What do we really need to make the best use of shared patient data? Some say that once we have adequate data sharing protocols in place (such as FHIR or Direct), organizing and using the data will be well within our capabilities. Other efforts assume that if we pulled together the right common data set, deciding how to exchange the data physically won’t be as big of an issue as it has been.
A new initiative from the American Medical Association seems to fall into the latter category The AMA has announced that it’s created a common data model which it says is missing in healthcare. The Integrated Health Model Initiative (IHMI), which has attracted the support of heavy hitters like IBM and Cerner, is a “shared framework for organizing health data, emphasizing patient-centric information, and refining data elements to those most predictive of achieving better outcomes,” according to an AMA statement.
The AMA and its partners said that the new model will include clinically-validated data elements which it says can speed up the development of improved data organization, management and analytics. Its initial focus will be on costly chronic diseases such as hypertension, diabetes and asthma.
The effort will include technical development efforts which will address interoperability problems, cumbersome or inadequate data structures and poor interface designs which forced physicians to click far too often, the trade group said.
From my standpoint, there’s a lot that’s hazy about this announcement, which was long on form but pretty short on substance.
For one thing, it’s not clear what Cerner, in particular, is getting out of this effort. It’s already an anchor member of the CommonWell Health Alliance which, having merged with rival group Carequality, arguably offers as mature an interoperability model as any out there today. Also, while even a giant like IBM needs continued press attention, I’m not sure how much benefit it will realize here.
Not only that, it’s hard to tell where the AMA and partners will take IHMI. The trade group has posted a set of data model specifications to its site. The group has also created a process wherein physicians review data elements and missions and decide whether they meet clinical applicability and consistency requirements. In addition, it’s creating technical and clinical communities focused on key sub-areas of interest. But it’s still not clear what all of this means and why it’s important.
Ultimately, the initial press release is as much a buzzword cloud as it is a statement of intent. Pardon my cynicism, but I doubt even a group with the AMA’s clout can fix interoperability problems, streamline data structures and foster more elegant UI design in health IT in one fell swoop.
The announcement does do something useful regardless, however. While I’m not personally qualified to say whether it will take universally accepted standards for data exchange, a widely-used reference set for health data or both, I believe someone should address these questions. As proposed interoperability solutions pop up on both sides, perhaps we’ll get some answers.