What’s happening with data standards and FDA submissions?  We know something is changing but are we all aware of what this means for us?  Our friend and colleague Dawn Edgerton does, and she’ll be explaining it in the next few blogs, starting now!  Thanks to Dawn for enlightening us!


Will your Study Data Meet FDA’s Technical Requirements for Submissions?

A Mini Introduction to CDISC Data Standards


Clinical studies starting after December 2016 MUST use the data standards as listed in the FDA’s Study Data Standards Data Catalog for submission of NDAs, ANDAs, and BLAs; but actually, the FDA already now strongly prefers submission data compliant with CDISC standards for all submissions.  One big reason for this is having standard predictable names for variables and values allows the FDA reviewer to focus on scientific review and not mundane data issues.
So, what are the components for FDA’s technical requirements? To answer that question we need to define some terms first.

CDASH – Clinical Data Acquisition Standards Harmonization.  This is the CDISC standard for collecting the data.  There is a whole library of CDASH forms available to use available free of charge from CDISC. Most EDC vendors have a way to leverage this.  You actually do not submit the CDASH data to the FDA in this format, but using CDASH CRFs will make the SDTM (next bullet) conversion more streamlined.

SDTM – Study Data Tabulation Model.  This is a tabulation of the clinical data.   As an example, the Adverse Events dataset (or domain in CDISC language) will always be named AE and the   verbatim name of an event will always be AETERM.

SEND – Standard for Exchange of Non-Clinical Data.  This is similar to SDTM but for non-clinical data like from toxicology studies.

ADaM – Analysis Data Model.  This the analysis datasets used to create the tables, listings, and figures.  This data model is facilities easier integration of completed studies for marketing applications due to its vertical format.

Define.xml – This is the metadata (data about the data) in XML format.  It contains all the data definitions in the above described CDISC datasets.  It is hyperlinked such that an FDA reviewer can easily drill down to the dataset and variable of interest.

Reviewer’s Guide – This documents describe any special considerations or directions that would facilitate the FDA’s review of the data.

FDA Data Standards Catalogue – The Study Data Standards Catalog provides a listing of the currently supported data standards with links to reference materials.  It also tells you when FDA support began and when FDA support ends for each version of a standard.

Study Data Technical Conformance Guide – This 43 page document provides the details on how to submit standardized study data using the FDA Data Standards Catalogue. This guide also gives explicit instructions on where each component lives in the eCTD and how they should be organized.

Now to answer the original question.  The table below specifies what is needed for each type of data submitted to the FDA.


If you want to dig into the details yourself, I suggest the following sites.
CDISC Website – www.cdisc.org
FDA Website – for example, www.fda.gov/forindustry/datastandards
PhUSE Website – www.phuse.eu

So, the above table tells you the basic components
How do you know if your submission is compliant?  We’ll cover that topic in the next blog. Stay tuned…

Dawn M. Edgerton, MBA, RAC