USCDI and Access of Health Data
On Wednesday, the HITAC met for their monthly meeting, and their US Core Data for Interoperability (USCDI) task force shared their recommendations on improving the ONC's USCDI, which ONC introduced earlier this year along with their TEFCA work.
You can find their material here, but I really felt their recommendations were solid and well-made. ONC has introduced the idea of USCDI and a process for its growth, but they only provided a light framework around it for determining how to identify emerging and candidate data classes. The taskforce provided some good suggestions for improving this framework and process, including expanding the stages for clinical data adoption. Their suggested stages are:
Stage 1: Proposed (New) Stage 2: In Preparation (New) Stage 3: Emerging Stage 4: Candidate Stage 5: USCDI Stage 6: Widespread Deployment (New)
I think they are wise to add some stages before Emerging as there is significant work before you can even consider a data element is even at the level of "emerging". I can see this connecting back to the Interoperability Standards Advisory (ISA) analysis of data classes and associating these stages with the standards process and implementation maturity levels already identified in the ISA.
They also introduced the idea of a hierarchy of data structure with top level data classes consisting of multiple data objects which themselves have specific data object attributes. For example, data attribute of Income is part of the data object Economic Stability which is grouped in the class of Social Determinants of Health. You need to think of data this way if you are going to properly identify, capture, share, and use it in a way that is ultimately meaningful to patients and providers and other stakeholders.
In listening to their presentation, I thought of the recent CMS MyHealthEData initiative allowing greater access to claims data. The great response to that effort shows that we need to enable all parts of a patient’s data involving their health care to be shared with the proper security and consent, and not just pure clinical data elements captured in the EHRs.
Also, I felt their summation of what prevents data from being shared (copied below) was a great quick analysis of the challenge of sharing data and better explains the nuances of “information blocking” than many things I have previously seen . Their points dealt with the data elements and did not really touch the exchange aspect via API web service channel or Direct or other methods. However, you can more or less switch “Transmission Exchange” with “Data” in these points and correctly address the challenge of information blocking in exchanges mechanisms too.
Data doesn’t exist.
Data exists but is not collected at all or in part.
Data is collected but there are no semantic standards for normalizing it.
Data is collected and there are appropriate semantic standards, but they are not being broadly applied.
Data is collected and semantic standards are applied; however, inconsistent application of semantic and other standards (eg., use of local or custom codes) by organizations inhibits interoperability.
Detailed and reliable workflows to share the data outside of the originating organization have not been established.
Of course, there are certainly other factors which are part of the information blocking problem. I would probably add a seventh point of workflow is established but clinicans and providers do not believe sufficient value in exchanging data due to cost or other factors. Health care organizations have to see the ROI in sharing data or they are not going to invest in the resources to make it a reality. Finally, an eighth point of health IT developers preventing data sharing due to market-based and financial concerns. This does exist although I feel it is overstated in terms of presenting EHR developers are as the main culprit behind lack of interoperability and data sharing. I personally believe the main bottleneck with information blocking is somewhere in the 3-7 stages, mostly in 6 and 7.
Overall, some really solid work by the task force and shows the value in having a Health IT Advisory Committee and health care stakeholders volunteers associated with the ONC work. I hope ONC takes it into consider as they look to finalize the USCDI later this year.