CDISC, the Clinical Data Interchange Standards Consortium, is a non-profit organisation that develops the data standards used to collect, exchange, submit and analyse clinical trial data. Its best-known standards are CDASH, SDTM and ADaM.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Clinical Data Interchange Standards Consortium
noun · clinical research
A global non-profit that develops open standards for clinical research data, including CDASH (collection), SDTM (tabulation), ADaM (analysis), ODM and Define-XML (metadata and exchange).
Related terms
In short
Explained
Data moves through a clinical trial in stages: a question is asked on an eCRF, an answer is stored, the answers are organised into submission datasets, and then derived into datasets suitable for statistical analysis. CDISC defines a standard for each of these stages so that a vital signs record, for example, has a recognisable structure whether it comes from a small academic trial or a global pivotal one.
The key point for newcomers is that these are different standards for different jobs, not synonyms. Collection forms and submission datasets are not the same thing, and neither is the analysis dataset.
The Clinical Data Acquisition Standards Harmonization guidelines set out recommended questions, field names and wording for common eCRF domains such as demographics, adverse events, concomitant medications and vital signs. Building forms to CDASH means the data you capture already lines up with the next step. See case report form (CRF) and how to design an eCRF.
The Study Data Tabulation Model organises collected observations into standard domains, each a table with defined variables: DM for demographics, AE for adverse events, VS for vital signs, LB for laboratory results, CM for concomitant medications and so on. SDTM is the format in which study data are usually submitted to the FDA. Datasets are delivered as SAS transport (XPT) files together with metadata.
Analysis Data Model datasets are built from SDTM for the statistical analysis, with derived variables, population flags and analysis visits traceable to the source. Biostatisticians write the specifications; the datasets support the tables, listings and figures in the study report.
The Operational Data Model is an XML format to exchange study metadata and data between systems. Define-XML is the machine-readable description of the datasets (variables, labels, codelists, derivations) that accompanies a submission, so a reviewer can understand each column without asking the sponsor.
eCRF forms
CDASH-style questions
Collected data
Edit checks, queries, audit trail
SDTM datasets
DM, AE, VS, LB ... as XPT
Define-XML
Dataset metadata
ADaM datasets
Analysis-ready, built by statisticians
At a glance
| Standard | Stage | What it gives you |
|---|---|---|
| CDASH | Data collection | Recommended eCRF fields and wording |
| SDTM | Tabulation / submission | Standard domains and variables for study data |
| ADaM | Analysis | Analysis-ready datasets with traceability to SDTM |
| SEND | Nonclinical | Standard for animal toxicology study data |
| ODM | Exchange | XML format for study metadata and data transfer |
| Define-XML | Documentation | Machine-readable description of the submission datasets |
| Controlled terminology | All stages | Standard code lists, so values mean the same thing everywhere |
Regulators support specific versions of each standard. Check the current FDA Data Standards Catalog (and equivalent lists for other agencies) before building.
Build a study in the free Capture sandbox, enter sample data and inspect the export. Every eCRF template carries an OID and an SDTM domain.
Why it matters
For US marketing applications, the FD&C Act gives FDA authority to require electronic submissions in a specified format, and the agency lists supported study data standards in its Data Standards Catalog. In that catalog, CDISC standards are the named formats for study data, and secondary sources report that FDA may refuse to file an application whose study data do not conform. Effective dates depend on the type of application and study start, and have been phased in over several years. Read the current guidance rather than relying on a summary.
Not every study is destined for FDA submission. Academic and investigator-initiated trials may never produce a regulatory dataset. Even so, CDASH-style forms and SDTM-style structure make datasets easier to share, pool and re-analyse, and make it simpler to answer reviewers' questions later. The cost of retrofitting is far higher than the cost of choosing the standard at the start, which is why mapping decisions are best made at form-design time.
Biostatisticians and SAS programmers are the people who most feel the difference. If you work with them, see EDC for biostatisticians and SAS programmers. Data cleaning also depends on the structure; see the clinical data cleaning guide and database lock.
In Capture
Every eCRF template in Capture carries an OID and an SDTM domain (for example DM, VS, EG, PE, LB, MH, AE, CM). CDISC SDTM export is built: SDTM datasets download as SAS XPT files with Define-XML. CSV and Excel exports include an automatic data dictionary and codebook, with raw and decoded files, date-range and by-site filtering.
Blank eCRF · no subject data
Visit by form matrix
| Form | SCR | D1 | W2 | W4 | W8 | EOS |
|---|---|---|---|---|---|---|
| Demographics | X | |||||
| Vital signs | X | X | X | X | X | X |
| Laboratory | X | X | X | |||
| ECG | X | X |
3. Was the ECG performed?
☐ Yes☐ No
If No, complete question 4.
CDISC is the Clinical Data Interchange Standards Consortium, a non-profit that develops open data standards for clinical research, covering how data are collected, tabulated, analysed and described for submission.
SDTM tabulates the data as collected into standard domains and is the usual submission format for study data. ADaM datasets are derived from SDTM for statistical analysis and include derived variables and analysis flags.
CDASH is the CDISC guidance for collecting data: recommended questions and field names for common eCRF domains, designed to map cleanly to SDTM.
For many electronic marketing submissions the FDA expects study data in the standards listed in its Data Standards Catalog, which names CDISC standards. Requirements depend on the application type and dates, so check the current FDA guidance.
A machine-readable file that describes the datasets in a submission, including variables, labels, code lists and derivations, so reviewers can interpret them.
Yes. Using CDASH-style forms and an SDTM-ready export makes data easier to share, pool and submit later, even if the study begins as an academic project.
Yes. Capture exports SDTM datasets as SAS XPT files with Define-XML, and every eCRF template carries an OID and an SDTM domain.
Keep exploring
CDISC SDTM export software
SDTM export in Capture.
EDC for biostatisticians
Working with analysis teams.
Case report form (CRF)
The collection stage.
Clinical data cleaning guide
Before lock.
Database lock
Freezing the data.
eCRF templates
Forms mapped to SDTM domains.
Contract research organization
Who often produces SDTM.
Free sandbox with every feature, including SDTM export. No credit card.