1

Data Access Framework (DAF)The Use of DAF for Clinical Research

July 21, 2015S&I Initiative Coordinator: John Feikema/Johnathan ColemanHHS/ONC Sponsor: Mera Choi

2

1. Background and Overview2. DAF Phase 1 & 2: Artifacts and PCORnet Applicability3. DAF Phase 3: PCORnet Collaboration 4. Questions5. Additional Resources

Agenda

3

Background and Overview

4

• In 2011, the Office of the National Coordinator for Health Information Technology (ONC), an office in HHS, launched the “Query Health Initiative.” The Query Health (QH) Initiative enabled providers and researchers access to aggregated data from multiple data sources (known as distributed queries)– During QH, multiple pilots were conducted and multiple standards for queries and results

were developed. The initiative later concluded in 2013 and was succeeded by the Data Access Framework (DAF)

• Lessons Learned from QH:– Barriers exist to extracting data from clinical data sources (EMRs, Warehouses, etc.)– Data Mapping challenges (specifically around vocabularies, semantic meanings, etc.)– The need to simplify standards for Query and Query Results – The need to create modular standards for Transport, Security, Query Structure, Query

Results and Information Models so they could be replaced as better approaches evolve

• In July of 2013, the Data Access Framework (DAF) was launched to address these challenges and also help enable data access for multiple use cases, which include: – Enabling data access within a organization (Local DAF) – Phase 1– Enabling data access from a single external organization (Targeted DAF) – Phase 2– Enabling data access from multiple organizations (Federated DAF) – Phase 3

Data Access Framework:

Background

5

Data AccessFramework

Local Access viaIntra-Organization Query

(Phase 1)

Targeted Access viaInter-Organization Query

(Phase 2)

Federated (Distributed) Access Across Multiple Organizations

(Phase 3)

Standards based approach to enable information access at all levels: Local, Targeted, and Federated (Distributed)

• Create and disseminate queries internal to organization

• Query Structure Layer • APIs for Data Access• Authentication/Authorization Layer

• Receive standardized responses• Query Results Layer

• Create and disseminate queries to single external Organization

• Query Structure Layer • Transport Layer• Authentication/Authorization Layer

• Receive standardized responses from external orgs• Query Results Layer

• Create and disseminate queries to multiple orgsgoverned by a network

• Receive aggregated or de-identified responses

• Focus on Information Model for the network and leverage standards from earlier phases.

DataSource(Org2)

DataSource(Org3)

DataSource(Org1)

Query Request

Query Response

X Hospital System X Hospital System

Y Hospital System

Data Access Framework:

Overview and Phases

6

DAF Phase 1 & 2: Artifacts and PCORnet Applicability

7

Data Access Framework:

Phase 1 & 2 Overview

Basic Transport Protocols

Application Transport Protocols

Query Structure

Query Vocabularies and Value Sets

Authentication/Authorization

Result Structure

Result Vocabularies and Value Sets

Information Models

Transport Layer

Security Layer

Query Structure

Query Results

Data Model to support queries

Data Access Framework Building Blocks

DAF Scope: Develop Implementation Guides (IG) using standards/profiles for specific use cases

• Local (Intra-Organization) access of clinical information from one or more systems

• Targeted (Inter-Organization) access of clinical information from a single known external organization

• Use cases deal with accessing both data elements (e.g. problems, medications, procedures) and documents (e.g. C-CDA, C32)

DAF Out-of-Scope• Trust establishment between

organizations for Targeted Query• Discovery of End Points for Targeted Query • Patient Matching rules and algorithms that

organizations may implement • Policies that allow/dis-allow disclosure of

patient data

8

Data Access Framework:

Phase 1 & 2 Query Stacks

Basic Transport Protocols

Application Transport Protocols

Query Structure

Query Vocabularies and Value Sets

Authentication/Authorization

Result Structure

Result Vocabularies and Value Sets

Information Models

Transport Layer

Security Layer

Query Structure

Query Results

Data Model to support queries

Data Access Framework Building Blocks

HTTP

TLS

OAuth2 profiles*

FHIR

FHIR + MU2 vocabularies

FHIR Resources

FHIR + MU2 vocabularies

FHIR

RESTful FHIR based Query Stack for Data Element based access

HTTP

TLS

OAuth2 profiles*

IHE MHD v2 FHIR based Profile

FHIR + XDS Metadata

FHIR Resources + CCDA

FHIR + MU2 (CCDA vocabularies)

FHIR + XDS Metadata

RESTful FHIR based Query Stack for Document Metadata based access

HTTP

Mutual TLS

IHE XUA (SAML)

XCA

XDS Metadata

CCDA

MU2 (CCDA vocabularies)

ebRIM + XDS Metadata

SOAP 1.2

SOAP Query Stack forDocument Metadata based access

DAF Data Element IG DAF Document IG

* - OAuth2 profiles still evolving

9

• Argonaut– Project led by industry in collaboration with HL7, with goals that include the following:

• Development of CCDA to FHIR Mappings• Security Guidance for FHIR • Accelerate FHIR Implementation and DSTU publication• http://hl7.org/implement/standards/FHIR-Develop/argonauts.html

Relationship: DAF Data Element IG is based on FHIR profiles and is being reused by Argonaut

• HEART– An MIT project supported by ONC as a Committed Member– The goal is to harmonize and develop set of privacy and security specifications/profiles to

enable individuals to control access to RESTful health related resources using OAuth, OpenId, UMA

– Develop open source reference implementations for the developed specifications/profileshttp://openid.net/wg/heart/charter/

Relationship: DAF Security layer could potentially reuse profiles/specifications developed by HEART

Data Access Framework:

Other Related Projects

10

• Can address the challenge of building specific APIs (ETLs) required at each data source – DAF HL7 FHIR profiles being reused by Argonaut project

Vendor 1 DB

Vendor 2 DB

Vendor 3 DB

PCOR Data Mart

(Problems, Meds,

Procedures…)

DAF API

DAF API

DAF API

Vendor 1 DB

Vendor 2 DB

Vendor 3 DB

PCOR Data Mart

(Problems, Meds,

Procedures…)

Specific API’s for each data source

API - 1

API - 2

API - 3

Data Access Framework:

Phase 1 & 2 Artifacts & PCORnet Applicability

11

• Can address the challenge of accessing data from a known external data source over the internet

Requestor (Systems or

Researchers, etc.)

Responder

Responder

Responder

• NOTE: Each Requestor/Responder interaction is unique

Data Access Framework:

Phase 1 & 2 Artifacts & PCORnet Applicability CONTINUED

12

• PCORnet could leverage the data structures and logical models developed by DAF• Models very well aligned with PCORnet CDM • Vocabularies and Value Sets aligned with national

standards for interoperability from ONC

• Open Source Reference Implementations available for HL7 FHIR APIs

• All DAF Phase 1 & 2 artifacts and technical support available for PCORnet pilot opportunities

Data Access Framework:

Phase 1 & 2 Artifacts & PCORnet Applicability CONTINUED

13

DAF Phase 3 and PCORnet Collaboration

14

• By leveraging work performed in DAF Phase 1 & 2, DAF Phase 3 aims to address PCORnet challenges

• Examples include:• Enabling access to data for researchers using Distributed Queries

• Enhancing existing data models (standards) to include claims and other data sources for queries

• Enhancing existing standards for aggregate, de-identified data collection

• Developing a policy framework that enables Distributed Queries

• Enabling patient controlled data release using internet standards such as OAuth2, OpenID Connect, and UMA

• Data quality improvement (e.g. Patient Matching)

Data Access Framework:

Phase 3 Overview

15

• DAF Phase 3 would request collaboration with PCORnet members to:

⁻ Identify challenges that could be addressed by ONC DAF Phase 3

⁻ Pilot solutions to challenges that are identified ⁻ DAF support team will provide technical support where

appropriate

⁻ ONC will use the solutions to inform/guide the standards created

⁻ Help ONC address any policy issues that require additional guidance

⁻ Provide expertise and guidance for the development of DAF Phase 3 artifacts

Data Access Framework:

Phase 3 PCORnet Collaboration

16

• Please sign-up for the DAF initiative to ensure you stay up-to-date with the latest information• Complete the “Join” form on

the DAF Wiki Page: http://wiki.siframework.org/Data+Access+Framework+Join+the+Initiative

• Join us as we begin the process of identifying the problem statement and identifying the technical/policy issues that need to be addressed by Phase 3 pilots

• Consider piloting DAF Phase 1, 2 or 3 artifacts

Data Access Framework:

How You Can Join DAF

17

Questions

18

Additional Resources

Useful Links• DAF Homepage

– http://wiki.siframework.org/Data+Access+Framework+Homepage• DAF Initiative Signup

– http://wiki.siframework.org/Data+Access+Framework+Join+the+Initiative• DAF Charter

– http://wiki.siframework.org/Data+Access+Framework+Charter+and+Members• DAF Standards, Harmonization and Implementation Activities

– http://wiki.siframework.org/DAF+Standards+Harmonization+and+Implementation

• DAF HL7 FHIR IG– http://hl7.org/fhir/2015May/daf.html

• DAF IHE White Paper– http://

ihe.net/uploadedFiles/Documents/PCC/IHE_PCC_White_Paper_DAF_Rev1.1_2014-10-24.pdf

19

Initiative Support Leads• For questions, please feel free to contact your support leads:

– Initiative Coordinator: John Feikema john.feikema@siframework.org /Johnathan Coleman jc@securityrs.com

– ONC Sponsors: Mera Choi mera.choi@hhs.gov– Support Team:

• Project Management: – Jamie Parker jamie.parker@esacinc.com – Gayathri Jayawardena gayathri.jayawardena@esacinc.com

• Technical Support: – Dragon (Nagesh) Bashyam nagesh.bashyam@drajer.com

• Standards SME:– Brett Marquard brett@riverrockassociates.com – Ed Larsen ed_larsen@comcast.net

• Privacy and Security SME:• Glen Marshall gfm@securityrs.com

• Standards Support: – Angelique Cortez angelique.j.cortez@accenture.com

20