A Primer on FHIR: Lightweight, Reusable Web Technologies Can Help Solve Substantial Real-World Health Challenges
This blog post is the second in a series outlining what Fast Healthcare Interoperability Resources (FHIR) are, how FHIR is being used at the Department of Health and Human Services, and how you can get involved. If you haven't already, check out the two-day FHIR code-a-thon happening April 1-2, 2016 in Washington DC. At the code-a-thon, developers from Federal and State-based organizations will have the opportunity to collaborate with entrepreneurs, startups and research organizations to build cool solutions that leverage the HL7 FHIR API.
What's the Big Idea?
Not long ago, health records were paper files locked away in cabinets. Thanks to advancements in technology and national incentives, these records are largely digitized and their roles have evolved beyond a means of documenting care. Now, people look to data captured in electronic health records (EHRs) to create a deeper understanding of wellness and diseases, identify threats to public health, and determine what interventions work best for whom. The data are also used to guide health-related policy decisions and track cost-effectiveness outcomes in real-world settings. Fast Healthcare Interoperability Resources (FHIR), or "fire" as it's pronounced, is critical to this paradigm shift. FHIR is a framework being developed by Health Level Seven International (HL7), which is an organization that sets global standards that spell out rules and patterns to improve the exchange, integration, sharing, and retrieval of health information. In the same way that the iPod evolved from records, 8-tracks, cassette tapes, and compact discs, FHIR evolved from and preserved the best features of HL7's earlier standards (V2, V3, and CDA). Importantly, however, FHIR offers many improvements over existing standards, most of which stem from FHIR's use of modern web technologies that are implementer friendly and already widely adopted while ensuring privacy and security. In doing so, FHIR makes health data more accessible and easier to integrate with existing systems and tools, which helps kick-start innovative solutions to real-world health challenges and make them far easier to deploy. FHIR serves many purposes, but it's useful to think of it both as a "standard" as well as an "architecture." Within the standards world, people form workgroups to determine what data elements are important to collect and share and how those elements should be structured and grouped together in messages, documents, or other purposes. Instead of trying to structure everything, FHIR focuses on the most common use cases representing roughly 80-percent of clinical practice. As an architecture, FHIR uses application programmable interfaces (APIs), among other techniques, to provide health information where it's needed in the form it's needed. Think of APIs as a chuck of software code that contains a set of instructions, which allows machines to "talk to" other machines behind the scenes, ultimately making data available and usable. APIs provide portals between electronic systems, even legacy systems, so that enhancements can be made and interoperability can take shape incrementally. FHIR is designed to be flexible and support multiple use cases by breaking complicated health concepts into modular components, called resources, which are easy to develop and implement in real-world settings. FHIR resources are open to innovation and are built to interact, similar to the way Legos can be combined in multiple ways to create whatever you imagine. Each resource consists of three main parts: the data, a human-readable description of the data, and extensions that allow resources to be adapted and customized for specific use cases not anticipated by the basic specification.
How Does FHIR Promote Innovation?
FHIR is laying a framework for digital disruption to occur. A big part of FHIR's popularity is that it's vendor-neutral and free to use, which allows innovators to do things that couldn't easily be done before. For example, before FHIR, health IT consisted of closed, siloed systems where the existing legal, technical, and management agreements complicated data sharing. With FHIR, several electronic health record vendors are participating in the Argonaut project and are agreeing to map to common interfaces. In doing so, they're building an ecosystem, upon which very innovative things are starting to emerge. The SMART on FHIR app platform and app gallery are great examples. Think of SMART on FHIR like the app store on a smartphone. Some of the apps are designed for physicians to use, such as the Growth Chart app developed by Boston's Children's Hospital. The app plots a child's height and weight against growth charts published by the World Health Organization and U.S. Centers for Disease Control and Prevention (CDC) so that physicians can track a child's growth over time and communicate this to the child's caregivers. Other apps in the SMART on FHIR gallery are patient-facing, such as the ClinDat application, which makes it easier for rheumatoid arthritis patients to document which joints are normal, tender, or swollen. These data are captured electronically and sent back to the medical record in real-time to support the clinical care patients receive. The beauty of SMART on FHIR is the apps are vendor neutral and can be 'plugged-in' to EHRs and other tools used on multiple devices (particularly mobile devices) that are already integrated into clinicians' workflows. Already, we're witnessing how FHIR can help HHS agencies solve real health data challenges. As Mark Scrimshire discussed in a previous blogpost, CMS is using FHIR to modernize Blue Button. Also, in my role as an Entrepreneur-in-Residence at CDC, I'm working with a multidisciplinary team that's exploring how FHIR can help improve the accuracy of information collected on death certificates. As I've written previously, mortality data are one of the most comprehensive sources of health information at the population level. However, in certifying causes of death, several kinds of errors or oversights are frequently made, often due to physicians not understanding how to complete the death certificate and not being able to readily use information in the medical record. Efforts to improve the quality of cause-of-death data must strike an appropriate balance between feasibility and complexity. To address these challenges, we at CDC are working with Georgia Tech on a proof-of-concept project that uses FHIR and advanced computing to identify critical information from the descendant's health record and notify the physician of conditions that were most likely associated with the decedent's death. We chose FHIR because it is vendor neutral and scalable. Our goal is to build an application that captures more accurate and timely data without being overly complex to implement or use. Stay tuned for more blog posts on this specific project as well as how other parts of the Department are using FHIR in addressing challenges in patient matching or leveraging open data APIs.
Where Can I Learn More?
If you're interested in learning more about FHIR, here are some resources (no pun intended) that helped me understand what it's all about:
- Video interview of Grahame Grieve (one of the chief architects of FHIR)
Information Week Articles on FHIR:
Online Course and Additional Materials:
How Can I Get Involved?
What excites me about FHIR is that we're already witnessing how it is being used to help solve seemingly intractable health challenges and spur further innovation. Using FHIR:
- Patients can access their health records across multiple providers, integrate this information into a website or smart-phone app, and share this information within their trusted care circles. This infrastructure provides a common platform where innovators can develop applications or other targeted interventions to promote healthy behaviors, prevent diseases, or even potentially reverse adverse health outcomes.
- Providers can better manage diseases by developing evidence-based goals, tracking how well their patients are performing against those goals, and having the best information to determine what needs to be done for each patient at the point of care.
- Public health officials can aggregate data from the care of millions of patients in real-time, identify what trends are emerging across populations, and help guide decisions (at the point of care as they're being made) about what interventions will best help people live longer, healthier lives.
- Payers can see risks across populations and have a better understanding of which patients need engagement to better control costs and promote positive health outcomes.
If you're interested in building (or piloting) FHIR-based solutions, like these, then join us at the HealthCa.mp on FHIR Code-a-Thonin Washington DC on April 1-2, 2016. You'll meet innovators across Federal and State-based organizations, startups, and research organizations. To get more background on the code-a-thon you can check out the HHS IDEA Lab Blog. See you in April.
Featured image photo credit: woodleywonderworks on Flickr