During the course of any clinical trial, there are often data which, while collected electronically, are outside of the scope of the eCRF . These data include central lab results like ECGs, PK/PD data and others. In this blog we’ll take a look at some key considerations in handling electronic data transfers and any subsequent integration with the core EDC database.
First of all, when data are being brought across from a third party, a Data Transfer Agreement (or DTA) needs to be put into place. As a minimum, the DTA should contain sections for:
- Contacts – who the relevant people are regarding this data and transfer
- Method of Transfer – how the data will be transferred from one entity to another
- File Format of Transfer
- Timing of Transfers ( and test transfers)
- Data Format (a detailed specification of what each variable in the transfer will look like, including variable and date formats.)
- Applicable Definitions (where coded lists will be detailed)
- Blinding of Data
- Data Manipulation (what if any manipulation is being done)
- QC steps Cleaning/Query process
The DMP (Data Management Plan) will tend to handle the steps to be taken once the data management group has received the data including blinding and cleaning considerations. The DMP will also address whether the data are to be integrated into the EDC system.
In our experience, data tend to be easier to understand and clean when integrated, since the integration activity initiates the cleaning process.
Once data are integrated, it’s possible to take advantage of automated edit checks and the data can become part of routine listings and cleanings. One benefit of bringing the data into the EDC system itself is that they become available for clinical evaluation. It’s critical though to be mindful that there are parts of the data which should not be available for all to view.
When taking the decision of whether or not to implement an integration, it’s important to understand that external data can change over time. Therefore explicit data handling directives need to be produced and shared with the teams involved. There are also timeline and budget aspects to be considered, and Cytel Clinical Data Management experts are able to provide guidance on the best approach to fit in with the demands of a particular trial.
As with any aspect of the clinical data management process, careful planning and clear communication are critical for success.
With thanks to Patti Arsenault, Director Clinical Data Management at Cytel