comments using the attached comments spreadsheet template to jevans@cdisc.org by [September 30, 2006]. Click here to download the draft MB Posting Package.

Analysis Dataset Model Team (ADaM)

ADaM Implementation Guide v1.0 Draft for Public Comment
Analysis Data Model v2.1 Draft for Public Comment
The ADaM Implementation Guide, Version 1.0 (ADaMIG v1.0), and the Analysis Data Model, Version 2.1 (ADaM v2.1), have been released by the CDISC Analysis Data Model (ADaM) Team for public review and comment.

Reviewers are asked to submit comments by the close of business on Friday, September 5, 2008.  Final versions are expected to be released by early 2009.

Submitting Comments
Comments should be submitted in the Excel spreadsheet (Comment Template for ADaMIG v1.0 and ADaM v2.1.xls), included in the Review Package, ADaMIG v1.0 and ADaM v2.1 Review Package.zip.

Please use only this mechanism to provide comments.

About the ADaM Implementation Guide, Version 1.0
The ADaMIG is a new document that is intended to guide the organization, structure, and format of analysis datasets and related metadata.  In addition to the subject-level dataset ADSL, the ADaMIG describes a new multiple-record-per-subject structure:  the ADaM basic data structure.  The majority of analyses, regardless of therapeutic area or type of analysis, can use this standard structure.  Standard metadata and implementation guidance are provided.  The application of the ADaM standard is illustrated in a number of commonly occurring analysis situations.

About the Analysis Data Model, Version 2.1
The ADaM has undergone revisions, most of which were organizational in nature.  Some of the content that was in Version 2.0 of the model has been moved to the new ADaM Implementation Guide, Version 1.0.  Other content has been rearranged, reduced, or clarified.  There have been no changes to the basic principles to be applied to the design of all analysis datasets and the four types of ADaM metadata.

Click here to download ADaM IG v1.0 Draft for Comment and ADaM v2.1 Draft for Comment.

Final Release of Analysis Data Model Version 2.0
The CDISC ADaM Team announces the final release of the “Analysis Data Model Version 2.0”. This model combines the principles outlined in previously released ADaM documents “Statistical Analysis Dataset Model: General Considerations Version 1.0” and “Subject Level Analysis Model Version 0.21”. Significant restructuring of the two previous released models was conducted to produce a single document giving the user both a guide to important considerations for the creation of analysis datasets and a specific example of the core analysis file that should be created if any analysis files are submitted.

The ADaM team is working on an implementation guide that will build on the considerations discussed in the Analysis Data Model Version 2.0.  This implementation guide will outline specific standards and recommendations for the structure and content of analysis data sets and will contain a library of examples of analysis data sets that would serve to support specific statistical methodology used within clinical trials, such as Change from Baseline, Time to Event, Categorical Analysis, and Adverse Event.

Comments on this released version can be posted through the CDISC Discussion Board at http://www.cdisc.org/discussions/discussions.html. Click here to download the ADaM V2.0.

Laboratory Standards Team (LAB)

Lab Model Status
The CDISC Lab Model Team expects to post Lab Model V1.1 for comment in Q2 2006. Existing Lab models are available by clicking here.

Protocol Representation Group

Important Note about Protocol Representation Model Version 1 Draft: The posting package has been revised as of 4 May 2009. Please click here for details. Comments due 15 June 2009.

The CDISC Protocol Representation Group has now posted the Protocol Representation Model Version 1.0 (PRM V1.0) for review and comment.  Please use the comment spreadsheet included in the posting package to provide comments to CDISC by 15 June 2009.

PRM V1.0 is represented in a domain analysis model and includes semantics for Study Design, Clinical Trial Registry, and Eligibility Criteria.

To review the document and register your comments, please click here. 

Case Report Tabulation Data Definition Specification (define.xml)

The CDISC define.xml Team has published the implementation version of the Case Report Tabulation Data Definition Specification (define.xml). Version 1.0 supersedes all prior versions and reflects changes from comments through the Health Level 7 (HL7) Regulated Clinical Research Information Management Committee (RCRIM) in December 2003 (www.hl7.org) and CDISC's website in September 2004 as well as the work done by the define.xml team in conjunction with the CDISC ODM team to add functionality, features, and additional documentation.

Click here for the Case Report Tabulation Data Definition Specification Version 1.0.

Standard for Exchange of Nonclinical Data (SEND)

SEND Implementation Guide V2.3 Released
Version 2.3 of the Standard for Exchange of Nonclinical Data (SEND) Implementation Guide has been released by the CDISC SEND Team. The SEND Model has been prepared by the SEND Team to guide the organization, structure, and format for non-clinical data submitted to the FDA. The focus of the SEND Team is on data collected from animal Toxicology studies. SEND v2.3 covers single-dose, repeat-dose and carcinogenicity studies. Please also refer to the Study Data Tabulation Model (SDTM) which represents the underlying conceptual model behind the SEND standard. Comments on any of these documents can be submitted through CDISC's public Discussion Forum at any time. Click here for SEND V2.3.

The FDA is not currently prepared to accept SEND datasets. Once they are, there will be a series of validations performed to ensure the datasets comply. The FDA has been working with PharmQuest PointCross under a CRADA to develop the validation checks as part of the ToxVision tool.

A Federal Register notice announcing a regulatory pilot project for the submission of SEND datasets is expected within the next month. Participants will be submitting SEND data along with the traditional PDF files as part of active INDs. The specifications for the pilot datasets will be based upon SEND v2.3, with some modifications for greater consistency with the SDTM v.1.1 and SDTM Implementation Guide (for Human Clinical Trials) v 3.1.1. Feedback from the pilot will also be taken into account in developing the SEND version that will ultimately be the basis for guidance and/or regulation. Because of pilot dependencies, it is difficult to predict when a final implementation-ready version of SEND will be published, but the earliest is estimated to be mid-2008.

Terminology Team

The CDISC Terminology Team has released a new terminology set for public review known as SDTM & Labtest Package 3. This set includes terms for Drug Accountability, Evaluator, Reference Range Indicator, Relationship Type, Specimen Condition, Specimen Type as well as new terms for Laboratory Test Results (LBTEST). Also, included in this set are terms for the CDASH-specific fields MHONG (Ongoing/Resolved) and SUNCF (Prompt for Substance Use). Click on the following web-links to access the terminology spreadsheet and public comment form.

SDTM & Labtest Package 3 Terminology Spreadsheet (September 2008)

Public Comment Form

The CDISC Terminology Team has finalized the production release of SDTM Package 2B and Labtest Package 2. This terminology set has been combined with prior production releases of SDTM Package 1, SDTM Package 2A and Labtest Package 1, bringing the total number of SDTM production terms to nearly 2300. SDTM terminology is maintained and distributed as part of NCI Thesaurus and is available for direct download at the following web-links.

NCI EVS Terminology Resources (access to CDISC, FDA and other terminology)

SDTM Production Terminology spreadsheet (September 2008)

Clinical Data Acquisition Standards Harmonization (CDASH)

The Clinical Data Acquisition Standards Harmonization (CDASH) project goal is to develop a set of ‘content standards’ (element name, definition, and related metadata) for a basic set of global data collection fields (also known as CRF, or Case Report Form, variables) that will support clinical research studies.  The initial scope of the project is focused on the ‘safety data domains’ (i.e. Adverse Events, Prior and Concomitant Medication, Demographics and Subject Characteristics, Medical History, etc.).

CDASH is a CDISC-led project, initiated by ACRO (Association of Clinical Research Organizations) and first announced by Janet Woodcock, FDA, at the 2006 DIA Annual meeting.  CDASH will focus on FDA Critical Path Opportunity #45, and is currently supported by a collaborative group of sixteen organizations: ACRO, ACRP, AMIA, Baylor College of Medicine, CDISC, Clinical Research Forum, FDA, NCI, NCRR, NIH, NLM, Critical Path Institute, PhRMA, BIO, SCDM and Duke Clinical Research Institute.

For more information on CDASH, click here.

Cardiovascular and Tuberculosis Data Standards

Package #1 for the Cardiovascular and Tuberculosis therapeutic specific data elements are now posted for public review.

The purpose of these projects is to develop clinical content data standards supporting both patient care and secondary data uses which include disease surveillance, quality improvement and clinical research. This work is funded through the NIH Re-Engineering the Clinical Research Enterprise Roadmap contracts. The resulting data elements and products were developed by a team from Industry; Standards Development Organizations, including CDISC and HL7; and Clinician and Government Stakeholders over the last 2.5 years.

The work focused on two separate clinical domains: Cardiovascular and Tuberculosis. Because these domains are large, they have been further broken down into packages. For the Cardiovascular domain, the Package 1 scope is Acute Coronary Syndrome and for Tuberculosis (TB), the focus is on Pulmonary TB.

The data elements with their definitions and permissible value sets were developed from Patient Use Case Scenarios and Activity Flow diagrams. These data elements were incorporated into class models, using Unified Modeling Language (UML), to illustrate the relationship of the data elements to each other. An example of how the data elements are produced in a form format is also included and the Tuberculosis data elements have been modeled in the SDTM format.

There are supporting documents to provide guidance on how to navigate through the products. There is an Index Document that provides a brief description of each document in the packet, and indicates which documents are for review and which are supporting documents. Hyperlinks are provided to help you navigate through each document.

The Project Document provides an overview of the project and a detailed explanation of the products in the packet. The Understanding Class Model Document gives an overview of the model and how to read it. The TB Implementation in SDTM provides guidance of how TB healthcare data standards could be implemented in clinical research using the CDISC Study Data Tabulation Model (SDTM) to aggregate data across research studies.

There is a comment form for each project to collect feedback and to indicate what materials you reviewed but did not have comments.

Through this work, we hope to provide and facilitate a path for others to follow as clinical content is defined with the semantic specificity necessary for interoperability.

Please review any of the products within the packages that fit your area of expertise.

For more details on these projects and to download the review packages, click here. Please use the comment spreadsheet included with the download to submit comments to jevans@cdisc.org by 15 April 2008.