HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D....

24
HL7: Today and HL7: Today and Tomorrow Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009

Transcript of HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D....

Page 1: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

HL7: Today and TomorrowHL7: Today and Tomorrow

HL7 UK 2007 Conference

Making Interoperability Work

W. Ed Hammond, Ph.D.Professor EmeritusDuke University

Chair HL7 2008-2009

Page 2: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 2

If we could first know where we are and whither we are tending, we could then better judge what to do and how to do it.

Abraham Lincoln

Page 3: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 3

The BeginningThe Beginning• Formed 1987 to create required standards to enable “best of breed”

approach to hospital information systems• Small, overlapping working group with few committees: focused on

admission, discharge and transfer (ADT); orders and results; patient administration; control and query – No top down or bottom up

• Independence of governance and working group activities– Chair, secretary and treasurer elected; rest of board appointed– Primary Stakeholders – vendors and providers; formed IAB and PAB– Hired management group in 1991

• Version 2.x driven by simplicity; implicit model: see it, do it– Basic building block were data items combined into logical segments– Messages driven by trigger events and consisted of segments in defined

sequence. Segments could repeat and some were optional.– New required elements were sequentially added to existing segments– New functionality created new segments.– Z segments permitted independence and expansion.– Implementation was at the interpretation and needs of the implementer.– Backwards compatibility demanded.

Page 4: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 4

The PastThe Past• Formed Technical Steering Committee with chair, vice chair• Evolved to international organization• Evolved to largely elected board (8 at large, including

international) (TC Chair and vice chair non-elected members)

• Convergence of governance and technical work; mainly bottom up and volunteer driven

• ANSI accreditation influenced governance and balloting• Grew rapidly in numbers: membership, technical

committees, special interest groups; scope expands• Addition of SGML/XML activity; vocabulary; clinical

interests; moved toward model driven approach resulting in RIM, v3, CDA

• Loss of communication among groups; increased time to create standards

Page 5: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 5

The past continuesThe past continues• Migration to model-based v3 standards

competed with existing base of v2 standards; co-existence demanded

• Increasingly unsolved issues impacted efficiency of organization

• Increasing competition among groups both national and international

• Interests from government and regulatory bodies• EHR standards activity – expanded participation• Increased importance of ambulatory care

standards• Wider audience, expanded set of stakeholders

Page 6: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 6

The PresentThe Present• Board becomes strategic; Technical Steering Committee

controls technical standard development• Hiring of a Chief Executive Officer, Chief Technical

Officer; Executive Director of Management Group became Chief Operating Officer

• New structure organized around common themes– Foundation and Technologies– Structure and Semantic Design– Domain Expertise– Technical and Support Services

• Increase in top down governance but still balanced by bottom up influence; still a volunteer organization

• Importance of global approach recognized• Increased efforts of joint activities among SDOs in

creating standards

Page 7: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 7

Still the presentStill the present• Pressures on HL7 from multiple groups, including US

Health Information Technology Standards Panels (HITSP)

• Movement towards testing of standards through Draft Standards for Trial Purposes

• Perceived increased complexity in implementing model-based standards

• Increased requirements for family of implementation manuals

• Increasing number of groups creating standards• Pressure to create standards more quickly• Increased interest in standards throughout healthcare

community• New focus on interoperability standards• New attention from clinical community

Page 8: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 8

Looking at the futureLooking at the future

Page 9: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 9

Recognize changesRecognize changes• Addition of genomic and proteomic data to

health record• Movement toward predictive and preventive

health model with personalization of care and treatment

• Modifications in workflow• Increased demands for quality, patient safety,

access, efficiency, effectiveness, integrity, privacy

• Integration or data from disparate sources; life and death decisions based on integrated model

Page 10: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 10

?

InteroperabilityInteroperability

Semantic Interoperability

BusinessInteroperability

Functional Interoperability

Privacy& Security

Interoperability

Page 11: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 11

No single focusNo single focus• Geographical

– International– National– Regional– Local

• Stakeholder– Providers, vendors, payers, governments, consultants, payers,

researchers, knowledge brokers, regulators, consumer, employers, suppliers, imaging (PACS), others

– Interoperable solutions must satisfy all

• Sites of care– Inpatient, ambulatory care, emergency care, nursing and long

term stay facilities, home care

• Multiple diseases; increasing focus on chronic diseases

Page 12: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 12

ChallengesChallenges• HL7 must view itself as part of a larger community

dedicated to a specific role in using IT to improve health care. HL7 must become involved with that broader community; we must form new relationships and new partnerships.

• Standards require both a policy and a technical framework to be effective. What is HL7’s role in policy? How do we influence policy?

• HL7 needs to understand the problems with today’s health care and how IT may improve the system.– Does more data mean better care?– What is the value of IT in health care?– What standards are necessary and what else is required for

sufficiency?• When does HL7 drive and when does HL7 follow?

Page 13: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 13

What does the future hold?What does the future hold?• Evidenced based medicine

– Standards for knowledge representation, knowledge extraction for data, knowledge transfer and knowledge use

– Standards for clinical guidelines, care plans, decision support

• Translational research– Reusable data: clinical trials fed by patient care data– Decrease time from bench top research to routine

bedside use

• Query standards that make it easy to access information; push and pull standards

Page 14: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 14

What is HL7’s role in semantic What is HL7’s role in semantic interoperability?interoperability?

• Ontology, terminology, vocabulary• Data elements and attribute sets

– Nationally or globally?– Clinical vs informaticists?– Data types, names, definitions, units, value sets

• Compound and complex structures– Templates, archetypes, CMETs, Clinical Statements

• Mapping vs single integrated set• Collaborate vs doing vs endorsing

Page 15: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 15

Future requirements?Future requirements?• Understanding content, similarities and

differences in different views of health care data– EHR, EMR, population or summary records, personal

health record– Variation in sites and purpose; linkages and

relationships

• Views, content and functionalities of– Regional health information systems (RHIO or HIE)– States or provinces– National (NHIN)

Page 16: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 16

Establishing trustEstablishing trust• Security, Privacy and Confidentiality

– Authentication– Authorization– Access– Audit Control– Digital signature– Integrity– Non-repudiation– Encryption– De-identification standards– Probability of risk vs value

Page 17: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 17

Accommodating technologyAccommodating technology• Move to service-oriented architecture

approach– Service components can be modified or

expanded independently

• Keep solutions as technologically neutral as possible

• Provide functionally rich and well-documented tool sets

Page 18: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 18

Other servicesOther services• Rule-based data exchange• Filters for data presentation• Mapping services in transition• Consent management• Identifying candidates for clinical trials• Record linkages• Digital identifiers• Notification services• Business processes• Natural language processes

Page 19: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 19

What is the future HL7?What is the future HL7?• HL7 no longer can exist as an

independent group of “techies” making standards as we please.

• How does much of the broader community view HL7? The image must change.

• HL7 must change its vision to understand its role in an expanding and evolving world; we must become part of the solution and claim that role.

Page 20: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 20

What is the role of HL7?What is the role of HL7?• HL7 needs to determine what standards

are required to meet national visions for use of IT in patient care, research, reimbursement, performance evaluation– What can and should HL7 do itself?– When should HL7 collaborate and with

whom?– When should HL7 endorse?– What should HL7 leave to others?

Page 21: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 21

Questions HL7 Must AnswerQuestions HL7 Must Answer• Should our vision be broad or narrow?• What is the proper balance between academic

and “real world” approaches?• What is balance between national and global?• In making a standard

– How many people are required?– How long should it take?– Is creating different than approving?

• What else is required beyond creating a standard?

Page 22: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 22

Questions for the futureQuestions for the future• What is the balance between top down and

bottom up governance?– Broader scope required; how dose HL7 influence

completeness among volunteers?– How do we deal with gaps?– How do we participate in joint initiatives?

• What support functions should HL7 provide?– Repositories (data elements, data structures, queries,

reports, decision support algorithms, templates, guidelines, use cases: some, all, or none)

– Certification (of what)– Testing (of what)– Tools (who and how defined)

Page 23: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 23

What level of cooperation is What level of cooperation is possible?possible?

• Is one SDO better than many?

• Are we willing to:– Create standards without worrying about who

gets credit?– Accept the best solution regardless of source– Contribute freely– Accept direction– Cheer other’s successes– Look for solutions beyond the obvious

Page 24: HL7: Today and Tomorrow HL7 UK 2007 Conference Making Interoperability Work W. Ed Hammond, Ph.D. Professor Emeritus Duke University Chair HL7 2008-2009.

21 Nov 2007 HL7 UK - Making Interoperability Work 24

The future awaits!