People whose careers have focused on health IT interoperability issue can sometimes feel like Sisyphus from Greek mythology — doomed to the eternal punishment of forever rolling a boulder up a hill. But the last few years have been more promising, with the advent of the FHIR data standard and federal regulations requiring EHR vendors to use standardized application programming interfaces (APIs). Other regulations have opened up patient access to data in digital format and targeted “information blocking” by vendors, health systems and payers.
By the end of 2022, all certified health IT vendors will be required to have FHIR-enabled versions, which is very significant, says Viet Nguyen, M.D., chief standards implementation officer at nonprofit standards organization HL7. “Requiring all the vendors to adopt FHIR will enable us to get to ‘EHR 2.0,’ by adding these Smart-on-FHIR components to improve data visualization, data collection and data sharing. It is really going to be powerful,” he says. Having stakeholders agree on a common set of FHIR profiles means that “when I request data from a certified health IT system, I know that what I'm getting back,” Nguyen adds. “That's going to be remarkable because it sets the first bar of what everyone is agreeing to do as we continue to advance the whole ecosystem. I think it is going to be a quantum leap for interoperability.”
Nguyen notes that the work that the HL7 Argonaut Project did five years ago around the common clinical data set requirements and creating FHIR profiles got the provider community on board with FHIR.
With last year's implementation date for the patient-access API, the payers were required to get on board. They are now using FHIR for Medicare Advantage and other programs. “Making data available is an important first step to developing the FHIR capabilities for any organization,” Nguyen adds. Making data available will start them on the path of working on challenges such as burden reduction, prior authorization, and quality measures. The HL7 Da Vinci Group is also developing patient cost transparency implementation guides to support the No Surprises Act and is starting work around risk adjustment.
HL7’s FHIR accelerators like Da Vinci bring communities of stakeholders together to identify opportunities to use FHIR and the FHIR data model to improve on certain workflows and efficiencies. Nguyen says that will lead to efficiencies much like we see in travel, e-commerce and banking.
Don Rucker, M.D., chief strategy officer at health tech startup 1upHealth, previously served as national coordinator for health IT. He gets credit, along with other ONC leaders over the years, with putting the policies in place to push interoperability further. In his speeches, he stresses bringing healthcare IT onto a level playing field with other industries. “What's being required here is basically the style of APIs that fuels the rest of the Internet, rather than having healthcare-bespoke APIs,” he says. “Having the type of standard that every app on our smartphone uses when they communicate with their back-end server is powerful. It allows you to use industry tools, industry expertise, and programmers, as opposed to them being purely healthcare-specific.”
Rucker says that industry stresses are forcing a reconsideration of payer/provider dynamics around utilization management, prior authorization, and quality measurement. That is going to require combining clinical and claims data into a single computable workspace that is available for AI and machine learning tools to analyze. “A lot of folks out there are going to be pointing these AI tools at healthcare, and the lingua franca is going to be FHIR,” he says. “When you look at the trends of consumer access to data in the app economy, when you look at the broader payer/provider space, when you look at the API rules, FHIR has a central role. When is this going to happen? My guess is a lot sooner rather than later, because the burning platform is the ongoing dissatisfaction with the healthcare system.”
Brendan Keeler is a senior product manager at startup Zus Health and was previously a product manager at Redox and Epic, working on interoperability issues. He says that when the FHIR API requirements kick in, some of the smaller EHR vendors are likely to struggle with it, just as some payers struggled with the patient access requirements from CMS last July.
Concerning the growth of EHR third-party app ecosystems like the Epic App Orchard and similar offerings from vendors such as Cerner and athenahealth, Keeler says that because of the information blocking regulations, the vendors really don’t have any leverage left to deny access to core FHIR APIs, even if they wanted to. As those ecosystems evolve, he is seeing several tiers develop where things like the patient access APIs are fully open. You can register your app and it's deployed in 12 hours. But there also are offerings EHR vendors are charging for that involve co-marketing deals, heavy technical support, a sandbox, as well as FHIR and other proprietary APIs or legacy methods of connectivity.
Keeler says the clinical and business use cases need as much help as possible from regulators. “Patient access is the most dramatically enabled by technology, both by the Cures Act, but also by the CMS regulations,” he says. “They very clearly give awesome new superpowers to pull data from payers and providers for consumer-facing apps. The business associates are radically enabled on the provider side, because they have these new ecosystems to join and create apps,” he adds, but it's actually provider-to-provider and payer-to-provider interactions that need more support.
Mark Savage was formerly the director of health policy at the Center for Digital Health Innovation at UC San Francisco and is now a consultant and the social determinants of health policy lead for the Gravity Project, an HL7 FHIR accelerator. He says the Gravity Project’s work defining social determinant data elements demonstrates some of the ways that the FHIR API ecosystem work can be a game changer.
Making it easier to share data across platforms is going to expand the ecosystem of who provides care, he says. “ACOs were an initial version of that, but now people are building the infrastructure to collect and exchange data that makes that broader ecosystem possible,” Savage says. “We started with clinical use cases and certified EHRs, but there are also many use cases for public health and social care.” This data-sharing ecosystem expands to include health and human service organizations, community-based organizations, social services agencies, referral platforms, patients, and their family caregivers. “Just think of the ecosystem we're talking about and what becomes possible with those FHIR APIs and the Gravity Project’s work to connect all of these different entities,” he says.
Many healthcare stakeholders also see great potential for work being done by ONC and CMS on “bulk FHIR” to lead to new ways to use clinical data in the EHR for measurement, rather than having to rely on claims data. CMS is looking at calculating digital quality measures electronically directly from the EHR. “You're not doing that patient by patient; you're doing that with aggregate and bulk data,” Savage explains. Other population health and public health use cases exist.
HL7’s Nguyen gives ONC and CMS high marks for the work they’ve done to date. “They are not promulgating regulation from on high; they're actually in the weeds with us. As we're developing implementation guides, they're paying attention to what the industry conversation is, and they have chosen discretionary enforcement when necessary,” he notes. “They recognize that there is a balance between driving the industry and recognizing that it takes time for standards to mature. We have to do a better job as an industry in helping folks understand that standards development is an ongoing journey.”