The rationale for changing the ‘all or nothing’ approach to USCDI certification
The EHR Association’s commentary illustrates the input ONC must balance as it aims to incrementally move the industry to using standards.
The Office of the National Coordinator for Health Information Technology (ONC) has grown in its understanding, over time, about the breadth of technical capabilities and maturity of organizations that must implement regulations and technical requirements to achieve national goals.
This is not an easy process and ONC has shown a willingness to be open and learn about infrastructure capabilities across the healthcare ecosystem. This is particularly true in the pursuit of national interoperability, where a variety of health information technology (HIT) will be involved in the management and dissemination of all electronic health information.
ONC has demonstrated increased adaptability over time, with a move from “complete EHR” certification to more flexible modular certification.
At the EHR Association, we anticipate new phases of interoperability where the clear goal is facilitating the exchange of all electronic health information (EHI).
ONC must move ahead with a nuanced approach that will ensure HIT is not burdened with requirements that could stifle innovation and essential flexibility.
One example of the challenges that ONC and the industry face is the move to standards supporting all of EHI, and how that works itself out in the nitty gritty of requirements. This topic demonstrates the nuanced way that ONC must approach its regulatory duty as it seeks to move a varied industry forward to achieve long-term goals.
The conundrum of USCDI
Currently, ONC requires health IT developers to certify for the whole of United States Core Data for Interoperability (USCDI). As the USCDI expands, that will create challenges for HIT – in particular EHRs – that would not otherwise need to support all of USCDI.
As the industry considers the expanding USCDI v2, v3, as well as the upcoming USCDI v4, there are examples of data not historically sourced in all EHRs, such as encounter data that specialty EHRs focusing on procedures would not require, or administrative and financial data. Yet certification is dependent on support for all USCDI. This leads to an environment in which developers must overbuild to remain certified, limiting resource availability to create other user-requested health IT functionality and reducing vendors’ capacity to innovate.
The HIMSS Electronic Health Record Association believes a better approach would be to allow each health IT developer to certify against only those standards that support the applicable subset of USCDI that their system actually manages, and which would be useful to their specific client base.
We believe that ONC is not quite aiming in the right direction yet. It is still unclear how USCDI and USCDI+ are aiming to address all of EHI and if it cannot, how ONC anticipates filling the gap for any remaining EHI the standards do not cover. This is also where USCDI can be larger than what a singular HIT would have to support; if USCDI would remain “stuck” at “core” and that “core” is more than what many HIT would actually need to support, then the aim and purpose are out of sync.
Our proposal suggests USCDI be a library of EHI supported by standards and implementation guides for specific use cases that are referenced by focused and modular certification criteria. At the same time, USCDI should continue to expand toward covering all EHI and electronic protected health information (ePHI) as part of setting a roadmap for enabling standards-based interoperability.
Current state challenges
The challenges surrounding the current “all or nothing” state of USCDI certification are twofold.
First, certification is required to full USCDI v1 or USCDI v2 using FHIR US Core and C-CDA, which can necessitate HIT or EHRs offering more capabilities than their intended purpose. For example, creating discharge summaries is not done outside of acute-care settings, while administrative systems typically maintain health insurance information and patient demographics. As other HIT pursues certification for access to EHI using common standards, neither EHRs nor other HIT should need to support USCDI beyond the data they actually manage for users.
These challenges are most evident for specialty EHRs, which must meet all certification criteria, even though only a small percentage of them being applicable to the providers using the systems. For example, laboratory systems deal with test results, not diagnoses and problem lists. Additionally, dermatology practices do not need to create an implantable device list. Yet all of these must be included in specialty EHR functionality to meet current certification requirements.
The result of the current process is that certified HIT developers are increasingly required to provide more functionality than their clients need or are prepared to manage. This adds not only to development costs, but also the timeframe for releasing certified technologies to the market and related implementations. As USCDI grows, HIT capabilities will have to continually expand to maintain certification.
The proposed future state
Instead, the Electronic Health Record Association proposes to reposition USCDI as a library of EHI/ePHI and other data critical for interoperability for which standards have been or are being defined. This sets the stage for health IT to certify against only those standards supporting the applicable subset of USCDI that the health IT in question manages.
We also recommend focusing on the EHI data classes as defined by a collaboration between EHRA, AHIMA and AMIA, which covers all USCDI data classes already in USCDI v1, v2, and v3, as well as other data not yet addressed in USCDI yet part of EHI, as the core target to further advance the necessary standards and implementation guides to support access and exchange of all EHI.
As implementation guides are established using relevant USCDI/USCDI+ data classes and elements for the use cases at hand, they and the standards drive what is necessary and typically include more data than what is defined in USCDI. In the end, the standards and implementation guides are what health IT would use to actually be certified.
Finally, scope certification criteria to well-defined interaction sets within standards and implementation guides that specific EHRs can support for just their role in the use case. Health IT that supports more roles can support more criteria, while not needing to support all criteria.
Addressing EHI/ePHI
These proposed changes would not hinder expansion of USCDI to cover all of EHI/ePHI, setting a roadmap for enabling standards-based interoperability. Interoperability without special effort for EHI – per the 21st Century Cures Act – remains a goal that the Electronic Health Record Association strongly supports.
To achieve this goal, all interoperability involving EHI should be standards-based. For example, it would have been preferable for EHI exports to be standards-based rather than the current approach. If USCDI in combination with USCDI+ is to guide the way to that goal, then EHI must be the ultimate target.
While not all health IT wants to or needs to be certified, wider participation would enable high-fidelity interoperability and predictable, consistent standards. The Electronic Health Record Association therefore suggests that USCDI be used as a roadmap to establish interoperability standards and implementation guidance for all EHI, not just a limited subset of EHI.
The EHR Association suggests establishing standards and implementation guides for relevant subsets of USCDI for specific interoperability use cases and interactions and ensuring certification criteria are sufficiently granular and flexible to enable any sized health IT system to certify to the interoperability use cases and interactions that make sense based on the subset of EHI their clients actually use.
Benefits across the board
Revising USCDI certification from the current all-or-nothing approach would convey benefits across the industry. By focusing on the scope of criteria that need to be met based on the health IT solution’s capabilities, the certification process could be streamlined and accelerated, likely increasing participation.
More USCDI-certified EHRs and other health IT systems on the market would reduce the effort necessary to interoperate across them. It also would speed adoption of new initiatives by ensuring certification criteria are better aligned to the necessary standards and implementation guides based on specific use cases.
Enabling health IT developers to limit the number of criteria they certify against to only those functions that impact their client base would accelerate development timeframes and reduce costs. It would also foster innovation, enabling a wider API-based health IT ecosystem.
Finally, providers would benefit from the industry’s ability to deliver greater innovation and a broader number of certified systems from which to choose – systems not overbuilt with functionality that goes beyond their scope of needs.
David Bucciferro (Foothold Technology-Radicle Health), is chair, and Hans J. Buitendijk (Oracle Cerner), is an ex-officio member of the EHR Association Executive Committee.