June 5th, 2020 NIC webinar was a bulls-eye for professionals whose bailiwick includes such technical aspects of interoperability and information-sharing as ontology, person-matching and privacy/security (among others). And, we’re happy to say yet again, that means we’re making substantive progress on our proof-of-concept initiative, Project Unify. If you’re a geek like us and want to catch up on the fun, check out the recording of the session.
Other highlights of the webinar – which wrapped up a three-part series on Project Unify – included a recap of the successful InterOpathon, held a week earlier by the Michigan Information Health Network (MiHIN); and insights from the Office of the National Coordinator for Health IT into the virtual ONC Patient Identity and Matching Working Group meeting. More information about both events is presented later in this blog.
The NIC webinar, which was appropriately titled “Project Unify (Part 3): Report from the InterOpathon and Next Steps,” included the following:
- MIHIN’s InterOpathon Recap
- Ontology/Vocabulary model proposal
- Person-Matching and ONC event recap
- Privacy and Authorization best practices
- Breakout sessions on Zoom (our first ever!)
For readers who don’t know much about Project Unify, it is a unique proof-of-concept demonstration designed to advance interoperability and information-sharing across Health, Human Services, Education, Child Welfare, Criminal Justice and other domains that impact health and well-being. Unify is a project of the National Interoperability Collaborative (NIC), which itself is an initiative of Stewards of Change Institute (SOCI). On behalf of SOCI and all of us working on Unify, thanks again to the last week’s participants.
Mary Kratz, Executive Vice President of MiHIN’s Interoperability Institute, and Eric Jahn, CTO of Alexandria Consulting, provided an overview of the event. There were five potential “tracks” for competition submissions, CARIN (CMS BlueButton 2.0), DaVinci (Clinical Data Exchange, Prior Authorization), Gravity (Social Determinants of Health), Project Unify (Health-to-Human services integration), and an Open Track. There were nine submissions across these five tracks.
Lessons learned from our participation in the multiple aspects of the InterOpathon include:
- Publish/Subscribe Push notifications would be useful for alerting apps and systems of any changes to Social Determinants of Health (such as chronic homelessness, living situation codes), which would be more useful for supporting real-time workflows.
- Optional syncing of client consent to a Distributed Ledger.
- Improvements to search - household listings in client search results returned by the Golden View need to be easier to get.
- Trusted App registration needs to be accelerated or completed before the event.
- More prep time for developers, so connection/understanding steps are out of the way. Basics of registering, call workflows, code lookups, etc.
- Automate more, improve the UI, and consider no/low code app tracks.
- Identify and/or define standardized ontologies and vocabularies for human services.
NIC’s New Project Proposal: Human Services Ontologies/Vocabularies
One of the lessons learned from the InterOpathon was the need for ontologies, or vocabularies, for Human Services. An ontology is a type of dictionary for a domain that enables everyone exchanging or processing information to have a common definition for given information, its potential values, and its relationships to related information. Eric Jahn proposed a new NIC/Project Unify effort to address this need in Human Services. He proposed that we:
- Catalog worldwide Human Services ontologies/vocabularies using RDF/OWL in coordination with Standards Development Organizations (SDOs).
- Identify gaps in vocabularies and propose new areas of work such as
-FHIR to NIEM (NIEM Health) and HMIS (HUD Homelessness Management Information Systems) to NIEM.
- Propose new APIs and workflows based on collected and gap vocabularies
- Define a Human Services functional equivalent to FHIR/RIM models
Finding the Right Match
Mark Vafiades, Senior Advisor to the ONC, provided a recap of the agency’s June 1 patient-matching discussion. The ensuing conversation at last week’s webinar was a lively, insightful one that included these key points:
- There are lots of intra-system solutions, but no solutions intended for cross-domain matching.
- Many of the ONC Patient Identification and Matching workshop presentations were either “Yay!” Universal Patient Identifier (UPI) cheerleaders (with specific proposals for how to implement UPIs) or “Warning!” NPI/UPI naysayers.
- UPI/NPI does not completely eliminate the need for effective patient identification and matching within the Health domain, especially with regard to Project Unify.
- NYS SHIN-NY patient matching was one of the few presentations that discussed the full range of multi-factor matching issues and Mark LaRow of Verato, Howard Sragow of AllScripts, and David Speights of Appriss each had nicely balanced presentations on referential matching. These approaches to multi-factor identification matching are not healthcare-specific and would work equally well across health and non-health domains.
Data Exchange Privacy Issues
We introduced a mural from SOCI that conveys the breadth of privacy issues involved in cross-domain, cross-organization data sharing. It’s as fun to look at as it is substantial.
Brian D. Handspicker – a co-author of this blog, a key participant in Project Unify and the new CTO for Open City Labs – discussed some of the issues involved, such as:
- Principle of Ownership by Patients/Clients of their wellness data including:
- Protected Identifying Information (PII) used for patient/client matching
- Patient Medical Records, Human Services Case Records, Educational Transcripts, etc.
- Balance of Obligations
- Personal healthcare
- Personal social services
- Public Health
- Research, etc.
- Ensure that access to sensitive and protected personal information can be directly or indirectly controlled by the patient/client to the extent that their circumstances requires.
Tom Silvious – a co-author of this blog, a key participant in Unify and Director of GDIT State and Local Solutions – provided an introduction to the SAMHSA-developed Consent2Share mechanism for real-time patient consent for sharing of medical records. This solution enables an affirmative, informed consent to share different kinds of information consistent with federal regulations like 42 CFR Part 2. This is one mechanism that will be used in Project Unify, but is not the only aspect of privacy that needs to be addressed, especially in Human Services.
The following were identified as issues to address in future meetings:
- Cross-Domain Provenance – where does data come from so that Consent2Share can be requested?
- Cross-Domain Trust – who defines trust relationships between organizations?
- Trust Enforcement – how do we ensure that a domain receiving data abides by either static privacy mark requirements or Consent2Share requirements?
- Data Ownership – beyond the question of consent, who owns the data, especially once in transition across domains with differing legal requirements?
- Governance – what organization has responsibility for defining cross-domain data policies, procedures and rules?
Finally, we experimented with Zoom’s breakout room technology to create smaller groups to discuss a series of questions:
- From your own professional and organizational perspective, describe a few key benefits of demonstrating that data-sharing across domains can be accomplished using standards, common models and tools.
- If Project Unify is successful, what impact would it have on your own work, on your agency/organization/field or on the people you aim to serve (clients, patients, etc.)? Please offer specific examples, if possible.
- What are the key impediments to success?
- Are there potential solutions elsewhere that should be explored?
Join Unify and Tell Us What You Think
If you’re interested in being a part of this exciting initiative, join the Project Unify Group. For more information contact the group's co-leader, Dave Walsh. This collaborative working group is looking for technical and subject-matter experts, developers, documenters and testers, along with anyone who just wants to gain a better understanding of how to provide interoperable, cloud-based solutions.
Please share your comments/questions in the comments section below, and/or by joining in our upcoming webinars.