Making Patient Data Usable — New Guide Points the Way Toward Semantic Interoperability
By Adam R. Davis, MD, and Bill Gregg, MD
For anyone working in health information exchange—which arguably includes everyone in health care and health IT—the why that drives our efforts is patient health. The how is interoperability.
Over the last 15 years, the health care sector laid a solid foundation for the how. Hospitals and providers digitized health records, health information networks are up and running, and the first designated Qualified Health Information Networks under the Trusted Exchange Framework and Common Agreement are not far off. In short, connectivity and the ability to exchange health information are well established.
It’s now time to make real strides toward semantic interoperability—defining standards for structuring health data to make it usable and meaningful for those who need it. To realize the actual benefits of the data, we need to gear up for the next major phase of our journey: Implementing use case-driven data usability guidance for best practice.
Moving from data usability theory into practice won’t be easy. This is complex, incremental, and iterative work, and we’ll need to enlist everyone with a stake in information exchange. Together, we’ll need to map out a plan for how the data usability guidance can be implemented and improved upon, one data element and class at a time. Like topping out a funnel, the pace of progress may feel slow at times.
The great news is that The Sequoia Project—an independent nonprofit dedicated to solving discrete challenges in interoperability—has given us all a head start.
The Power of Usable Data
In the broadest sense, usable data classes and elements are data that help optimize patient care and improve health outcomes and population health.
A single query via a health information network now often returns more data than can efficiently be reviewed—sometimes hundreds of pages. The results can be unorganized and riddled with duplicates, making it impossible to filter or search. The receiving user typically cannot see who entered data or when, and the receiving system cannot surface important clinical information. Narrative notes and clinical context may be missing.
Consistent structure and coding will help resolve these challenges, improving the consistency and meaning of the data, as well as user trust. Data usability is the linchpin to realizing the value of interoperability and building a learning health system.
Improving data usability is also essential for future advances that require data aggregation, analytics, artificial intelligence, and machine learning, which can advance clinical decision-making, population health management, public health analytics, and research.
Where to Start: Getting Consensus on Priorities and Existing Standards
In 2020, as part of its Interoperability Matters initiative, The Sequoia Project kicked off a large public workgroup focused on data usability (which we proudly cochair). Workgroup participants represent a wide range of perspectives—including health IT vendors, standards developers, health information networks, public health agencies, plans and payers, providers, and patients.
Together, the Data Usability Workgroup reviewed 34 major pain points that interfere with the usability of health data. It also reviewed existing standards that are designed to alleviate those pain points. Drawing on input from the 350+ workgroup members, feedback from clinicians, and public comments, the workgroup compiled practical guidance to help health care and health IT stakeholders implement those standards in a more consistent and predictable manner.
The final version of the guidance, an Implementation Guide, was released at The Sequoia Project’s Annual Meeting in December 2022.
No Surprises. No New Standards. Just Actionable and Achievable Recommendations The workgroup’s charter was not to create new standards but to raise the visibility of what already exists. The implementation guide offers actionable and achievable recommendations to begin solving the following six critical challenges:
• data provenance and traceability;
• effective use of codes;
• reducing the impact of duplicates;
• data integrity, format, and trust;
• data tagging and searchability; and
• effective use of narrative.
In each of these areas, greater consistency and specificity will make health data more usable in a profound way.
Designed primarily for health IT implementers, product development teams, software developers, and content testing groups, the Data Usability Implementation Guide evaluates usability from human, machine, and interorganizational perspectives. Implementing the recommendations will help make data easier to receive, parse, sort, filter, index, read, interpret, and use.
An Important First Step
It’s important to acknowledge that this new implementation guide is only a starting point. But in the context of the proverbial journey of a thousand miles, it’s an important first step.
As the health care sector progresses toward a more ubiquitous health information exchange, the Data Usability Workgroup and The Sequoia Project are committed to continuing this vital work on the Data Usability Road Map, to taking direction—and uncovering additional pain points and insights to inform their resolution—from those in the trenches.
No single organization alone can make data usability a reality. It will take a critical mass of the health IT community implementing this interoperability guidance to do the work and make the data exchanged valuable.
It’s a destination we only reach when we navigate and travel together.
— Adam R. Davis, MD, and Bill Gregg, MD, cochair The Sequoia Project’s Data Usability Workgroup. With more than 350 members from across the health IT community, this public group develops pragmatic guidance to facilitate health information exchange.