Meaningful use of pharmacogenomics in health records: semantics should be made explicit.

The recent emphasis on 'meaningful use' of electronic health records in health information technology reforms (e.g., as in the US stimulus package) can leverage the pharmacogenomics field. In order for clinical trials outcomes, based on pharmacogenomics research, to be meaningfully and effectively used in clinical practice there is a need to make health semantics explicit. Often, semantics is merely implicit in both the research and practice worlds and is buried in unstructured and disconnected descriptions of the data or just in the heads of human experts. Meaningful semantics includes rich metadata, but more importantly, the context of each discrete data item and how it relates to other data items in a specific dataset, as well as how it fits within the entire health record of an individual and how it references up-to-date clinical knowledge. Properly-built electronic health records systems based on standards could provide meaningful semantics on the healthcare side, while the fields of research and clinical trials need to come closer to healthcare in its data and knowledge representations in a way that lends itself to personalized medicine. The purpose of this review is to explore how evidence created by pharmacogenomics can be meaningfully delivered to healthcare through new approaches, such as electronic health records systems and information models.