October 2001
CDISC Laboratory Data Team Progress
The Clinical Data Interchange Standards Consortium (CDISC), formed in 1997, is an open, multi-disciplinary, non-profit organization committed to the development of worldwide industry standards to support the electronic acquisition, exchange, submission and archiving of clinical trials data and metadata for medical and biopharmaceutical product development.
The CDISC mission is to lead the development of global, vendor-neutral, platform-independent standards to improve data quality and accelerate product development in our industry.
To carry out its mission, CDISC is organized with several different organizational units which guide and support the contributions of the many CDISC participants who perform standards-related activities. One of these groups, the Laboratory Data Team (LAB) is defining requirements for laboratory data, recommending standard approaches to laboratory data processing for clinical trials and testing model performance with representative laboratory data.
The LAB team, formed in August 2000, has sought membership to represent a broad spectrum of stakeholders who deal with laboratory data generated in clinical trials. Currently, team members represent three pharmaceutical companies, one biotechnology company, one CRO, four central laboratories and a group leader. In addition, the LAB team has a representative of HL7 who can provide the perspective of a standards group experienced in handling laboratory data from a clinical (if not clinical trials) perspective. This membership roster incorporates an academic perspective and a European perspective, and also brings to the LAB team the multidisciplinary expertise needed in handling laboratory data generated during clinical trials: statistical, data management, data interchange, information technology, information systems, clinical trials and clinical laboratory expertise. In addition, there is representation from two of the other CDISC data modeling teams: Operational Data Modeling (ODM), working on a model for operational data acquisition, and Analysis Dataset Modeling (ADaM), working on metadata models and sample analysis datasets used for regulatory submission. The goal of the LAB team is to extend the ODM data acquisition model, enabling the model to handle the unique requirements of laboratory data.
The work of the group has been conducted through bi-weekly teleconferences and monthly face-to-face meetings. In the process of doing its work, the LAB team has considered other laboratory data standards that have been proposed in the past. They wish to take advantage of the standardization work that has already been done and to understand the reason that past standards have not been more widely implemented. The CDISC LAB team is striving to develop flexible standards that provide clear advantages to stakeholders in facilitation of data exchange.
Many levels of issues and concern have been considered by the LAB team. Development of truly useful lab data standards will consider global issues such as language, character sets, (e.g., Cyrilic, Turkish, Japanese) and global variations in units and test names. Other issues identified by the LAB team include:
Special need for identification and tracking of lab specimens.
Tracking at several levels (e.g., kit versus specimen).
Requirements for representation of complex test indexing and reporting (e.g., urine microscopic or microbiology testing).
Identification and tracking of collection date and time, receipt date and time, reporting date and time and perhaps, for PK data, tracking of projected versus actual collection parameters.
Specification of a variety of normal ranges (e.g., ranges based on ethnicity, sex, age) or a variety of normal range results (e.g., text only, numerical range, upper limit only, lower limit only).
Transactional changes in patient demographics, test orders, test results.
Descriptors of specimen condition (e.g., hemolyzed, QNS).
Reporting of a variety of flags to describe test results (e.g., telephone, panic).
As a first step in our efforts, the LAB team has worked to define the content requirements of laboratory 'super dataset', including types of data, definitions of each data type, characteristics (e.g., field length) and a logical structure for representing the data. The team is currently working to test the dataset to ensure that our requirements and definitions are comprehensive and work in handling real data (both 'sends' and 'receives'). As a part of our efforts the LAB team is looking to identify potential opportunities for further standardization such as use of standardized code lists (e.g., various HL7 code lists).
After testing, the LAB team plans to release the laboratory requirements document for public comment in order to get broader feedback, from interested parties. Our goal is to incorporate relevant feedback and then identify the data model or models that will best handle laboratory data requirements.
Susan Bassion,
Ph.D.
If you would like to be a member of the Review Committee that will be asked to comment on the laboratory requirements and laboratory data models and standardization, please complete the Participant Questionnaire on the home page of the CDISC website (www.cdisc.org) or contact the CDISC Director of Operations,
Shirley Williams (512-341-9885 or swilliams@cdisc.org).
Indicate your interest in being added to the Review Committee for the LAB team. Your name will be added to the CDISC participant database and you will be contacted when documents are released for public comment. You will also be made aware of CDISC news and updates