FHIR for FAIR - FHIR Implementation Guide
0.1.1 - STU 1 Ballot International flag

FHIR for FAIR - FHIR Implementation Guide, published by Health Level Seven International - SOA Work Group. This is not an authorized publication; it is the continuous build for version 0.1.1). This version is based on the current content of https://github.com/HL7/fhir-for-fair/ and changes regularly. See the Directory of published versions

FHIR for FAIR Home Page

Official URL: http://hl7.org/fhir/uv/fhir-for-fair/ImplementationGuide/hl7.fhir.uv.fhir-for-fair Version: 0.1.1
Active as of 2022-04-12 Computable Name: FhirForFairIG

Scope

This guide aims to provide guidance on how HL7 FHIR can be used for supporting FAIR health data implementation and assessment to enable a cooperative usage of the HL7 FHIR and FAIR paradigms. Other kinds of health-related artefacts, such as clinical guidelines, algorithms, software, models are out of scope.

What does FAIR mean?

FAIR stands for Findable, Accessible, Interoperable and Reusable

FAIR data - Wikipedia

Figure 1 – FAIR [SangyaPundir / CC BY-SA]

The FAIR principles - a minimal set of community-agreed guiding principles and practices - were first introduced by Wilkinson et al (2016) in their seminal paper (doi.org/10.1038/sdata.2016.18).

The authors intent was to provide guidelines to improve the FindabilityAccessibilityInteroperability, and Reuse (FAIR) of scientific data. Since their first introduction in 2016, FAIR principles were well received in international community and rapidly adopted by researchers.

The FAIR principles put specific emphasis on improving the ability of machines to automatically find and make use of the  research (and other) data, as well as to support its reuse by the human researches. Therefore, acting as a guideline for those wishing to gain much greater value from the future reuse of their scientific data and relevant metadata.

We provide more detailed introduction to the FAIR data principles as well as the relationship to the HL7 FHIR standard in later sections of this IG.

There are existing concepts for operationalization and indicator for assessment of FAIRness , for example Research Data Alliance or EOSC recommendations. We discuss their implications in a special section HL7 FHIR and RDA Indicators.

What this project is aiming to?

The FAIRness for FHIR project, this Implementation Guide is part of, is intended to be the result of an active collaboration between the RDA and HL7 communities.

This project has as main goals to:

  • Facilitate the collaboration between the FAIR and the FHIR communities

  • Enable a cooperative usage of the FHIR standard and FAIR principles.

  • Support the assessment and implementation of FAIR health data by using HL7 FHIR

What problems are FAIR and this guide trying to address?


  • Allow researchers to make available under specified conditions of use set of data, derived from a data source, that have been collected and consolidated for a specific purpose.
  • Allow researchers and data scientists to look for and access previously collected data sets to answer specific questions



Audience

The audience for this Implementation Guide includes:

  • Researchers : People who generate, process or use research health data
  • Health Data Providers : Healthcare providers that populate clinical data warehouses; Clinical study groups, operators of registries or epidemiological cohorts, surveillance or Public Health
  • Technical Implementers : Vendors of EHR systems, data repositories or EDC systems
  • Government agencies : Funding institutions that want to ensure the sustainable usability of their projects; Legislators issuing guidelines for the use of personal data (GDPR,Data Governance Act); Regulatory bodies providing best practice guidelines.
  • Public : Citizens who want to use their data for their own purposes or donate data for research purposes

Structure of this guide

This guide has been structured in the following parts:

  1. A Backgroud section describing the context of this work , including the methodology followed. This section summarizes also the FAIR data principles, the RDA indicators for assessment of FAIRness; and some general considerations about data and metadata, and Globally Unique, Persistent and Resolvable Identifiers.
  2. A Real World Cases section describing a set of real world cases used as case study for this guide.
  3. A Recommendations section, the core part of this guide, including:
    1. General recommendations
    2. Best practices for native HL7 FHIR architectures aiming at being FAIR
    3. An overview of the relationship between the FAIR Data Maturity Model described in the RDA Indicators and HL7 FHIR.
  4. A set of HL7 FHIR conformance resources and examples.

Ballot Status

This Implementation Guide will be balloted as STU with the intention to go normative in subsequent ballot cycles.

Authors and Contributors

Role Name Organization
Author Giorgio Cangioli HL7 Europe
Author Alicia Martinez-Garcia Andalusian Health Service
Author Ian Harrow Pistoia Alliance
Author Kees van Bochove The Hyve
Author Matthias Löbe IMISE University of Leipzig
Author Philip van Damme Amsterdam UMC
Contributor Anupama Gururaj NIH/NIAID
Contributor Belinda Seto NIH
Contributor Brian Alper Computable Publishing LLC
Contributor Catherine Chronaki HL7 Europe
Contributor Edward Eikman  
Contributor Olga Vovk Samvit Solutions
Contributor Oya Beyan Koeln University
Contributor Steve Tsang NIH/NIAID