From CVRG Wiki

Primary data capture & sharing

Within most clinical trials, clinicians capture baseline and follow-up visit information using paper forms and medical files. Subjects are referred to phlebotomists, cardiologists, radiologists, medical geneticists, etc., for ancillary information meant to provide a full background for each subject. While all of this primary data is tied to clinical trial subjects, the storage mechanisms at each of the clinical locations varies based upon the equipment and services available at the locations. In addition, while the clinical locations can forward their information to a coordination center, the coordination center only has a copy of the primary data, not the original data captured. This can lead to errors in the information at the data coordination center and lead to misinterpretation by subsequent analyses.

Within the CVRG, the basic infrastructure streamlines data capture, using open source software to make the capture itself electronic. This ensures that clinical visit information can be submitted to the data coordination center instantaneously, removing delays that occur from previous submission methods. This also complements a number of the value-added benefits that the CVRG itself provides.

The CVRG software platform also reduces the level of technical expertise required at each clinical site, allowing the clinicians to focus on the subjects, rather than the systems. The CVRG platform can be stored solely at a data coordination center, with access provided to each of the clinical sites. The CVRG platform software can also be distributed among clinical sites, allowing each site to control access to its data. This flexibility allows clinical trials to reduce deployment costs while maintaining access control.

Primary data analysis & derived data sharing

Within most clinical trials, reading centers serve as the resource for interpretation of primary data. While the reading center teams are experts in their field, their coding/annotation of their interpretations are also captured in paper forms. In addition, some reading centers have developed algorithms to analyze the primary data, assisting them in their analysis. While these algorithms may be beneficial to their respective trials, the derived data calculations may be proprietary and understood by a select group of members rather than all the members of the clinical trial.

Within the CVRG, derived data are automatically associated with the primary data and are co-located. This improves the efficiency of communication between the clinical sites and the reading centers, making the analyses available to the clinicians faster.

The CVRG also focuses on making algorithms available for use by the community as a whole. The CVRG maintains an algorithm repository, allowing the community to contribute algorithms and provide detail on them. There is a minimum amount of information that must come in the detail, but this does not mean that the algorithm itself must be disclosed (i.e., intellectual property can be maintained while the algorithm analysis is made available to the community).

Sharing derived data

Within most clinical trials, epidemiologists mine the data, searching for biomarkers to share with the biomedical community regarding specific diseases. While this information can be requested, the data coordination center must aggregate the information requested and provide it to the epidemiologists. In addition, the requests themselves must be approved by the clinical trial's steering committee, adding time to the approval process.

Within the CVRG, access control is maintained by the individuals/groups who submit the data, with control over individual pieces of information serving as the basis for security. Ownership can be transferred (i.e., a technician can transfer ownership to a physician, a reading center coordinator can transfer derived data control to the data coordination center, etc.), allowing for faster sharing of data.

The CVRG also focuses on making data integration easier. With security and data integration combined, the data coordination can provide approved epidemiologists access to certain portions of the primary & derived data. The epidemiologists can then use data integration tools provided by the CVRG to retrieve the data themselves, reducing the burden of the data coordination center while maintaining access control.