Modern Healthcare technology reporter Jessica Kim Cohen caught up with the Trump administration’s top health information technology official Dr. Donald Rucker, who leads HHS’ Office of the National Coordinator for Health Information Technology, to talk about the industry’s momentum on interoperability, addressing long-standing problems with patient-matching, and what could be next for the agency. The following is an edited transcript.
Modern Healthcare: 2020 kicked off with healthcare executives wondering when the final interoperability rules would come out—which ultimately happened in March, right as COVID-19 really hit the U.S. Fast-forward to the end of the year and they’ve been delayed into 2021. Are you concerned that the industry is losing momentum on interoperability?
Rucker: We had an event in early December where developers were talking about what they were doing in the API (application programming interface) space and I think it’s actually quite the reverse. The API world that is at the core of our rule is frankly exploding. Microsoft has released a FHIR service on their Azure cloud platform. Google announced the same, and I believe Amazon will be doing the same. So, we will have all of the big cloud platforms facilitating these APIs. Most of the big EHR vendors have already built out some version of the FHIR APIs. And it’s going everywhere—some of my colleagues showed me student projects at Georgia Tech were building FHIR servers and testing against our test bed. When students are doing it in class that’s a pretty good sign. I’ve been in this field for 30 years, so to see this now is pretty gratifying.
MH: One of the reasons you cited for delaying the interoperability rule was that hospitals’ resources had been stretched thin early in the pandemic. But many hospitals are still struggling financially. Are you worried about how that financial strain will impact their ability to support IT efforts, such as APIs and interoperability?
Rucker: What we’ve tried to do—and I think have done—from a computer science and from a resource point-of-view is relatively technically lightweight lifts. What’s expensive for hospital systems are things that require manual intervention. But for these changes, the docs and nurses don’t have to enter new data, you don’t have to replace your electronic medical records—it’s really providing a server endpoint, which is a pretty common activity in the modern world. What the rule, in its simplest form, says is that you will stand up a server endpoint that connects securely to your electronic medical record database. That basically means the vendors will put a database front-end, or a database connector, that attaches to an internet server. That’s really the (technical) lift. In institutions there’s always policy and lift there, so the (new compliance) timing is really reflecting that some of the same IT folks might, potentially, be needed for both tasks, but realistically we believe it’s the EHR vendors who will be standing up the servers.
MH: There’s no shortage of tasks to complete in healthcare right now. Any advice for how healthcare CIOs can keep interoperability and related IT efforts on the radar of the CEO?
Rucker: If you’re a CEO and you’ve built up a capital infrastructure and invested infrastructure in your system that’s based on certain assumptions about how you’re going to be able to move traffic into your system—that world is changing. It’s changing not just because of the ONC rule and the CMS rule, but I think it’s changing because of broader ability of patients to learn about healthcare. It’s being changed by broader expectations of what we expect in an app economy. Consumers are going to be empowered. I can’t imagine a CEO of an enterprise, of any sector of the economy, these days not having a pretty keen eye on the digital.
MH: Some people have raised concerns that inadequate interoperability and patient-matching in healthcare could pose hurdles for tracking COVID-19 vaccinations. Is that something you’re concerned about?
Rucker: I’m not a COVID-19 vaccination person, just by way of disclaimer. But I would encourage immunization information systems to work with their local health information exchanges to share vaccination data and maybe piggy back off the enterprise master patient index tools that the HIEs have already established and have up at the scale of millions of adults. I think there’s an opportunity there. To the broader issue of patient matching, I think the question that people often anchor conversations on is: “I need another federal number,” beyond the Social Security number, beyond the Medicare number, beyond the driver’s license. But having another number, or a better ID card, doesn’t solve what we need to solve to have a seamless patient-controlled economy. I think the way to look at patient matching is really in a modern, app-type of way, a service-type of way, as a combination of authentication, authorization and consent.
MH: Any closing thoughts on opportunities you see for ONC in the next year?
Rucker: That’s for others to decide, ultimately, but I think we have a strong bipartisan tradition of moving the data standards along, which is really central to all of this. I think if we’re going to get rid of some of the big burdens on providers—so things like quality measures and prior authorization—we’re going to have to:
A) merge clinical and financial data, so those aren’t totally disconnected streams. It’s hard to imagine we can shop for value if there’s no connection between price and product. We have a lot of work there. And,
B) a lot of the burdens both on patients and on providers—with things like quality measures, prior authorization—are because we haven’t had any other way of measuring data. These big data, bulk FHIR APIs (a way to export data on a population of patients that’s included in the interoperability rule) are going to give us a much better sense of what providers are up to. Are people doing undertreatment? Are they doing overtreatment? Are they doing inappropriate treatment? Are they not doing prevention or skipping treatment? Everybody’s been talking about the search for value and quality in healthcare, but we have never, ever had an elegant way of getting the data, in a robust way, to actually truly figure out quality. With the bulk FHIR API, you’re going to be able to look for the patients that you have a legal right to see under HIPAA, and you’ll be able to get outcome data, intermediate variable data, spend data, diagnostic data, treatment data.
There’s lots of folks working on this. It’s not just ONC. It’s HL7 (Health Level 7, the standards development organization that oversees the Fast Healthcare Interoperability Resources framework), researchers, the big tech vendors who are putting cloud APIs out there, the EHR vendors that have really responded with FHIR API. I think there are a lot of folks who have worked together in the past and I’m pretty confident will work together in the future.