A precise and thorough approach to planning is key for success in data management.
The Data Management Plan (DMP) is a critical document in any data management project. It outlines all of the data management work to be done, the timelines and milestones to be achieved, as well as the outputs to be produced. The DMP lets all of the stakeholders know what to expect, how to expect it and when to expect it.
The Society for Clinical Data Management (SCDM)‘s publication, Good Clinical Data Management Practices (GCDMP) (1), provides a complete chapter on Data Management Plans. (The GCDMP is available, even to non-members of the society, at their webpage). It is important to note that while DMPs are not regulated documents, they are in fact so commonly used across the industry they have become auditable, and therefore scrupulously close attention needs to paid to getting them right.
We outline 4 key points to bear in mind when creating or reviewing a Data Management plan.
Use a Standard Template for Consistency
To a great extent, the DMP can, and should be standardized across projects for a consistent approach. When using a centralized biometrics model, where data services( data management, statistics, statistical programming) are conducted by a single provider, the development of such standard documents can represent an efficiency in the study set up, and also reduce the oversight burden for the sponsor. Indeed, for any trial project, a robust Data Management Plan template provides a solid starting point. One of the important challenges facing industry professionals today is the increasing complexity of clinical trials, and as such, great care needs to be taken to ensure the DMP accurately documents what actions will be taken with the trial data. Having a highly experienced data management team working on your project, with a track record of implementing innovative and complex trial designs, therefore, becomes increasingly important in this environment.
Take Note of Data Standards
Data standards play a crucial role in structuring and promoting the long-term value of clinical data, and can improve the efficiency of the data management and data analysis process. Clinical Data Acquisitions Standards Harmonization or (CDASH) is now commonly used as part of the CDISC initiative. Developed with participation from all three ICH regions, CDASH provides recommended data collection fields for 16 domains-> DEMOG, AE etc. Where such standards are being used, they should be referred to and documented in the DMP.
Managing the Review
The GCDMP notes that the DMP should be developed ‘in collaboration with all stakeholders to ensure that all responsible parties understand and will follow the processes and guidelines put forth in the DMP from study initiation to database closeout.’ Once developed, it needs to be reviewed and approved by the relevant stakeholders, and put in place before the work described within it is begun.
Project management and communication now form a core part of the Clinical Data Manager’s role, and one of the important tasks is to educate all the stakeholders about the importance of the DMP, achieve agreement on its contents and obtain sign off. Again, less experienced Clinical Data Managers may find this kind of stakeholder education challenging so it is important to ensure that your CRO team has seasoned professionals guiding the data management aspects of the project.
Finally…
It’s a Living Document
A Data Management Plan is a living document, not a static one- as such it is likely to change over the course of the study. For example, for one clinical trial project our data management team created a data transfer agreement with a laboratory that was providing certain results. As the project got underway, the laboratory ceased to be the provider and was replaced by another laboratory vendor who provided their results in a different manner. This had two implications for the DMP- firstly, a new data transfer agreement needed to be executed with the second lab provider. Then, a plan needed to be made for integration of the data already provided by the original lab provider with the data which would come from the second lab. This plan needed to be agreed, executed and documented within the DMP. This is one example which illustrates why and how a DMP needs to evolve, and the importance of documenting any amendments to the plan moving forward.
Cytel’s experienced data management team works with global biopharma and medical device clients to perform efficient database builds and relevant data processing, successfully implement powerful clinical trial software packages, data standards(CDASH) in case report forms and reporting(SDTM), and design operational support functions allowing end users to accomplish their tasks in a more streamlined manner. For more information about our Clinical Data Management services click here.
Liked this blog? We compiled our readers' favourite Data Management blogs from 2016 into a handy ebook. Click the button below to access your copy and gain more insights from our data management experts.
References
SCDM Good Clinical Data Management Practices 2013
Further reading on the Cytel blog
6 steps to timely database lock
Ensuring Quality Data no Matter the Phase
