HL7 SDWG Meeting · There is a master conformance resource - list of resources and operations needs to be supported. What data elements should come back? #153 How is this MU data

HL7  SDWG  Minutes  Jan.  19th  -­‐22nd  2015  

 Attendance  Sheets  at  Bottom  of  Document  

Monday Jan 19 Q1 Business Meeting Agenda Review

Proposal that we make Mark Roche an interim co-chair for this week Against: 0 Abstain: 1 For: 29 motion passes

Update SDWG 3 Year Plan

Structured  Documents  Work  Group  2014  -­‐  Three  Year  Plan  

January  19,  2015    

• New  release  of  CDA  • C-­‐CDA  Maintenance    &  Update  

o Examples  Task  force  o Value  set  ownership  and  maintenance  o Enhanced  conformance  testing  for  C-­‐CDA  

• Templates  based  on  detailed  clinical  models  • Support  

o Help  Desk  o Certification  Testing  o User  Groups  

• Disease  Specific  IG  • Registry  Reporting  (Co-­‐sponsorship)  • Public  Health  Reports  • CDA  Product  Families  

o Governance,  Management,  Methodology  establishment  • FHIR  

o Ongoing  Coordination    • International  Version  of  the  C-­‐CDA  /  IHE    • Template  Repository    • Analysis  followed  by  Governance  oversight  for  implementation  of  Template  DSTU  • Tracking  of  all  know  CDA  R2  extensions  • US  Realm  Projects  

o Structured  Data  Capture  o Data  Access  Framework  o Data  Provenance  o TBD  

SDWG  SWOT  Analysis  January  19,  2015    

Strength  -­‐  Highly  relevant  committee  with  a  lot  of  interest  and  plenty  of  participants  -­‐  Cited  in  MU    2014  edition  -­‐  Broad  scope  and  templating  strategy  -­‐  Expanding  international  use  and  sharing  of  IGs  -­‐  Stable  and  widely  implemented  standards  -­‐  Tools  exist  for  building  implementation  guides  -­‐  Examples  task  force  supporting  the  CDA  implementer  community    

Weakness  -­‐  Bandwidth  limitations  (work,  meetings)  -­‐  Lack  of  user-­‐friendly  tools  &  documentation  to  support  implementations  -­‐  Unclear  CDA  strategy  moving  forward  -­‐  Focus  on  new  work  instead  of  sustaining/maintaining  existing  work  -­‐  Lack  of  a  international  registry  /  repository  for  CDA  Templates  and  historical  artifacts  -­‐  The  size  of  C-­‐CDA,  is  a  challenge  to  support  in  a  purely  volunteer  environment.  -­‐  Lack  coordination  with  Publishing/Electronic  Services  &  Tooling    work  group  

Opportunity  -­‐  Collaboration  with  other  groups  

-­‐  IHE  -­‐  ONC,  DOD,  VA,  FDA,  CDC,  CMS,  …  -­‐  HIMSS  HealthStory  -­‐  DICOM  -­‐  Professional  Societies  

-­‐  AMIA,  ASCO,  ACP,  …  -­‐  Have  major  impact  on  the  global  healthcare  community  and  healthcare  delivery  -­‐  Working  with  transport  mechanisms  to  have  a  complete  solution  -­‐  FHIR  –  aligning  FHIR  documents  with  CDA    -­‐  New  CDA  Release  (2.1)  -­‐  Support  implementation  concerns  and  improve  interoperability    -­‐  Showcasing  International  CDA  Implementation  

Threat  -­‐  Crumbling  under  too  much  interest  /  work  -­‐  CDA  R2  semantic  expressivity  limitations  -­‐  Competing  areas  of  interest,  pulling  resources  away  -­‐  Lack  of  volunteers  and  support  for  maintenance  of  legacy  standards   C-CDA Maintenance & Update Expiring Standards – deferred, as there was a report error in this listing.

Review PBS metrics and work group health • Project Report Card is looking good • The Work Group Health looks good

Update on current ballot - number of comments and planned ballot reconciliation

QRDA – 143 DAF – 291 Birth Registry - not posted Trauma Registry FHIR – 14 comments Upcoming ballots HAI – Normative Ballot – possible - refactoring FHIR – Composition Resource CDA in FHIR Presentation on any new projects – none

Co-chair availability – discussed & updated on the wiki site SDWG co-chair openings - reviewed Discuss Paris Agenda – 9 Participants Mission & Charter

Need to speak to FHIR relationship Austin will make changes. Decision Making Practice Discussion of Providence Q2 MEETING

Project Argonaut relationship to SDWG FHIR work C-CDA on FHIR Brett made a presentation about the C-CDA on FHIR and the Argonaut project DAF ballot reconciliation - Comment 291 comments 130 negatives 160 affirmatives I&M Process for ballot resolution MARCH deadline Propose disposition today with provisional votes by SDWG, review with I&M if this is OK Implementation Guides – DAF Emma #79 – Reference of the identifier on Patient – and issue with reference MRN – the patient identifier field How do you define a URI for defining the scope of the MRN identifier? Resolution: Add an example on how an MRN could be represented in FHIR. Once with an OID and one with URN. Business or Domain Identifiers the other is a resource identifier. Any time we constrain something we need to have a note that does not conflict with the base definition. Rule: If you don’t constrain it, don’t carry the note forward. If you do constrain or modify, add note for the item. Persuasive w/ Mod

Question about the usability of profiles

There is a master conformance resource - list of resources and operations needs to be supported. What data elements should come back? #153 How is this MU data - Allergies From where did they derive the data in the Allergy Intolerance profile? Agree The table will grow into three columns: 1. As defined in MU 2. Priority DAF elements 3. FHIR Resource Mappings

Persuasive with mod

The DAF and C-CDA on FHIR need to coordinate their work. #133 (#75 – similar comment) – What is the definition of must support? Found it, does the profile support the definition of must support. Balloter requested that the comment be changed to a comment w/ questions. Need to review with FMG on the Must Support and how to highlight in profile. How does Must Support work in FHIR? 1..1 – mandatory and must support vs. 0..1 and must support There is no way to summarize the Must Support rules First time profiler and first time reader #151 – abatement Boolean [x] Need to add a note to indicate the other options have been removed.

Complete discussion and will bring forward the issue of snapshot table.

Need to be real clear what the snap shot table and the differential table mean. It is 7 or 8 weeks before the DSTU ballot needs to be done. The calls – I&M Q2 Tuesday will determine and post to both SDWG and I&M #167 - Procedure – codeable concept – shouldn’t be further constrained to other vocabularies – need to align with C-CDA Should we align with C-CDA or are there value sets that are missing? (Preferred, extensible)

update dafEncounterType to dafProcedureType For SNOMED, restrict to Procedure hierarchy. Vote on the items discussed in Q2 from the ballot Rick made, Gay second 4 comments – against – 0 abstain 2 for: 31

Q3 MEETING

C-CDA R1.1 Implementation Issues - open discussion Large CDA documents (length) The overwhelming volume makes it unusable. One problem – on receiving end mange the order of the materials presented. The reason for referral is not that useful Content that is incorrectly categorized Medications that are not relevant Relevant is not computable There is a lot of dirty data What is generated is a data dump – we do not provide guidance on this The work that we are doing for data providence, security and privacy will only enlarge the data set. We need to provide guidance on how to do reconciliation or consume. There is a process for the issues that needs to be considered There is likely a need to create best practice guidelines, which can be shared with the industry to improve interoperability. There is a problem – as a producer has filtered the content there is not way to say what filtering was used. There may need to be footer at the bottom, which says how the information was filtered. How do you organize the data out of the EMR to make it effective to use. A new section, which is called synopsis, might need to be provided. As physicians are reconciling the medication lists, it would be useful to clearly identify which medications that are no longer active. CDA does not require that the prescriber needs to be identified Vs Active We need to take some PHI de-identified examples as a source document, to see the issues that are being created in the industry. A couple of issues identified were entered on the DSTU Comment Site for the C-CDA R1.1 DSTU. #568 Update the samples to conform to the standard C-CDA R1.1 3 issues Errata Windy made Rick second --- against: 0 Abstain: 0 For: 29 motion passes #567 Will remove 8720 conformance statement - errata

Windy made Second Rick – against: 0 Abstain: 1 For: 30 motion passes #566 – procedure note template error 8507 - There is a technical error in the formulation of the constraint Errata – move the attribute binding to a child constraint Windy made, Second Rick Against: 0 Abstain: 1 For: 30 motion passes

Other implementation issues All the documents are CCD and none of the other documents are sent. Cover letter – is faxed and sent Next steps – what can we do to improve the systems 1. Large documents, - have a project to make the documents relevant to use cases for

physicians. – Keith want to help with this. Do you or do you not include your full note in your CCD – inclusion of raw clinical notes? Another issue is the inclusion of patient education materials – why does the doctor need those materials. It was recommended that SDWG stay engaged with the Standards Content work group. Do we want to set aside time on the call on Structured Documents? CMIO feedback and other groups – find the right committee and find the time. Feb 5th – for this issue

Examples task force – developing samples Would it be a good idea to seek samples from others? Pushing out the samples from what we have learned Enhancing the Smart CDA Trying to collect more de-identified samples that are coming from others. We might have John Devork make a presentation on samples that have been generated. There is a repository on site – ONC – a number of samples

The co-chairs thanked the members for their participation on the discussion and indicated that we would continue to discuss these issues and work on plans to help resolve them for membership and the industry.

SDTC extension management and ongoing maintenance of CDA schema - Lisa Nelson We voted to approve a number of extensions on the SDWG

Would like to add a date to manage the approval of the extensions and where version of the schema. It was suggested that we need a more formal process for extensions, so that when they are approved, members would know where to find them. Austin indicated that GForge is would current platform for the publication of this materials – Austin indicated that he would be willing to create a subdirectory on GForge that could be used to store and provide members with access to the current extensions for CDA. Q4 HOST PHER

Send representatives to vocab to discuss value set stewardship (Brett+) Refactoring the HAI implementation guide Karen had a couple of topics Birth Defects IG – reconciliation will occur this week – Tuesday Q3 Birth and Death Registry IG Cancer Registry IG - We will plan on meeting with PHER in Paris Decision Making Practices - Once every few years we need to review and establish our default decision-making practices. Calvin made, Second Thomas - move to retain the 2 co-chairs and 3 other members’ rule for quorum. Against 0, Abstain: 3, For: 5 for motion passes Calvin moved, Second Abdul M. To adopt the Decision Making Practices as revised. Opposed: 0, Abstain: 1, For: 7 motion passes. Tuesday Jan 20 Q1 MEETING joint with CQI

Send Representatives to CQI (CQI Hosting) (These are notes the official minutes will be posted by CQI) Tacoma Ballot Update Clinical Quality Framework – Presentation on the public / private partnership Harmonize the standards used for clinical quality and clinical decision support.

1. Common Expression Logic Standard – CQL a. Shared sample syntax for the CQL b. Balloted in 2014, DSTU this cycle

2. Common Data Model Standard – QUICK* a. Quality Information Data Model – a profile of FHIR b. They are aligning with FHIR resources c. Alignment with DAF and CIMI and C-CDA profiles d. Req. May 2014, Comment only Sept 2014, DSTU ballot May 2015

Additional ballots on HQMF and others 3. Common Metadata Standard

Pilots - providing feedback on the standards as they are being developed. Midway thought the standards development and review part, they have wrapped up the initial phase of pilots and wrap up in 2015. They are looking for participation by others and are using the www.cqframeworkinfo Expression Language team and Data Model team meeting on Wednesday. Reconciliation – FHIR Profile for Quality (30 min) CQI – will do their reconciliation in their working group CQL – a few items (15 minutes) A couple of items for CQL Data model – QUICK / FHIR quality profiles given that the timelines for the items to be coordinated do not link up. It would be useful to know how CQL will work with C-CDA. Comment by Gay The thought for the specification is to include examples in the documentation on how this would work, but not providing a mapping – CQL is data model agnostic. QRDA – 1 item for reconciliation (20) Q2 Open Quarter for SDWG QRDA Ballot Reconcillation #124 - priorityCode does not seem specified. Has the wrong value set in examples – need to work with the measure and place a note in the spec. Q3 HOST ITS/MnM/Publishing/Tooling (Rick /

Topic: • CDA R2.1 questions regarding adopting newer RIM vocabulary • CDA Template modelling and vocabulary questions (S. Gaunt)

Problem Type, LOINC code names have “...HL7.CCDAR2”. CCDAR2 is method. Issue is resolved in updated LOINC db. How do we represent “I did not do any of the above” (e.g. none of above medications, or problems...etc). Currently, this is expressed using negationInd. We added extension in CDAR2 with a value set OID and a nullFlavor (Check with Vocab/MnM on a periodic basis). 3 options for nullflavor: OTH, UNKN, and NA Votes “Yes” votes for least offensive option: NA (25/36), UNKN (5/36), OTHR (6/36)

Q4 HOST FHIR

CDA on FHIR Using clinical documents in FHIR Composition resource will be deferred to SDWG Mapped CDA elements to FHIR Composition Resource available at “CDA Mapping” Google sheet. Next step: Argonaut mapped C-CDA to FHIR Put together a presentation/tutorial by Sep 2015, on how C-CDA maps to FHIR. Review during CDA on FHIR calls on Tuesday.

Wednesday Jan 21 Q1 JOINT with CQI (CQI Hosting)

QRDA Ballot Recon Calvin to meet with II WG Q2 No SDWG MEETING Q3 MEETING

(In SDWG) Refactoring the NHSN HAI CDA IG (A Kreisler) Send Representatives Pharmacy (Not a JOINT meeting) SubstanceAdministration Free Text Sig Proposal (Keith) Medication statusCode Discuss Loading Dose

Q4 No SDWG MEETING Thursday Jan 22 Q1 Host Imaging Integration

TOPICS: Radiology CDA IG / LOINC code use in DIR Revising  document: “Supplement 155: Imaging Reports using HL7 CDA” (revision and replacement of PS3.20)” (ftp://medical.nema.org/medical/dicom/supps/Drafts/Sup155_2014-06-25.pdf ) went through public comments, made substantial changes after public comments, most of changes to stay aligned with structured docs templates. Using SNOMED CT to encode field names for “Measurements” such as “Area Measurement”, “Length Measurement”, etc. Using SCT as opposed to LOINC due to the 17-yr agreement with SNOMED CT that predates recent agreements within HL7 to use LOINC as code system for all observables. UCUM are used for  Units of Measure (since 2002). Also, created new data elements for CDS purpose that enables verification whether radiologic study was correctly ordered (was procedure appropriate to indication). As a result, the IG needs new LOINC codes for these new CDS-related data elements. What does SDWG think about  proposed structure (CDS-related as explained above)? Discussion on whether SNOMED CT or LOINC codes should be used for observations. For now, it should be SNOMED CT because that is what most systems in world do today. Discussed whether Radiology report template “Document Type” in CCDA R2,0, Table 34: “LOINC Imaging Document Codes” (OID: 1.3.6.1.4.1.12009.10.2.5) should use one

generic LOINC code (e.g. Radiology Report) or several more granular LOINC codes (e.g. 24754-4 - Administration of Vasodilator into catheter of Vein). Harry Solomon to provide DSTU comment on C-CDA R2.0 to update the Value set (Table 34) based on DICOM preferences Also Table CID x7021: “Imaging Report Document Title” LOINC codes in ““Supplement 155: Imaging Reports using HL7 CDA” (revision and replacement of PS3.20)””, to be aligned with Table 34. Conclusion:

Discuss progress of discussion and updates to documents at May 2015 meeting in Paris.

Q2 MEETING

JOINT w/PC, Templates (PC Hosting) Allergy/Intolerance discussion with Patient Care (Gaye Dolin)

Q3 HOST FHIR

Review FHIR ballot comments on composition resource (cbeebe)

Motion made by Calvin, Kevin second Adopt the modeling of section as depicted by Calvin on the screen.

Opposed: 2 Abstain: 9 For: 8 Motion passes

Q4 MEETING

CDA R2.1 project Where we are at with vocabulary

I&M / ITS / Electronic Services Tooling Abdul M. – Pick the version of the RIM to bind the model to. Ted Cline - provides – it is possible to do, but might be dangerous. Non-structural Codes – Participation Types, Act Relation, Role Codes If we cherry pick, we would have to hack the schema. M&M – look at the analysis SNOMED – LOINC – IP will be required Request to discuss - cultivating Vs owning – FMG to formally define Who and How to decide 2.47 RIM Structured Document – CDA R2.1 RIM 2.07 -Vocabulary of a current version M&M -- get on the M&M list Meet with Vocabulary – Readness – for March 22 – doubltful – OO FMG / FGB A topic for discussion to harmonization Facilitor Meeting Harmonization Meeting n March 3,4,5 possible dates

Initial proposals in by Feb 1st – for discussion on topic

and technical review Feb 2-5 Final are due the 22nd – final materials for discussion

Non structural vocabulary – use the latest release not Doodle Poll – for the timeEntry à Grouper Look at the WIKI to work on the ballot – secure wiki site Katherin - Discuss Paris  Agenda  

      EU  /  US  –  CDA  Project         Austin  updated  the  WIKI  for  the  Paris  SDWG  Meeting.    Friday      Jan  23    Q1   Joint  w/Templates,  PC  (Templates  hosting)  These  are  notes  from  the  joint  meeting  with  Templates  and  PC.    The  official  minutes  will  be  posted  by  Templates  WG.      Templates  DSTU  1    Art  Décor      

 Paris  –  Friday  Q1    Meetings  during  the  week    CGIT  and  FHIR  and  Templates  -­‐ meta-­‐data  sync    -­‐ core  principles  

o Kia  –  wrote  a  document  about  it  o Core  Principles  –  sync    o A  template  which  is  a  set  of  constraints    

§ FHIR  does  not  support  the  connections    • This  set  of  constraints  on  a  profile  map  to  a  template  

§ There  are  issues  with  the  mapping  that  is  needed.  § Graham  –  Cycil  –  profile  mapping    § Make  it  computable    

o There  is  a  need  to  improve  the  documentation  on  Profile    o Template  defintions  –  can  refer  to  definitions  in  another  model  

§ CIMI  /  Detailed  Clinical  Models  • Mapping  element  in  resource  goes  to  model  

§ There  is  a  place  holder    § Is  there  an  XML  tag  that  is  used  for  mapping  to  source  domain  models  or  

detailed  models.  o They  are  using  an  ID  in  each  and  every  functional  model  for  the  attributes  in  the  

models.    -­‐ Templates  DSTU  has  ids  assigned  for  the  element  /  attribute  definition  -­‐ DCM  –  Detail  Clinical  Models    -­‐ Vs  CIMI  models    Discussion  with  Templates  about  how  we  might  be  able  to  map  a  general  model  to  a  detailed  model  and  fill  in  some  of  the  attrbutes  based  on  a  link  to  a  LOINC  code  to  fill.    

1.3 release  governance  groups  –  can  manage  templates  and  model  access      FHIR  strategy  will  be  provided  in  1.3    IHIC  –  Prague    Art  Décor  developer  day      DSTU  Comments  for  Templates    Templates  WIKI      Meeting  ended  early,  as  there  was  a  fire  alarm,  which  required  that  we  leave  the  hotel.