The Standard

Déjà Vu All Over Again

My last tooling update was titled Focus on Finishing. Thus, in homage to the inimitable Yogi Berra, it would be hypocrisy to change focus now. Focus on Finishing is still the principal theme for the year, building on essentialism, my other guiding light, as expressed in the axiom “Do less, better.”

Toward that end, we continue to move ahead with our transition to our core collaboration tool stack and processes based on workflow-driven online forms. As of this writing, we’re completing final improvements to make the online PSS available to all later this spring. We’ll be working to finish automating most other key form-driven processes after that.

In addition, we hope to finish our transition to a new JIRA-based balloting system, which is also being piloted as of this writing. This, together with the recent transition from GForge Tracker and the STU Feedback web page to JIRA, puts all of our specification feedback in one repository moving forward.

While finishing our transition for balloting is critically important, we also have to update and replace some peripheral systems supporting the balloting process for members, as well as our core business systems for managing membership, events and operations. While we don’t expect to complete this transition to a new Association Management System before the end of 2021, we’ll be focused on finishing this as rapidly as possible since it’s an essential foundation to further systems improvements for the HL7 organization.

Primary author: John D'Amore, Co-Founder, Diameter Health; Co-authors: Brett Marquard, Principal, Wave One Associates and Wayne Kubick, CTO, HL7 International

While HL7 FHIR® (Fast Healthcare Interoperability Resources) is today’s hottest healthcare standard, clinical documents are already exchanged in the billions today. HL7 published the Consolidated Clinical Document Architecture (C-CDA) in 2011 to support care coordination and patient engagement. The ONC 2014 certification rule named C-CDA R1.1 and adoption exploded. The current version of C-CDA, R2.1, remains backwards compatible to that version today. While C-CDA is a flexible, robust standard to record patient care longitudinally or for an encounter, it is structurally complex. The C-CDA standard itself is over 1,000 pages long. Applying the standard to the latest US requirements for clinical data exchange, known as the US Core Data for Interoperability, also routinely requires information from the C-CDA Companion Guide.

HL7 publishes the C-CDA standard and its Companion Guide as PDF documents. That will remain the official version of the standard, but a small team from the CDA Management and Structured Documents Working Groups saw an opportunity to make the C-CDA content easier to use by developing a new web tool. Since the C-CDA has reusable parts, known as “templates,” it seemed logical to make each template searchable and distinct via unique web pages so the standard would be much easier to absorb.

The new HL7 C-CDA Online Search Tool is a searchable, web-based navigation resource for the Consolidated Clinical Document Architecture (C-CDA 2.1) and its Companion Guide. The solution includes over 240 unique web pages that make each template accessible on the web with linkage back to the original PDF content. This navigation is managed through a single searchable index page, like a smart table of contents.

Using the template webpages, health IT vendors can ensure that their C-CDAs are conformant and high-quality for document exchange.  The search tool enables users to search the C-CDA by description, template object identifier (OID) or conformance number. Conformance number searches are particularly valuable for C-CDA implementers, since validators will return these numbers when there’s a violation or warning associated with C-CDA testing. 

Doing Less, but Better

Imagine you’re struggling in a long race – maybe an ultra-marathon over a winding, hilly course. You’ve been running for many hours, and you’re tired, sore and hungry. You’re running up a hill, hoping the end will soon be in sight. But when you get to the top, you only see a turn, not a finish. And after that turn – oh, no! – another hill. We’ve had that feeling during the long pandemic, and, for some of us, we’ve had that feeling even longer with respect to tooling at HL7. We’ve covered a lot of ground, and climbed a lot of hills, and we can feel the end should be in sight very soon. But we’re still running.

Fortunately, we have a team of supporters handing out Gatorade, clapping and cheering us on, and we’ve got our fellow runners pulling us along. And so it is with the HL7 community. We ask a lot of you to help us move forward, with support and understanding; sometimes contributing your valuable time to help us with development or testing, or to struggle patiently with change and the unexpected discoveries of new technology rollouts. While we don’t see that finish line yet, we see plenty of blue skies and greener fields beyond. We won’t always make it on the podium, and sometimes we stumble along the way. But the important thing is to keep moving forward and getting better.

The View from Above

We may not always seem to be progressing very fast, but we’ve really come a long way in the last few years thanks to the important contributions of many of you as well as the ongoing generous support of the US Office of the National Coordinator for Health IT (ONC), which has funded many of our retooling efforts. To list a few prominent examples:

Doing Less, but Better

For many of us, this desperate pandemic year has led to plenty of introspection.  This has also been true for the HL7 Board, which has been contemplating the future of the HL7 organization after emerging from the current crisis. Among a set of core principles adopted by the Board are agility and focus. To be agile, we need to simplify and refine the organization and core processes as well as provide support with continued improvements to our tooling. This also requires getting our global community to better understand and use the processes more consistently and effectively, so we can better focus on our core work of developing and implementing interoperability standards.  This a perfect segue back toward my long-held core belief in essentialism.

Back to Basics

I first espoused the concept of essentialism to an enthusiastic Board and Technical Steering Committee back in 2016. While we’ve only made small incremental progress in the four years since, it has been guiding our process improvement and tooling initiatives. Essentialism was a driving force behind our adoption of Confluence and JIRA as well as efforts to simplify our product portfolio. Of course, we operate in a complex field, and there were many confounding forces acting at the same time. The HL7 community is more adept at introducing new processes, tools and content than at retiring or eliminating the old stuff. Thus, our commitment to essentialism faded over time, tempered by inertia and continuing demands, not the least of which has been the black swan events of 2020.

Perhaps it’s time to once again review the key elements of essentialism and discuss how it fits with our ongoing tooling strategy and plans.

Keeping Us Together While Apart

During this time of global crisis, it’s worth remembering the critical importance of what we at HL7 are doing to improve global health by making health information more available and useful. While so many are struggling with social distancing, we already have a culture that long ago learned to work together remotely on our common goals. But now with fewer opportunities to meet together in person, we need to move ahead to finish much of the work we’ve been doing over the past few years – so we’re better prepared to work together even more effectively, while knowing we must be further apart geographically.

Our ongoing tooling journey at HL7 continues, neither as a sprint nor a marathon.  For us, it’s really more like an odyssey – an ongoing journey where there is always something more to be done, another path to explore, and a final destination (retirement, for example) seems far out of reach.  In the case of HL7 tooling, a fair number of tooling retirements are well overdue.   

Despite the wait, it’s gratifying to see when tangible progress is actually achieved. On the Confluence front, we’re in the home stretch of phase 1 of the rollout, though there’s a whole new course to pursue just around the bend.  We now have all work groups on Confluence (!) and have also migrated many more projects, committees and collaborations. New functions and help features in Confluence (including a major facelift for confluence.hl7.org) are being added regularly, and you can keep up with these by checking the CTO Tooling Update page. This enabling platform is already unleashing many new opportunities within the HL7 community. Our next target is to work toward optimizing our processes with online forms and workflow. The online project scope statement (PSS) pilot is now available and will give us an opportunity to speed up reviews and approvals as well as make new projects more visible to the community in the hope we can avoid last minute catchups. 

Every good story has a beginning, middle and end. First, we get hooked on the opening, which drives us ultimately towards a conclusion, but the real time and effort comes along the way.  While the middle is where most things happen, we can sometimes feel like we’re in a holding pattern there – until something tangible finally happens that directly affects what we do. 

This rings true with HL7’s transition to our new collaboration tooling environment built on Confluence and JIRA. The good thing is that we’re making steady progress on multiple fronts, with many more work groups in Confluence and the killer apps of JIRA Ballot and Unified Terminology Governance (UTG) becoming more palpable. On the other hand, we’re clearly still en route, perhaps able to imagine but not yet actually taste the promised rewards. This is understandable, since the mission of HL7 is the creation of standards, not the creation of tooling to help us achieve that. However, it’s tooling that directly affects us in the ways we develop HL7 standards.