Location: Montreal, Canada / Date: Monday 9 May 2016
Time Q 2
Facilitator / Patient Care / Note taker(s) / Mitra Rocca
Attendee / Name / Affiliation
Quorum Requirements Met:
Anita provided an overview to the CIC agenda for the week.
James Topping (Duke) provided an overview to the MAX tools. Max tools addresses the limitations of Enterprise Architect (EA) and back and forth with the EA model.
Example: Max tools focuses on the import and export and moving data in and out of the EA model.
James discussed the use case scenarios. MAX tools will help with moving data in and out.
Two main components of MAX tools are:
· EA Extension
· Excel XML Schema
XSLT is used to transform the exported EA XML.
· MAXSchema_Excel
· Source
Uses to set the MAXSchema Excel as the XML map
Steps to success:
o Define and prioritize MAX requirements
o Build MAX prototype
o Pilot MAX on a DAM project
o Refine MAX based upon pilot feedback
o Prepare MAX user guide and education
o Post MAX to HL7 tooling page
With EA, you can export in XMI, bring all separate instances is difficult to do in XMI. MAX tools will help with this.
Update from Child Health WG (Pele Yu)
· Update on projects
Child health will meet on Tuesday Q1, Q3 and Q4
· Have worked with Electronic Health Record (EHR) Functional Model (FM) Release 1 R1 and HL7 Quality Reporting Document Architecture (QRDA). This workgroup developed a Child Health Functional Profile.
· The workgroup is currently working on a derived profile for the EHR model that is focusing on developmental screening for kids and toddlers (0 – 3 years of age). The developmental Screening Functional Profile (FP) is based on the Electronic Health Record System Functional Model Release 2 (EHR-S FM R2). The child health WG is working on finalizing the work and hopes to ballot this functional profile in the HL7 Workgroup Meeting (WGM) in September 2016. The team is working with Enterprise Architect (EA).
· Q: During the developmental screening mapping exercise, the Child Health WG is working with EHR-S FM R2 and one Conformance Criteria (CC) maps to multiple CC in EHR-S FM R2. For example a CC for developmental screening, the EHR-S should provide an alert to the healthcare provider (HCP) with age, sex, etc. which maps to Clinical Decision Support (CDS), and other sections of the EHR-S FM. Do we map to all of the EHR-S FM R2 or only to the best function? In EA, you need to decide how to link it to the EHR-S FM. Child health is the first WG working on a derived profile. What questions should the system ask at a certain age group? We need to map granular to higher level Functions in EHR-S FM.
o A: No good answer.
Update from Clinical Quality Initiative (CQI)
· Floyd provided an update to the Health IT enabled Quality measurement and improvement Ecosystem. CQI is looking at quality measurement and trying to improve care. We look at research, guidelines, CDS, clinical care, measurements and reporting.
· Types of data Re-Use can be research, Public Health and population health management, quality improvement.
· CQI has Health Quality Measures Format (HQMF), QRDA, Quality Data Model (QDM) and other standards. We have a common data model to define the elements and a common way to use logics.
Update from EHR Workgroup
Gary Dickinson provided an update on the Functional profiles based on R1 or R2.
HL7 CIC – New DAM Project THEMES (Exercise Standard)Location: Montreal, Canada / Date: Monday 9 May 2016
Time Q 3
Facilitator / Anita Walden / Note taker(s) / Mitra Rocca
Attendee / Name / Affiliation
Quorum Requirements Met:
THEMES
· University of California Irvine (UCI) (Dan Cooper is the PI), brought a team together to develop a set of exercise measures (CPET (Cardio, Pulmonary Exercise testing)) to leverage it for healthcare.
· There was a face to face meeting in March at UCI.
· Q: where are we in this project? And the DAM?
o A: we are at the early phase. There have been 3 meetings for this project.
· Q: Is there funding secured for this project?
o A: Identify the data elements and scope for this project and request for additional funding for the January 2017 ballot cycle.
o The team understands what a data element is and the project lead is taking the list of data elements back to physicians.
o Q: what is happening with this project at CIC?
A: CIC is providing them with a data modeler and guiding principles. We will develop a DAM including their data elements. We will include the data elements in CIMI. We will ballot it in January 2017.
o Pele will ask pediatric pulmonologists to see if any data elements are missing.
o The scope is still not identified. Select a protocol to use (bicycle to use or another device), conduct the exercise, and interpret the results.
· Q: what is the group that is providing the domain expertise?
o A: This is not managed by a professional society. There are 12 clinicians participating in this project.
· Q: When we get the data elements and definitions of them, is there a group who can validate them?
o A: Yes.
· Q: is there an association within sport medicine or pulmonologists?
o A: there are pulmonologists but not sure about sport medicine specialist.
o We need to identify the use cases and build them in the EHRs. Most of the focus is around exercise bike and treadmill.
o We would like them to use mobile Health devices.
o The DAM will extract the concepts, to talk about the device, vs. treadmill or bike. This will enable enhancing the model to include any device (Fitbit, digital scale,).
o We will have a draft of a DAM by this summer and will bring it to the September HL7 WGM.
o The funding is currently a problem.
o Pele does not see these currently in his patients.
o Sport physical or athlete, or if there are injuries to the sport medicine doctor and get a report. It can be part of a Consolidated Clinical Document Architecture (C-CDA).
· Q: Is there anything HL7 can do, FHIR resource, CDA?
o 2 bodies of stakeholders
o Domain experts who provide input
o DAM experts
o Ask a Clinical and Translational Science Award (CTSA) body to look at this.
Action Item 1: Anita will ask Dr. Cooper, if they are interested in presenting to the CTSA Informatics task force.
FHIR Project Scope Statement (PSS), Lenel James
Since April 25th, there has been an effort to develop a PSS a FHIR extension and profiles. How can these be taken by other groups? For CDA there is a process. In CDA we did not have work outside. In FHIR there are a lot of projects going on.
The ask is, that we believe, that is uniquely a US REALM challenge. The known documented need for a FHIR repository is a need. The payers are seeing this. There is no process.
Comment: this is a process we have been asking for the past 25 to 26 years. We are asking for a registry for data types of HL7 standards or z-segments in HL7 V2.
Let’s bring it to a small use case.
Comment: There is a US REALM for profiles. We need to be extremely precise. The core resources are managed by HL7. This is about profiles and extensions to existing resources. This is similar to HL7 optionality.
Correct term: Repository process and requirements. How does the tool help us?
Argonaut will develop a resource.
In an ideal world it will be managed by the FHIR management team.
Lenel is asking CIC to join the attachments WG for the PSS.
Take it from CIC to DESD and then to TSC.
Decision: change the PSS to universal domain.
Comment: the need is here and focus on implementing this in phases.
Attachment, CIC and financial management and CQI WGs are all interested in this PSS.
End game needs to be clear and the end product is a white paper. Agree and vote, you are willing to be a prime co-chair.
What do we need CIC to be primary?
Comment: small scope and will be a larger effort managed by the larger FHIR management team.
Look at what has worked and what has not worked.
Look at eMerge.
Look at what NCI and Dianne Reeves has developed.
Rules in extension of FHIRs and process should be included similar to chapter 2 of the EHR-S FM R2.
Action item 2: CIC will find a quarter for the FHIR Resources project for the September 2016. Schedule for Tuesday Q4 for Baltimore.
HL7 CIC – RCRIM and BRIDG: Discuss Bipolar Disorder Project and other Mental Health StandardsLocation: Montreal, Canada / Date: Tuesday 10 May 2016
Time Q 1
Facilitator / Anita Walden / Note taker(s) / Mitra Rocca
Attendee / Name / Affiliation
Quorum Requirements Met:
· Anita provided an overview to the CIC agenda.
· Anita asked the ACC representative about their interest in participating in the patient registries project. A member from the SOA WG was among attendees. They have been leveraging a modeling approach to do traceability and gaps in information models. It is based on an OMG model and is looking at gaps in FHIR, FHIM and other models. It will provide SMEs with where the issues are. Bottom up perspective, concepts and cardinality, the team deals with relationships between semantic concepts. Ex: For the Federal Health Information Model (FHIM), we annotate FHIM with these semantic concepts.
· AMS has been working on use cases.
· Bipolar DAM
· Schizophrenia is now integrated into CDISC SDTM and available on CDISC’s Web site.
· CDISC took the Schizophrenia DAM and developed SDTM. CDISC also did the same for TB and built it into SDTM.
· Standardize the data elements at the smallest unit possible, and they can be reused by EHRs. And can be used reported out for quality improvement, etc.
· For Bipolar disorder, we will continue with ballot reconciliation and update it by July 2016.
· General Anxiety Disorder DAM
· Synthesis of data elements from FDA’s submission (CRFs, DSM, SCID and NIH).
· CIC is using a new tool (MAX Tool)
· James provided an overview to MAX tool:
· Overview: MAX tool is a program that allows you to export the data from EA into Excel and import them back to EA. It saves time with copy and pasting.
· Anita asked James to export common data elements from previous TAs in MAX tools and included them in General Anxiety Disorder.
· Q: How did you identify the common data elements?
A: once we had schizophrenia, we looked at Bipolar, we reviewed them with the FDA reviewers and also by other clinicians.
· Q: where sources of data included in EA? Do you capture provenance, esp. with CDEs?
A: In BRIDG, we capture, where something came from. We can query data elements behind BRIDG.
A: Duke is not currently capturing the provenance. If something comes from CRF, we do not indicate that it comes from CRFs, but we cannot share them. That is propriety.
We have a model in EA at SOA, and it is UML 2.0 compliant.
Comment: we need a provenance from EA to Excel.
We want to move away from Excel and use EA.
Anita: we have a lot of anticipation for MAX tools.
EA has table viewer from their models.
Excel is something that clinicians (SMEs) are familiar with. MAX tools will not be provided to SMEs.
There are different modeling perspectives.
Where you have permissible values vs. yes/no, do you have definitions for them?
A: we try to add definitions for them.
· Q: Do they tie back to standard questionnaires?
A: ask them to look at questionnaires. We ask the clinicians to look at data elements and questionnaires.
We use ISO 11179.
In data elements review, review the unnecessary data elements
Update on BRIDG
BRIDG was balloted in May 2016 and the 2 negative comments are being discussed in the BRIDG WG:
1. Vocabulary binding and to have value sets
2. Smaller packaging of BRIDG concepts
In BIRDG 4.1 we published 4 or 5 smaller views.
We are harmonizing BRIDG and DICOM. We are modeling by reference. There are other expertise, and put those pieces into BRIDG.
Modeling by referencing.
BRIDG is also talking to the clinical genomics WG.
What are the touch points where we can harmonize with the CIC DAMs?
Process: we met with Julie Evans to find out what is useful for BRIDG. CDISC is working on CFAST and they are building BRIDG into the concept maps.
We are working with BIRDG and CDISC on common concepts.
Move non-formal diagrams (Concept maps) to formalized phase.
The OWL representation of BRIDG is available on bridg.org
HL7 has 100s of models, like PID in HL7 V2, and patient role in V3, patient segment in FHIR.
BRIDG is looking at FHIR.
Model driven approach for gap analysis and traceability
Comment on BRIDG: there were a few classes without description.
Is there a need to indicate which diagnosis was associated with a particular patient episode?