Conversion of Legacy Data to CDISC (SDTM)
When a sponsor/applicant submit legacy data to the FDA (Food and Drug Association) it is important that this data is in a standardized format.
Today, there are many sponsors with compounds developed during the past decade, with data not yet standardized according to CDISC (Clinical Data Interchange Standards Consortium), therefore this article is a summary of FDA´s recommendations to approach the conversion of legacy data to CDISC based on SDTM (Study Data Tabulation Model).
FDA’s current thinking on the submission of standardized electronic study data
Standardizing study data makes the data more useful. Data that are standardized are easier to understand, analyze, review, and synthesize in an integrated manner in a single study or multiple studies, thereby enabling more effective regulatory decisions. As a result, FDA has publicly announced that it intends to propose a new Federal regulation that would require the submission of standardized electronic study data. Until this rule is finalized, the FDA (Food and Drug Association) recommends that sponsors and applicants submit study data in a standardized electronic format as establishes in the draft “Guidance for Industry Providing Regulatory Submissions in Electronic Format - Standardized Study Data”1.
In order to standardize study data submissions, applicants and sponsors should refer to the FDA Study Data Standards Resources Web page 2 to obtain an up-to-date inventory of the various data standards supported by CDER ( Center for Drug Evaluation and Research), CBER (Center for Biologics Evaluation and Research), and CDRH ( Center for Devices and Radiological Health ).
CDER collaborates with CDISC in the development of standards to represent study data submitted in support of regulatory applications. For representation of clinical trial, CDISC study data standards include SDTM (Study Data Tabulation Model) which is highly recommended to follow by sponsors and applicants. During these past years CDER has received many “SDTM-like” applications, due to lack of knowledge/experience of SDTM Implementation Guide from sponsors or applicants, which resulted in a big difficulty for CDER reviewers. I order to help sponsor/applicants on the standardization of study data, CDER has created the “CDER Common Data Standards Issues Document”3
Standardization of Previously Collected Nonstandard Data:
It is not always easy to manage a complete standardization of clinical or nonclinical data that were not captured in standard format.
In order to successfully convert to a standard format, each data element originally collected should be mapped to the corresponding data element according to the SDTM Implementation Guide4. Some studies will follow this conversion straightforward and will result in a complete standardized format. But some other studies even following the SDTM Implementation Guide4 may not be possible to map an originally collected data element to the standard data element. If this is the case, the submission should document the reason why data elements could not be fully standardized. If the data was collected on a CRF but was not included in the converted datasets, this element should be identified on the annotated CRF.
If a sponsor decides to convert Nonstandard trial data to SDTM, the applicant should make sure that the resulting SDTM data supports the accompanying analysis data sets and sponsor’s reports (study reports, etc.). CDER has received applications in which the converted SDTM data sets were not consistent with the submitted analysis datasets and study report analyses, thus causing confusion during application review.
Submission of data in SDTM format does not eliminate the need for submission of analysis datasets, whether in the ADaM4 format (preferred) or in an alternative format. Analysis files are critical for the FDA to understand on a per patient basis how the specific analyses contained in the study report have been created.
FDA´s data validation rules.
Data validation is a process that attempts to ensure that data are both compliant and useful. Compliant means the data conform to the applicable data standards. Useful means the ability of the data to support the intended use (e.g., regulatory review and analysis). Data validation is one method used to assess data quality.
FDA generally recognizes two types of validation rules:
- Technical validation rules help ensure that the data are conform to the data standards. For example, a technical validation rule for Clinical Data Interchange Standards Consortium (CDISC) Study Data Tabulation Model (SDTM) data would check that the value in the DOMAIN column of all datasets matches the name of the domain.
- Business validation rules help ensure that the data will support business processes that rely on the data (i.e., support meaningful analysis of the data).
For example, a business validation rule for a human study may require that each value for AGE fall in within a pre-specified human physiologic range. Once a data standard is defined, the technical validation rules are rather static. They are not expected to change substantially unless the standard itself changes. Business validation rules may evolve over time as new analysis requirements are identified and incorporated into data validation processes.
The validation rules used at FDA can be found on the Study Data Standards Resources Web page 2. The data should be validated by sponsors according to the published validation rules before submission and correct any validation errors or explain why specific validation errors could not be corrected. The version of the validation rules used should be specified in the submission. Upon submission, FDA will conduct its own data validation, using the same version of the validation rules used by the sponsor. If necessary, FDA will report serious data validation errors to the submitter for correction.
The Center for Drug Evaluation and Research (CDER) is strongly encouraging sponsors to submit data in standard form as a key part of its efforts to continue with advancement of review efficiency and quality.
SGS Life Science Services is a leading contract service organization providing clinical research, analytical development, biologics characterization, biosafety, and quality control testing. Delivering solutions for bio-pharmaceutical companies, SGS provides clinical trial management (Phase I to IV) services encompassing clinical project management and monitoring, data management, biostatistics, and regulatory consultancy.
For a successful submission including standard data, clients can count on SGS’s large experience on SDTM conversions (with experience converting 160 trials leading to three successful FDA submissions) as well as being a CDISC solution provider and member of the CDISC Organization. SGS’s staff are fully involved and committed to the CDISC organization, including Tineke Callant who is CDISC Authorized Instructor and Joris De Bondt who is member of the CDISC Advisory Council.
Ruth Alonso Meseguer
SGS Life Science Services
i. “Guidance for Industry Providing Regulatory Submissions in Electronic Format - Standardized Study Data” 1can be found at
ii. FDA Study Data Standards Resources Web page2 . See FDA Study Data Standards Resources
iii. “CDER Common Data Standards Issues Document”3. See FDA Electronic Sumissions
iv. SDTM Implementation Guide4. Sponsors should refer to CDISC.org for the latest version of the implementation guides for SDTM and ADaM