Note: This blog is part of a series highlighting presentations at the 2013 BioClinica User Conference, a free event for BioClinica customers May 14 – 15 in Philadelphia, PA.
Data integration can be a confusing and time consuming process. At this year's BioClinica User Conference, attendees will have the opportunity to participate in a breakout session focusing on this issue. The session will explore:
- Types of data that are critical for integration
- At what point the data should be imported into Express
- Requirement needs from a specification and testing stand point to ensure a successful integration of data
- Subsequent data reconciliation process that is required to ensure all integrated data is done so correctly
- The impact integration has on exporting data from Express
What do we (BioClinica) have to think about when performing mapping for importing data? How can users help us do our job? We need any/all information on what data providers are sending us, i.e. test data, which will show us what the files will look like, how files will be named, the file format, header information, file structure, etc… We need consistency in the transfers, and a data transfer agreement will help ensure that. We also need to work directly with vendors in order to map the key variables.
Data integration is ultimately a collaborative effort between different groups, and is both a technical collaboration and process collaboration. The technical collaboration is in the following of standards in the way the data is formatted across applications, and the way we communicate via web services. The process collaboration represents the human side, integrating business analysts across departments and disciplines. When this collaboration is successful it results in streamlined access to data from various silos, where specialists for each system can handle their own data while viewing data from other sources. Integration still allows specialization of resource pools working within each application.
It's important for users to understand that any importing/integration of data should to be worked on as soon as possible and that it helps to have data managers working directly with data providers. The sooner we begin planning for the importing/integration, the sooner we can get the technical and process aspects defined, and we can ensure that ultimately, the sponsor has access to all of the data they need.
Michael Schmidt is a Senior Clinical Data Manager at BioClinica and will present on the topic "Data Importing/Exporting & Integration Considerations" at the BioClinica User Conference.