1. What is CDISC
The full name of CDISC is Clinical Data
Interchange Standards Consortium, which is the Clinical Data Interchange
Standards Association. It is a non-profit organization. Committed to providing
data standards for the development of medical and biopharmaceutical products.
What we usually call CDISC refers to the
data standard developed by this association, that is, the CDISC standard. These
standards are used for the acquisition, exchange, submission and archiving of
clinical and non-clinical data.
2. Why adopt the CDISC standard?
The CDISC official website lists the
following benefits of implementing CIDSC standards:
•Improve efficiency
•Complete traceability
•Enhance innovation
•Improve data quality
•Promote data sharing
•Reduce costs
•Improve predictability
•Simplify the process
Let’s take a practical application
scenario.
With the popularization of artificial
intelligence technology, we have found that the threshold for technology
application is gradually lowering, but the large enough data required to apply
technology is difficult to solve in the short term.
First of all, a large amount of medical
data exists in different systems and research. Due to the adoption of
non-standard or standards in a certain field, these data are like information
islands, which are difficult to connect and share. Secondly, within these
information islands, due to the lack of unified "construction"
standards, if you want to reuse it, you must know the internal
"construction" details. If you want to reuse thousands of research
data, the workload will undoubtedly be very large. huge.
The two problems faced above are exactly
what the CDISC standard can solve.
First, develop unified data standards for
the construction of data structures. Whether for humans or machines, the
readability and reusability of data are greatly improved. Second, develop data
exchange standards for different systems. Data exchange improves data
integrity.
3. Composition of CDISC standards
The CDISC standard is constantly evolving.
According to the current structure of the CDISC official website, the CDISC
standard can be summarized into the following five parts:
CDISC family portrait
1. Foundational Standards
The basic standard is the core of the
entire CDISC standard. Defines the model, domain and specification of data
representation (Data Rrepresentation).
CDISC standards in the clinical research
process
The basic standard consists of the
following parts:
•SEND (Standard for
Exchange of Nonclinical Data)
It is the implementation standard for SDTM
for non-clinical studies. SEND specifies methods for collecting and presenting
non-clinical data in a consistent format.
•Protocol Representation Model PRM
(Protocol Representation Model)
Provide standards for the design of
research protocols.
•Clinical Data Acquisition Standards
Harmonization (CDASH)
Content standards for data collection fields
in case report forms. For example, the theme EDC system of our series should
build a database in accordance with the standards of CDASH and SDTM.
•Study Data Tabulation Model SDTM (Study
Data Tabulation Model)
The so-called tabulation, in layman's terms,
is a standard for data organization and formatting. The purpose of CDISC
developing SDTM is to submit unified standard data to regulatory agencies and
improve review efficiency.
•Analysis Data Model ADaM (Analysis Data
Model)
Establishes standards to be followed when
creating analysis data sets.
•BRIDG
BRIDG is a domain analysis model. The main
role of BRIDG is to exchange information between medical information systems
(such as HIS, LIS, and PACS) that comply with HL7 standards and clinical research
systems that comply with CDISC standards. BRIDG is simultaneously a CDISC, HL7
and ISO standard.
2. Data Exchange Standards
As the name suggests, the role of data
exchange standards is to exchange, share, submit and archive structured data
between different systems. Data exchange standards are vendor-neutral and
platform-independent. The XML mark means that the standard file format is XML
(Extensible Markup Language).
Data exchange standards consist of the
following standards:
•ODM-XML
XML-based Schema defines content and format
standards for obtaining, exchanging, submitting and archiving clinical research
data in accordance with regulatory requirements. A common infrastructure is
provided for Define-XML, Dataset-XML, SDM-XML and CTR-XML, and they all extend
ODM-XML.
•Define-XML
ODM-based metadata standard used to
describe metadata for SEND, SDTM and ADaM in order to inform regulatory
authorities which data sets, variables, controlled terms and other specific
metadata are used.
•Dataset-XML
Dataset-XML is used to describe SEND, SDTM
and ADaM data sets. CDISC developed Dataset-XML with the purpose of replacing
SAS V5 XPORT. Why replace it? Because it is a format standard unique to
commercial companies. Dataset-XML makes some improvements over SAS V5 XPORT,
removing many SAS V5 transfer file limitations (the current file format
required by the FDA and PMDA), such as 8-character variable names and
200-character text field lengths. Dataset-XML and Define-XML are complementary
standards, and Define-XML metadata describes the contents of the Dataset-XML
data set.
•CTR-XML (Clinical Trial Registry XML)
Promotes standards for the submission of
clinical trial information to global clinical trial registries. Examples
include the World Health Organization (WHO), the European Medicines Agency
(EMA) and ClinicalTrials.gov.
•SDM-XML（Study/Trial
Design Model XML）
The Study/Trial Design Model is an
extension of ODM that defines a machine-readable, interchangeable standard for
describing clinical study design.
•LAB
The LAB standard is specifically designed
for the exchange of laboratory data obtained in clinical trials. Provides a
standard model for the acquisition and exchange of laboratory data.
•RDF (Resource Description Framework)
The full name of RDF is Resource Description
Framework. It is a set of markup language technical specifications proposed by
W3C, which is used to describe and express the content and structure of network
resources. CDISC's RDF defines the basic standards of CDISC and is expressed in
the Resource Description Framework (RDF).
3. Therapeutic Areas
It is an extension of the basic standard
and develops data standards related to specific disease areas. At present,
standards for more than 30 specific disease areas including Alzheimer's
disease, Ebola, Parkinson's disease, asthma, etc. have been released.
4. Terminology
In brief, the Glossary explains the
definitions and terminology related to the collection, exchange, and submission
of data during non-clinical and clinical research phases. Contains all terms
involved in the CDISC standard.
5. CDISC SHARE
The full name of SHARE is Shared Health And
Research Electronic library, which is a set of tools and services. When we
create databases and applications for CIDSC standards, classes, fields, and
variables are manually extracted from PDFs, and SHARE facilitates the
automation of standards by providing variables and other CDISC standard
metadata in a machine-readable format. management and implementation.
4. Regulatory agencies and CDISC standards
Currently, FDA submissions require the
following CDISC standards:
•Controlled Terminology
•SEND
•SDTM
•ADaM
•Define-XML
Documents submitted by PMDA require the
following CDISC standards:
•SDTM
•ADaM
•Define-XML
•Analysis Results Metadata (ARM for
Define-XML)
Currently, SFDA does not require submitted
documents to comply with CDISC standards.
“In order to improve the quality of
clinical trial data and the quality and efficiency of statistical analysis, and
facilitate the exchange and summary analysis of data, when applying for new
drug marketing registration, it is recommended to submit the original database
and analysis database using CDISC standards.” - "Clinical Trial Data
Management Work" Technical Guide
References
1. CDISC official website.
https://www.cdisc.org
2.CDISC WIKI.
https://en.wikipedia.org/wiki/Clinical_Data_Interchange_Standards_Consortium
3.Sam Hume.CDISC Standards Beginning to End
(B2E),21-Feb-2017
4.Wayne R. Kubick.CDISC SHARE: A
BRIDG-based Metadata Repository Environment
5.Giri Balasubramanian,Edwin Ponraj
Thangarajan.CDISC Transport Standards - A Glance.Digital Innovation in
Healthcare,8th-11th Oct 2017.
6. Wu Chongsheng, Bao Wenjun, Wang Jun,
Peng Ruiling, Deng Yazhong, Zhang Zibao. Introduction to CDISC standards and
their application in China [J]. Acta Pharmaceutical Sinica, 2015, 50(11):
1428-1433.
7.Samuel Hume, Anthony Chow, Julie Evans,
Frederik Malfait, Julie Chason, J. Darcy Wold, Wayne Kubick, Lauren B.
Becnel.CDISC SHARE, a Global, Cloud-based Resource of Machine-Readable CDISC
Standards for Clinical and Translational Research.
8. Wen Tiancai, He Liyun, Liu Baoyan. BRIDG
model: a bridge for the exchange of medical information and clinical research
information. China Digital Medicine, 2010, 5(8):0-0.
评论
发表评论