A major challenge to adoption of personalized medicine approaches, such as pharmacogenomics, is that data to address many critical questions relevant to clinical application, economics, policy and regulation are typically unavailable or widely dispersed. The evidence base we tentatively call Evaluation Data for Assessing Personalized Medicine Translation (EDAPT) would generate the foundation for a database of information. This article begins with a description of the rationale and conceptual framework for this effort.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.2217/17410541.3.4.411 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!