At ViVE, Micky Tripathi Maps Out the Near-Term Future of Interoperability

With regard to that first category of “building the foundation,” Tripathi said that there are three key elements involved: “promoting data standards; filling in HIT gaps, and the coordination of federal agencies. We continue to move forward with USCDI,” Version 4 of the United States Core Data for Interoperability, a standardized set of health data classes and constituent data elements for health information exchange. The USCDI establishes a baseline set of data that can be commonly exchanged across care settings for a wide range of uses. The draft of Version 4 of the USCDI was released in mid-January (see here for Senior Contributing Editor David Raths’ report on that release).

With regard to USCDI, Tripathi told the audience, “We’re moving really fast and I’m pushing really, really hard, because I want all of you to be able to do all the great things” in terms of interoperability. “We often hear the old adage that we need business to do these things, because business is fast and government is way too slow. But actually, I’m hearing from people that ONC is moving too fast with USCDI. In the past, our cadence was once every four years. Now, we’re producing updates every year.”

What’s more, Tripathi said, high priorities for ONC in the ongoing forward evolution of USCDI will be incorporating ideas around “health equity, health insurance information, patient status; those are all really important foundational elements of core data. We’re also working with our federal partners on what we call US CDI Plus. We want to better integrate federal programs.” There is a public health version of USCI being developed with the Centers for Disease Control and Prevention (CDC), a version being developed with sister agency CMS (the Centers for Medicare and Medicaid Services), and a cancer version, with the National Cancer Institute. “This is a fertile field for innovation,” he emphasized.

Meanwhile, “The second area within that founding involves making interoperability easier,” Tripathi said. “Think about ten years for the most complex part of the most complex economy in the world and from a starting point of 5-10 percent of EHR [electronic health record] implementation. Now, 96 percent of hospitals, 80 percent of ambulatory care providers. As we know, there are big, important parts of the healthcare continuum—LT-PAC [long-term are] and behavioral healthcare didn’t get that opportunity. We’re working in the Department [of Health and Human Services] to say, what are the levers we have across all the HHS agencies, to support those types of providers with adoption of HIT?”

Further, he added, “We tend to lump behavioral health providers together. But you’ve got outpatient medical centers, inpatient psychiatric facilities, addiction recovery centers. And they have different needs. There are innovative ways we can deploy technology to serve the needs of those organizations. So we’re working with our behavioral health IT partners to help implement behavioral health IT.”

With regard to FHIR-based APIs [application programming interfaces compliant with the Fast Healthcare Interoperability Resources standard], “We’ve been moving forward on federal health IT alignment,” Tripathi noted. “In July 2022, the Secretary put in place a policy requiring all HHS agencies to align around data, FHIR APIs, etc., for better alignment, and making that a requirement for funding. So that cuts across FDA, CDC, CMS, HRSA. We’re working really hard in building up our cross-agency muscle at ONC to help other agencies” work forward together in concert. “All that is to help complete the digital foundation,” he emphasized.

Meanwhile, Tripathi said, “Now, let’s talk about the second category—making interoperability easier. And that involves FHIR APIs and TECFA”—the Trusted Exchange Framework and Common Agreement. “With regard to APIs,” he noted, “ONC’s regulation went into full effect Dec. 31 that all certified EHR vendors have a standard FHIR API that allows access to interoperability. In the control of the customer. Vendors had to make it available to the customer with the customer in control of the API. That includes other providers, health plans, etc. Now, 95 percent of EHR vendors actually certified to that by mid-January. I’m really encouraged that the industry picked that up and worked with it,” he added.

“But how do we know what’s really going on? There’s still a ton a of friction. We’re doing a couple of things to help. One is Project Lantern, an open-source search algorithm. We essentially sweep across all the known endpoints available out there, and deliver back, here are the endpoints that each vendor has made available, and here are the FHIR supports. If you google Lantern and go to the HHS site,” he added, “you’ll find the Lantern dashboards. It’s open-source code. You can implement it yourself.”

What’s more, “One of the things we’re doing to help build out this rich API ecosystem—we’re going to start looking at payer APIs,” Tripathi said. “We’re working together with CMS, CDC, and HRSA, on a FHIR strategy. ONC requires it in EHR systems. CMS requires that regulated payers make available data to members, that started in 2021. Now in CMS’s interoperability rule, they’re requiring that regulated payers make available a provider-facing API to get claims information, and a FHIR-based payer-to-payer API based on that same FHIR-based specification. So ONC is pushing and CMS is requiring, EHR-supported APIs available to a none-=patients, providers, other payers. And then CMS telling payers, you’re required to make an API for claims access, payer-to-payer data exchange.”

A little while later, Tripathi asked, “So what will we do with all these APIs living out in the wild? How do I connect this API to something else? And that’s where TEFCA comes in. We’ve got network interoperability across the country today. It still doesn’t do a whole bunch of things we need: filling in the gaps for the 30 percent of hospitals who don’t participate in those networks because it’s too hard; still doesn’t serve patients; public health; patient-payer uses. And TEFCA is designed to support that. On February 13, we announced the first six QHINs [qualified health information networks], and there are more in the pipeline; we’ll be making further announcements soon,” he promised.

“And the idea is that you want network infrastructure to support FHIR APIs int h wild. Where are all the endpoints for those organizations? How can I find an individual’s records, as authorized, patient search? And security protocols between APIs. We’ve got the cert (certification), same rules of the road. That’s the scalability of FHIR APIS, WE need TEFCA-type infrastructure to support it.”

Tripathi went on to discuss “the third major area: how do we encourage mor information-sharing? The information-blocking rules. We’ve put the rule into effect; it’s in full effect as of Oct. 1, 2022. And just because you build it doesn’t mean they will com. That’s where the information-blocking provisions of the 21st-Century CURES Act come into place. The OIG of HHS will issue a report; and we’ll be involved in the release in September of a draft rule for offenders.”

And, he said, “The last thing is this: ensuring appropriate use of data. We’ve got a tremendous amount of data that’s starting to flow. We’ve got questions about equity. How do we make sure the information supports health equity as a design feature, as a core construct? We’re supporting standardized data elements with USCDI. And supporting consistency of capture of the SDOH elements. Secondarily, algorithms. We’ve published a blog series on starting to think more about the use of algorithms in healthcare. Chat GPT just debuted, and last night over dinner, I heard ten ideas I had never heard before—after a month and a half since Chat GPT was introduced—for new ways to use algorithms. It’s both exciting and frightening. And it will require governance at the local level. The users need a better understanding of what the algorithm is doing.”