Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
Doug Fridsma, MD PhDPresident & CEO, AMIA
Interoperability
Getting to Interoperability
2AMIA PowerPoint Template
• To get to interoperability (or to avoid information blocking) we need a common understanding of the problem
• Can’t solve the problem all at once• Need to break it into pieces that allow different groups to work on different parts of the
problem
• How we frame the problem will affect the solution(s) we get• Ultra-large scale system• Definition of Interoperability
It’s not architecture, it’s city planning
Software Engineering Institute Carnegie Mellon University 4500 Fifth Avenue Pittsburgh, PA 15213-3890
Phone: 412-268-5800 Fax: 412-268-5758 www.sei.cmu.edu [email protected]
© Copyright 2006 Carnegie Mellon University.
Ultra-Large-Scale SystemsThe Software Challenge of the Future
Ultra-Large-S
cale System
s: The Softw
are Challenge of the Future
It’s not architecture. It’s city planning
Interoperability (IEEE)• Ability of two or more systems to exchange information• Ability of those systems to use the information that has been exchanged
Interoperability only makes sense in the context of what you want to DO
Transitions of Care
Care sum
mary
Clinical Trial Eligibility
Public Health R
eporting
Public Health R
eporting
“Silos of Interoperability”• Define Interoperability in terms of
the ”thing you want to do”• Without coordination across
there is no shared “building blocks”
• This creates confusion• Quality measures• Data Registries• Clinical Decision Support• EHRs• Non-health data• etc
You need a framework to pull together the different use cases
8
“You can’t build a skyscraper by nailing 10,000 dog houses together”
Patient Practice Population Public
Think HORIZONTALLY rather than vertically
Informatics, standards, testing, business drivers, governance
Big Data and population health
Electronic Health Record and
Quality
Consumer empowerment
Clinical and Translational
Research
Information Blocking and the Socio-Technical Stack
Legal Responsibilities (e.g. HIPAA, 42 CFR Part 2)Public PolicyContractual Decisions (e.g. Epic App Orchard)Intellectual PropertyMarket-based Motivations (e.g. ACOs) Business DriversWhen to apply the data (e.g. lab test results)Workflow (dynamic)How to apply the data (e.g. Admission v. Discharge Summary)Context (static)Purpose-specific APIs and services that leverage the other four layersServicesTerminologies, Structured data, coded (e.g. ICD-10, SNOMED)SemanticMessage formatting (e.g. CCDA v2)SyntacticHow the message move from A to BTransportHow we ensure that messages are secure and privateSecurity
Traditional technology stack
Implementation
Policy and Business drivers
API
Information Blocking and the Socio-Technical Stack
Legal Responsibilities (e.g. HIPAA, 42 CFR Part 2)Public PolicyContractual Decisions (e.g. Epic App Orchard)Intellectual PropertyMarket-based Motivations (e.g. ACOs) Business DriversWhen to apply the data (e.g. lab test results)Workflow (dynamic)How to apply the data (e.g. Admission v. Discharge Summary)Context (static)Purpose-specific APIs and services that leverage the other four layersServicesTerminologies, Structured data, coded (e.g. ICD-10, SNOMED)SemanticMessage formatting (e.g. CCDA v2)SyntacticHow the message move from A to BTransportHow we ensure that messages are secure and privateSecurity
Traditional technology stack
Implementation
Policy and Business drivers
API
Information Blocking and the Socio-Technical Stack
Legal Responsibilities (e.g. HIPAA, 42 CFR Part 2)Public PolicyContractual Decisions (e.g. Epic App Orchard)Intellectual PropertyMarket-based Motivations (e.g. ACOs) Business DriversWhen to apply the data (e.g. lab test results)Workflow (dynamic)How to apply the data (e.g. Admission v. Discharge Summary)Context (static)Purpose-specific APIs and services that leverage the other four layersServicesTerminologies, Structured data, coded (e.g. ICD-10, SNOMED)SemanticMessage formatting (e.g. CCDA v2)SyntacticHow the message move from A to BTransportHow we ensure that messages are secure and privateSecurity
Traditional technology stack
Implementation
Policy and Business drivers
API
Information Blocking and the Socio-Technical Stack
Legal Responsibilities (e.g. HIPAA, 42 CFR Part 2)Public PolicyContractual Decisions (e.g. Epic App Orchard)Intellectual PropertyMarket-based Motivations (e.g. ACOs) Business DriversWhen to apply the data (e.g. lab test results)Workflow (dynamic)How to apply the data (e.g. Admission v. Discharge Summary)Context (static)Purpose-specific APIs and services that leverage the other four layersServicesTerminologies, Structured data, coded (e.g. ICD-10, SNOMED)SemanticMessage formatting (e.g. CCDA v2)SyntacticHow the message move from A to BTransportHow we ensure that messages are secure and privateSecurity
Traditional technology stack
Implementation
Policy and Business drivers
API
Information Blocking and the Socio-Technical Stack
Legal Responsibilities (e.g. HIPAA, 42 CFR Part 2)Public PolicyContractual Decisions (e.g. Epic App Orchard)Intellectual PropertyMarket-based Motivations (e.g. ACOs) Business DriversWhen to apply the data (e.g. lab test results)Workflow (dynamic)How to apply the data (e.g. Admission v. Discharge Summary)Context (static)Purpose-specific APIs and services that leverage the other four layersServicesTerminologies, Structured data, coded (e.g. ICD-10, SNOMED)SemanticMessage formatting (e.g. CCDA v2)SyntacticHow the message move from A to BTransportHow we ensure that messages are secure and privateSecurity
Traditional technology stack
Implementation
Policy and Business drivers
APITransitions of C
are
Care sum
mary
Clinical Trial Eligibility
Public Health R
eporting
Public Health R
eporting
SO HOW DO WE GET THERE?
AMIA PowerPoint Template 15
THE PATH OF LEAST REGRET: THREE THINGS WE NEED
BUILD DOCUMENTS FROM DATA, NOT THE OTHER WAY AROUND
What is needed: A common format for granular data• Common data formats for
• Text data• Categorical data• Numerical data• More…
• This is the beginning of the “horizontal coordination” across different use cases
• Quality Measures• Clinical Decision Support• Registries
FULL EXPORT OF THE PATIENT RECORD (AND NARRATIVE)
Restore the importance of the narrative and unstructured textPatients are more than a collection of discrete data
• Disease vs. illness• The importance of the narrative to understanding patients
The unstructured data is where discoveries are made• Precision medicine
Restore the balance of power for access to data• Empower patients, researchers, public health officers
TEST BOTH SIDES OF EXCHANGE
Postel’s principleWhen you send, conform to the standard
• Pick a set of options and make sure it conforms to a valid instance of the standard
When you receive, accept ANY version that conforms to the standard• Any variation that is valid
Create the conditions of interoperability when you create the standard, rather than when you implement it
AMIA
@AMIAinformatics
Official Group of AMIA
AMIA informatics
www.amia.org
#WhyInformatics
Doug Fridsma MD PhD
23
THANK [email protected]
AMIA
@AMIAinformatics
Official Group of AMIA
AMIA informatics
www.amia.org
#WhyInformatics
AMIA is the professional home for more than 5,400 informatics professionals, representing frontline clinicians, researchers, public health experts and educators who bring meaning to data, manage information and generate new knowledge across the research and healthcare enterprise.
24AMIA 2017 | amia.org