Beginning in December of last year, all clinical trials are required to adhere to CDISC Standards. The Clinical Data Interchange Standards Consortium (CDISC) is a global not for profit organization with a mission to develop global standards. Through standardized methods, the stages of submission for study data can become more streamlined, efficient, and reliable. Documentation must be submitted in formats supported by the FDA.
In addition to standards allowing easier processing and review, having information readily in a comparable format allows scientists to gather and compare study data from multiple sources. So, let’s take a quick look at those standards.
Clinical Data Acquisition Standards Harmonization (CDASH)
CDASH creates a standard way to collect data across studies and sponsors so that formats and structures provide clear traceability of submission data into the Study Data Tabulation Model (SDTM). This provides transparency to regulators and others who conduct data reviews. CDASH includes things such as guidelines for implementation and regulatory references. They recommend certain data collection fields such as demographics and adverse events.
Study Data Tabulation Model (SDTM)
This extensive model defines a standard for clinical and non-clinical study data tabulations that are to be submitted as part of an application to a regulatory authority. When data is collected from trial participants, it is described by a pre-set classification of variables. Observations are collected in a series of domains that contain related observations. These domains each have a two-letter code that should be used throughout submission.
Though there is an extensive list of domains, most fall into three general classes; Interventions, Events, and Findings. The Interventions domain class includes items such as Concomitant Medications (CM), Exposure (EX), or Substance Use (SU). Observations under the Events domain could include data from Adverse Events (AE) or previous Medical History (MH) among others. The Findings domain contains a vast list including general observation of things like Death Details (DD), ECG Test Results (EG), Laboratory Test Results (LB), and Vital Signs (VS).
As I mentioned the data tabulation model is extensive to say the least. There is a lot of information to look at and as I have just brushed the topic, it is worth exploring further.
Analysis Data Model (ADaM)
The Analysis Data Model ties in very closely to the SDTM. In fact, the SDTM is where ADaM gets its data from. While the SDTM maps the data that is collected, the ADaM preps it for analysis. It outlines how to create datasets and other accompanying data. This allows for the traceability of data by reviewers to be easily traced back to their source. The five key aspects of ADaM is that the datasets must be:
· enable clear and unambiguous communication
· ready to provide traceability for analysis data and its source data (SDTM)
· readily usable by common software tools.
· accompanied by metadata
Standard for Exchange of Nonclinical Data (SEND) for nonclinical data
SEND is an implementation of SDTM as a way for nonclinical data to be presented consistently. Utilizing a common model allows for the easy sharing of information between sources. A SEND package is created to detail what is included in the submission and map out the various domains included. For more detailed information, check out http://www.cdisc.org/send
Define-XML was developed for the metadata that accompany SEND, SDTM, and ADaM datasets. Define-XML transmits metadata for SDTM, SEND and ADaM datasets; it is the metadata file sent with every study in each submission. This informs regulatory authorities on what datasets, variables, controlled terms, and other specified metadata were used.
This barely scratches the surface of the details that the CDISC Standards really contain. However, there is a plethora of information available detailing domains and datasets. What I often hear as the most challenging aspect is how frequently updates are made to things such as domains or controlled terminology Given the lengthy life cycle of pharmaceutical production, a lot can change between when studies are first done and when data is complied for regulatory submission. The best thing you can do is stay on top of it. Know how your data will be submitted during the process and don’t wait until the very end. The earlier you implement new changes the less work needs to be completed to have it in the right format for submission.
Learn more about CSOFT’s dedicated Life Sciences blog: The Word Lab.