Difference between revisions of "DAF Phase3 FHIR IG Proposal"
Line 84: | Line 84: | ||
<!-- Provide a listing of the types of scenarios to be represented in the examples produced for this Profile. They should demonstrate the full scope of the Profile and allow exercising of the Profiles capabilities (full element coverage, inclusion & omission of optional elements, repeating and singleton repeating elements, etc.) --> | <!-- Provide a listing of the types of scenarios to be represented in the examples produced for this Profile. They should demonstrate the full scope of the Profile and allow exercising of the Profiles capabilities (full element coverage, inclusion & omission of optional elements, repeating and singleton repeating elements, etc.) --> | ||
+ | |||
+ | * A PCORNet site would like to extract data from multiple different data sources supported by different EMR vendors to populate a data mart that will be used for PCORNet research studies. | ||
+ | * A researcher would like to retrieve the characteristics of a data mart to determine the total number of patients represented in the data mart, the platforms that are supported for queries etc. | ||
+ | * A researcher would like to submit a query to determine the total number of patients with diabetes within the age group of 20 and 70 stratified by race, age groups, postal codes. | ||
===Scope of coverage=== | ===Scope of coverage=== | ||
− | |||
<!-- Define the full scope of coverage for the Profile. The scope must be clearly delineated such that it does not overlap with any other existing or expected Profile. The scope will be used to govern "what is the set of potential applications to consider when evaluating what elements are 'core' – i.e. in the 80%" | <!-- Define the full scope of coverage for the Profile. The scope must be clearly delineated such that it does not overlap with any other existing or expected Profile. The scope will be used to govern "what is the set of potential applications to consider when evaluating what elements are 'core' – i.e. in the 80%" | ||
Line 97: | Line 100: | ||
As a rule, Profiles should encompass all of these aspects. | As a rule, Profiles should encompass all of these aspects. | ||
--> | --> | ||
+ | The Scope of the IG is as defined in the previous sections. | ||
===Realm=== | ===Realm=== |
Revision as of 21:37, 22 May 2016
Contents
DAF Research Data Access IG
This DAF implementation guide will specify the necessary conformance requirements, profiles, extensions and vocabularies necessary to enable researchers to access data from one or more data sources. This implementation guide will complement the existing DAF FHIR IG. While the existing DAF FHIR IG is expected to be used to enable use cases which deal with a single patient, the DAF Research Data Access IG will deal with extracting data for multiple patients from data sources such as EMRs, local databases.
Implementation Guide balloting plans
The balloting of the DAF Research Data Access IG will follow the FHIR STU3 timelines. The DAF milestones are listed below
- Submit content for ballot - September 2016
- Complete ballot Reconciliation - November 2016
- Submit publication Request - December 2016
Implementation Guide Details
The DAF Initiative which is sponsored by the Office of the National Coordinator (ONC) is currently piloting with PCORNet (Patient Centered Outcome Research Network) sites to enable data access for researchers using FHIR constructs. As part of the initiative the following capabilities are being developed
- Capability to use FHIR resources and profiles to extract data from data sources for multiple patients
- Capability to populate data marts from multiple data sources to make the data available to researchers
- Capability that defines metadata about data marts that will be useful for researchers to formulate hypothesis, trial concepts, define queries
- Capability that allows researchers to submit different types of queries (FHIR Search based queries, SASS queries, SQL queries, Other proprietary queries)
- Capability that allows data marts to send back aggregate data to researcher based on queries
- Capability that allows data marts to send back patient level data to researchers based on queries
- Capability to use FHIR Search/Query mechanisms to build cohorts
Each of these capabilities will be implemented using PCORNet sites and the implementation guide content will be based on the feedback obtained from these implementations.
Profiled Resource(s)
The following are resources that are being considered for defining the data extraction, metadata representation, query packaging, query submission and retrieving query results.
- Bundle: http://hl7-fhir.github.io/bundle.html
- Task : http://hl7-fhir.github.io/task.html
- Operation Definition: http://hl7-fhir.github.io/operationdefinition.html
- Conformance: http://hl7-fhir.github.io/conformance.html
- Consent: http://hl7-fhir.github.io/consent.html
- Parameters: http://hl7-fhir.github.io/parameters.html
- Observation: http://hl7-fhir.github.io/observation.html
- MeasureReport: http://hl7-fhir.github.io/measurereport.html
Constraints to be Applied
- The IG will define how to instantiate a data extraction task for extracting bulk data for research purposes.
- The IG will define how to characterize a data mart that can be used by researchers to formulate queries.
- The IG will define how to use Task, Operation Definition and Parameters to submit queries.
- The IG will define how to use Observations or Measure Reports to retrieve aggregate data.
- The IG will define how to use the existing DAF IG patient level profiles.
- The IG will provide data mapping between FHIR and PCORNet CDM.
Extensions to be Applied
- Extensions will be needed to characterize data mart meta data.
- Extensions may also be needed for the other resources, they will be based on the actual implementation needs.
Example Scenarios
- A PCORNet site would like to extract data from multiple different data sources supported by different EMR vendors to populate a data mart that will be used for PCORNet research studies.
- A researcher would like to retrieve the characteristics of a data mart to determine the total number of patients represented in the data mart, the platforms that are supported for queries etc.
- A researcher would like to submit a query to determine the total number of patients with diabetes within the age group of 20 and 70 stratified by race, age groups, postal codes.
Scope of coverage
The Scope of the IG is as defined in the previous sections.
Realm
It is a US Realm IG
Ownership
Owning committee name
Contributing or Reviewing Work Groups
- Infrastructure and Messaging
- RCRIM
Expected implementations
- PCORNet CDRN REACHNet
- PCORNet CDRN pSCANNER
gForge Users
- Nagesh Bashyam
- Brett Marquard
FHIR Profile Development Project Insight ID
Plans
Timelines
- Will be following FHIR STU 3 timelines, along with meeting the timelines from FHIR Infrastructure WG.
Balloting Requirements
Choose one:
- Ballot with next FHIR STU
Desired Ballot Date
- September 2016