Skip to content

Why FHIR

Fast Health Interoperability Resources (FHIR) is the future of healthcare information and one of the core standards of LifeOmic's Precision Health Cloud (PHC). It provides an interoperability specification for the efficient exchange of healthcare information. The goal of the standard is to make a patient's electronic health records available, discoverable, and understandable. For electronic health records to be shared among different systems and to support data processing and automate clinical and diagnostic information, healthcare information needs to become data with a common structure. The FHIR standard provides that structure. It defines data elements and formats and provides an API for data exchange.

FHIR is unique because it has been designed over many years by a large group of health data professionals to be a single data format that can encompass all of the health and medical data about a person. FHIR provides a standard for coding systems and data representations that allows you to harmonize and use data from disparate sources.

Collaborative and Open Source Standard by HL7

The FHIR standard is an open standard developed by the non-profit Health Level Seven International (HL7). HL7 is a standards organization that uses a collaborative approach and a balloting system to continue the evolution of FHIR. HL7 published the most recent FHIR Release, 4.0.1, in October 2019. The standard is free to implement and many open source FHIR components and implementations are available.

Broad Support

Earlier implementations of healthcare information standards, such as HL7 Version 2 and Version 3, failed to gain momentum with vendors and the medical community. FHIR built on lessons learned from these standards and has gained an unprecedented level of adoption and a growing momentum. It has support from the Federal government, the major technology companies, medical software companies, and the leading medical research institutions.

21st Century Cures Act

One of the biggest drivers for the adoption of FHIR is the 21st Century Cures Act (Cures Act), signed into law by Congress in 2016. This law is designed to help accelerate medical product development and bring new innovations and advances to patients who need them faster and more efficiently. Specific to FHIR, the Cures Act Final Rule, supports seamless and secure access, exchange, and use of electronic health information. The U.S. Department of Health and Human Service rules for implementation of the 21st Century Cures Act specifically mandated the use of FHIR to support seamless and secure access, exchange, and use of electronic health information. Health IT developers and Electronic Health Record (EHR) vendors are required to achieve the Office of the National Coordinator for Health Information Technology (ONC) Health IT Certification as part of the interoperability rule. This certification includes real world testing of interoperability.

The Federal Government's mandate to use FHIR continues to expand. In 2020, the Centers for Medicare & Medicaid Services (CMS), delivered the Interoperability and Patient Access final rule. This rule requires the use of FHIR by a variety of CMS-regulated payers, including Medicare Advantage, organizations, state Medicaid programs, and qualified health plans in the Federally Facilitated Marketplace. These payers are required to implement and maintain a secure, standards-based (HL7 FHIR Release 4.0.1) API that allows patients to easily access their claims and encounter information. Another example of the government's adoption of FHIR is the Centers for Disease Control and Prevention (CDC)'s eCR Now FHIR app. This app uses FHIR to help EHR systems report COVID-19 cases to state and local public health authorities.

Support from Big Tech

Major technology companies have also widely embraced the adoption of FHIR. During a White House meeting in 2018, Amazon, Microsoft, Google, IBM, Oracle, and Salesforce signed a joint commitment to remove barriers for the adoption of technologies for healthcare interoperability. In their statement, they cite the need for the industry to converge and embrace emerging standards for healthcare data interoperability, such as HL7 FHIR and the Argonaut Project.

An even stronger sign of support than the companies' joint statements are the FHIR-related products and solutions developed by Apple, Google, Amazon, and Microsoft. Apple has pushed hard into the healthcare space with apps developed for healthcare professionals and patients and support for developers of these apps. The use of FHIR is highlighted by the popular Apple Health app's Health Records feature. This feature lets patients use a user-friendly app on their iPhone or iPad to access their health records from multiple institutions. Apple notes that communication between a patient's Electronic Health Records (EHR) and the Health app uses FHIR APIs defined by the Argonaut Project. The connection also uses OAuth 2.0, the authentication standard for FHIR. None of the health data is seen by Apple, as the data flows on an authenticated connection between the app and the EHR APIs.

Google supports FHIR with the Google Cloud Healthcare API. The Cloud Healthcare API provides a fully managed, highly scalable, enterprise-grade development environment for building clinical and analytics solutions securely in Google Cloud. According to Google, the solution allows you to store, manage and gain actionable insights on your data in FHIR format. Google and the Mayo Clinic have a 10-year strategic partnership to secure and store Mayo Clinic's patient data in Google Cloud.

Amazon AWS offers FHIR Works on AWS. This solution is an open source software toolkit that can help software developers integrate FHIR standard APIs into their own products or build connectors between legacy interfaces and the FHIR standard.

Microsoft's Azure features the Azure API for FHIR. This is a managed, standards-based, compliant API for clinical health data that enables solutions for actionable analytics and machine learning.

Support from Health Care Research and Industry

In addition to the major technology companies, leading research and healthcare institutions support FHIR. This widespread support is shown in the HL7 Argonaut Projected referred to by Apple and the joint statement. The Argonaut Project is a private sector initiative to advance industry adoption of modern, open interoperability standards. The purpose of the Argonaut Project is to rapidly develop a first-generation FHIR-based API and Core Data Services specification to enable expanded information sharing for electronic health records and other health information technology based on Internet standards and architectural patterns and styles. The list of Argonaut Project Sponsors includes major technology companies, health care information companies, and leading medical research institutions.

Argonaut Project Sponsors
Accenture Allscripts Apple athenahealth Beth Israel Lahey Health Cerner
Change Healthcare eClinicalWorks Epic Humana Intermountain Healthcare Mayo Clinic
MEDITECH Microsoft Optum Partners HealthCare System SMART at the Boston Children’s Hospital Surescripts

Healthy FHIR API Ecosystem

APIs provide the standards and protocols for communication between modern information systems, and FHIR APIS have flourished. Cerner has implemented FHIR in its Ignite APIs for Millenium. Epic is also a strong supporter of FHIR and makes its HL7 FHIR standard APIs publicly available with strong developer support. And many health plans are now required by CMS and ONC rules to have FHIR-compliant public patient access APIs and provider directory APIs.

Another important API that builds on the FHIR standard is the SMART on FHIR API, also an HL7 standard, run out of the Boston Children’s Hospital Computational Health Informatics Program. SMART on FHIR integrates apps with FHIR servers and electronic medical records systems that have FHIR interfaces. According to SMART, the SMART on FHIR API has been built into the major EHR products, has been used by Apple to connect its health app to hundreds of healthcare systems, and has been used for app launch on the Microsoft Azure product. Countless other app and platform developers, including Google, Amazon, and LifeOmic use it as well.

Supports Rapid Development

One of the biggest drivers for FHIR adoption is that the standard is easier for software developers to implement than earlier standards and uses the technology of the modern web. A wide variety of implementations and developer resources are available for FHIR including open source. This allows developers of new healthcare software to adopt a widely adopted standard rather than develop their own proprietary system.

HL7 lists the following benefits for developers:

  • A strong focus on implementation: fast and easy to implement

  • Multiple implementation libraries, many examples available to kick-start development

  • Specification is free for use with no restrictions

  • Interoperability out-of-the-box: base resources can be used as is, but can also be adapted as needed

  • Evolutionary development path from HL7 Version 2 and CDA: standards can co-exist and leverage each other

  • Strong foundation in Web standards: XML, JSON, HTTP, OAuth, etc.

  • Support for RESTful architectures, seamless exchange of information using messages or documents, and service-based architectures

  • Concise and easily understood specifications

  • A human-readable serialization format for ease of use by developers

  • Ontology-based analysis with formal mapping for correctness is under development by HL7

How Does PHC Use FHIR

PHC supports STU 3 (FHIR 3) and is working towards full STU 4 (FHIR 4) support. PHC also uses the SMART on FHIR API.

The PHC use of FHIR allows you to store any type of health or medical data from any source without losing fidelity, such as dropping data elements that are unique to a single source,while at the same time doing analysis and visualization on common data elements across sources. More traditional systems lock you into a rigid data schema that you design up front. This rigid data scheme doesn't allow you to represent new data elements from other data sources or satisfy new requirements as your needs change.

LifeOmic has developed powerful and reusable analysis and visualization tools that can operate on FHIR resources of any type and shape, so you can answer questions about your data regardless of the data source. PHC leverages many data shapes with FHIR, including:

  • Clinical time-series data points: examples are diagnoses, blood pressure, heart-rate measurements, blood glucose, medications, or scheduled visits
  • Genomic Variants, Gene expression, Proteomics, Pharmacogenetics: file formats such as VCF, BAM, CSV, TSV
  • Documents, Images, and Audio: file formats such as JPEG, PNG, DICOM, or PDF

PHC FHIR for Clinical Data

FHIR is designed for clinical data. A specific example is HER2 tests for breast cancer. HER2 status is represented as either "positive" or "negative" depending on whether the cancer over expresses the HER2 protein or not, and this determines the treatment course and likely outcome of the cancer. But in reality there are two tests used to determine HER2 status: IHC and FISH. IHC produces a HER2 score that is a number than ranges from 0 to 3+, where 0 to 1 means HER2 negative, 1 to 2 means borderline, and over 2 means HER2 positive. For IHC HER2 borderline tumors, a FISH test is usually performed which directly measures protein expression by counting cells and looking at which ones fluoresce due to bound probes.

In FHIR, both test results can be represented by an Observation resource with multiple components. For IHC the components will be a HER2 status value (positive, negative, or borderline) and a HER2 score. For FISH, the components will be a HER2 status value (positive or negative), a number of cells counted, a number of cells that fluoresced, and a ratio. Because both types of Observation resources have a HER2 status component, a researcher can search for or aggregate subjects by HER2 status regardless of which test they had. They can do this while analyzing specific details unique to each test type when needed, for example the number of cells counted for FISH tests across a cohort.

The structural flexibility of FHIR compares favorably to the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM). FHIR is expressive and supports a large number of resource types. This allows you to work with a wide variety of data types. OMOP CDM is a flat data table and not as extensible as FHIR.

FHIR Resource Example

Click the example below to see a FHIR resource in JSON format. The sample is a pharmacogenomic Observation resource with multiple components.

FHIR Resource Example

{
  "resourceType": "Observation",
  "meta": {
    "profile": ["https://www.hl7.org/fhir/genomics.html#observation-genetics"],
    "tag": [
      {
        "system": "http://lifeomic.com/fhir/dataset",
        "code": "19e34782-91c4-4143-aaee-2ba81ed0b206"
      }
    ]
  },
  "code": {
    "coding": [
      {
        "system": "http://loinc.org",
        "code": "53040-2",
        "display": "Pharmacogenomic variation's effect on drug metabolism"
      }
    ],
    "text": "FCAMR"
  },
  "subject": {
    "reference": "Patient/8cb82aa0-7f2c-4fdb-bf91-0ed1b315392c"
  },
  "issued": "2017-07-20",
  "valueCodeableConcept": {
    "coding": [
      {
        "system": "http://loinc.org",
        "code": "LA10315-2",
        "display": "Ultrarapid metabolizer"
      }
    ],
    "text": "Ultrarapid metabolizer"
  },
  "component": [
    {
      "code": {
        "coding": [
          {
            "system": "http://loinc.org",
            "code": "51971-0",
            "display": "Drug metabolism analysis overall interpretation Qualitative"
          }
        ]
      },
      "valueCodeableConcept": {
        "coding": [
          {
            "system": "http://loinc.org",
            "code": "LA10315-2",
            "display": "Ultrarapid metabolizer"
          }
        ],
        "text": "Ultrarapid metabolizer"
      }
    }, 
    {
      "code": {
        "coding": [
          {
            "system": "http://snomed.info/sct",
            "code": "60132005",
            "display": "Gene"
          }
        ]
      },
      "valueString": "FCAMR"
    }, 
    {
      "code": {
        "coding": [
          {
            "system": "http://snomed.info/sct",
            "code": "9109004",
            "display": "Genetic allele 1"
          }
        ]
      },
      "valueString": "T"
    }, 
    {
      "code": {
        "coding": [
          {
            "system": "http://snomed.info/sct",
            "code": "9109004",
            "display": "Genetic allele 2"
          }
        ]
      },
      "valueString": "T"
    }
  ],
  "id": "56b63548-0cfc-46de-8345-daecc0c57c2c"
}

LifeOmic FHIR Web Service

LifeOmic FHIR Service is a managed service that makes it easy to store and exchange health care data and can easily scale to hundreds of millions of FHIR records. The service provides a comprehensive RESTful API and includes advanced searching and advanced analytics using technologies like Elasticsearch, EMR, and Neptune (for Ontologies). The FHIR Service also offers SQL capabilities and audit trail (data-versioning) support.

PHC FHIR Design Elements

FHIR elements are interwoven throughout the design and operation of PHC. When you use the PHC subject search or patient viewer, you are searching and viewing FHIR data. A FHIR resource is the basic building block of FHIR and used to exchange and store data. The data on the Subjects page is populated by FHIR resource types and this FHIR data is retrieved by an API call to the FHIR API in the FHIR service. The search filters on the left side of the page, such as Medications, Conditions, and Observations, are Resource types defined in the FHIR standard. FHIR Subject


Last update: 2021-02-11