Healthcare IT: Why is this so hard Can we build meaningful solutions

1
Healthcare ITWhy is this so hard?Can we build
meaningful solutions?
The great thing about standards is that there
are so many to choose from. --- anonymous

• Charles Mead, MD, MSc
• Senior Director, Healthcare Strategy
• Oracle Corporation
• charlie.mead_at_oracle.com
• Health Informatics Society of Ireland (HISI)
• Dublin, Ireland
• November 16-17, 2005

2
What is the (perceived) problem?

• Ineffective, inconsistent, inefficient,
  non-efficacious healthcare delivery
• Too much information for timely integration
• Ever-changing information base
• Mobile patients/global society
• Complex payment/reimbursement frameworks
• Time-consuming bureaucracy
• Increasing financial pressures
• ?? Unrealistic expectations ??

3
What is the (envisioned) solution?

• ?? Networks, Email and Web Browsers ??
• The United States National Health Information
  Infrastructure (NHII)
• Inform clinical practice
• Interconnect clinicians
• Personalize care
• Improve population health
• The Electronic/Enterprise Health Record
• Bedside to Bench and Back

4
A caBIG Example(from Covitz et al,
Bioinformatics, V19, N18, P2404)

• Patient presents with headache, focal weakness,
  history of seizures
• Workup reveals glioblastoma multiforma subtype
  astrocytoma
• Is this tumor histology is associated with gene
  expression abnormalities?
• Yes, in the p53 signaling pathway including BCL2,
  TIMP3, GADD45A, CCND1
• Is there documented evidence of aberrant
  expression of (e.g.) CCND1?
• Yes, SAGE tags for cyclin D1 appear with 3x
  greater frequency in cancerous vs normal brain
  tissue
• Are any gene products of the p53 signaling
  pathway known targets for therapeutic agents?
• Yes, TP53, RB1, BCL2, CDK4, MDM2, CCNE1
• Are any of the agents known to target these genes
  being specifically tested in glioblastoma
  patients?
• Yes, trials xxx and yyy are currently underway
• Research data at the point of care, Clinical
  data at the point of research

5
Cut to the Chase

• Implementing an Enterprise/nation-wide anything
  is hard
• Implementing an Enterprise/nation-wide EHR is
  particularly difficult because
• the essence of the problem is about computers
  processing data that has traditionally been
  processed by people(e.g. clinicians,
  researchers, administrators, accountants, etc.),
• human beings will still be involved in any
  successful (relevant) Enterpriseion/na-wide EHR
  implementation,
• and so enter the dreaded concepts
  ofUsabilityRelevanceEfficiencyCostComputabl
  e Semantic Interoperability

6
Two Types of Tools Are Needed

• Networking/connectivity tools
• Enterprise-wide means moving data in time and
  space between multiple sources
• Variable hardware / software profiles
• Synchronous / Asynchronous communication
• Persistence requirements (archiving)
• Auditing requirements (audit trails)
• Software Development / Data interchange tools
• The patient gets chest pain when they walk past
  the frozen food section of the store
  Patient has angina aggravated
  by cold.

7
The Good News

• The network/connectivity tools exist
• Hardware / Firmware
• Networks and Exchange Protocols (e.g TCP/IP)
• The Internet and its derivative technologies
• Email, Intranets, VPNs, etc.
• Software
• Browsers reading XML / HTML
• The data interchange standards exist
• HL7 Version 3 (and its various equivalence
  mappings)
• Terminologies (SNOMED, LOINC, etc.)
• Domain-specific representation standards (DICOM,
  MAGE)

8
The (Pretty) Good News

• The Software Engineering techniques and tools
  exist to build usable, relevant system
• Unfortunately, within healthcare, they are most
  often not applied correctly or to the degree that
  they need be to ensure that a truly relevant
  system is designed and built
• Requirements definition and management
• Quality requirements poorly defined (e.g.
  usability, performance, etc.)
• The Communication Pyramid
• Project Management

9
The (not so) Bad News

• Even with contemporary tools and expertise, its
  still a difficult problem to solve
• Expensive
• Requires specialized expertise
• Modeling, requirements specification, interface
  design, etc.
• Requires a top-down commitment to standards
  that is often not natural to many healthcare
  providers, institutions, etc.
• Market pressures in combination with government
  incentives
• Use of Guidelines / Outcome-based reimbursement
• Unique person identifier / suspicion of
  goverment
• However, other industries have adopted
  standardization and built complex systems

10
The Really Bad News

• The inherent problem is hard, i.e. it will never
  be easy!!!

11
The Semiotic TriangleHow Humans Communicate

• Protocol Example

12
Protocol and the Semiotic Triangle
Symbol Protocol
Source John Speakman/Charlie Mead
13
Interchange vs Interoperability

• Main Entry interoperability ability of
  a system ... to use the parts or equipment of
  another system Source Merriam-Webster web site
• interoperability ability of two or more
  systems or components toexchange information and
  to predictably use the information that has been
  exchanged. Source IEEE Standard Computer
  Dictionary A Compilation of IEEE Standard
  Computer Glossaries, IEEE, 1990

Syntax ? Structure Semantics ? Meaning
14
The Pillars of(Computable Semantic)
InteroperabilityNecessary but not Sufficient

• Common reference model across all
  domains-of-interest
• Information model vs Data model
• Model grounded on robust data type specification
• Methodology for binding terms from concept-based
  terminologies
• A formally defined process for defining specific
  structures to be exchanged between machines, i.e.
  a messaging standard

The Version 3 Tool Kit
15
HL7 V3 Reference Information Model (RIM)
An instance of an Entity may play zero or more
Roles. Each instance of a Role may, in turn,
play zero or more instances of a Participation in
the context of an instance of an Act. Each
instance of a Participation participates in a one
and only one Act for the duration of that Act.
Acts may be related to each other through
instances of Act Relationship.
ActRelationship

• Has component
• Is supported by

0..
0..
1
1
Entity
Role
Participation
Act
1
1..

• Referral
• Transportation
• Supply
• Procedure
• Consent
• Observation
• Medication
• Administrative act
• Financial act

• Organization
• Place
• Person
• Living Subject
• Material

• Patient
• Member
• Healthcare facility
• Practitioner
• Practitioner assignment
• Specimen
• Location

• Author
• Reviewer
• Verifier
• Subject
• Target
• Tracker

16
Collection, Context, and AttributionBuilding
Complex RIM-based structures

• A diagnosis of pneumonia (observation Act)
  related to three other observations Acts. Each
  Act is fully attributed with its own context of
  Entity-Role-Participation values.

AR is supported by
OBS Temp 101F
Attribution
is source for
has target
AR is supported by
OBS Dx Pneumonia
OBS Abnormal CXR
Attribution
has target
is source for
AR is supported by
is source for
Attribution
OBS Elevated WBC
has target
PARTICIPAT Subject
ENTITY Person
ROLE Clinician
Attribution
PARTICIPAT Author
ROLE Patient
17
Information vs Terminology ModelsIntersecting
and interleaving semantic structures
Terminology Model
Domain-Specific Terms specifying Domain-Specific
Semantics
Binding/Interface
Information Model
Terminology Model
Common Structures bound toDomain-Specific
Structures specifyingDomain-Specific Semantics
Domain-Specific Terms specifying Domain-Specific
Semantics
18
Summary

• Computable Semantic interoperability is essential
  if we are to solve the myriad of problems facing
  national healthcare delivery system revision
• Computable Semantic interoperability is difficult
• Computable Semantic interoperability is possible
  if the Four Pillars are addressed
• HL7 Version 3 and the associated RIM provide a
  framework satisfying the Four Pillars
• Oracle is building HTB, a standards-based
  interoperability value-add platform based on HL7
  v3, the RIM, and the Four Pillars