Tripathi is no stranger to the thorny issues at the heart of healthcare IT, such as medical record interoperability, standards, and certifying electronic health record software. His previous 20 years of work in the field let him hit the ground running. He most recently worked as chief alliance officer for population health management technology company Arcadia, previously served as president and CEO of the Massachusetts eHealth Collaborative, and has been on the board of directors of the HL7 FHIR Foundation, CommonWell Health Alliance, and The Sequoia Project.
Tripathi recently granted an interview to HealthLeaders. Following are excerpts from the discussion, lightly edited for space and clarity.
Now we’re putting into effect and trying to do everything we can to accelerate it, because it’s proceeded a little bit too slowly from my perspective. So now, of course, there are timelines already built into regulations, so it’s not as if we changed those. But I’m doing everything I can to try to evangelize with the industry that they should be moving ahead of the actual regulatory timelines, to the extent that they can.
HL: With the rise of telehealth, especially in the last year, what role will ONC play to ensure that the barriers that have been lifted will be permanently changed, so patients can access their doctors this way going forward?
Tripathi: ONC doesn’t directly play a role there. The reason, and this is sort of a subtle point in the health IT regulatory space, is that ONC is really an enabler of other agencies who have business needs. We don’t go out and just certify systems on our own. Telehealth would be a great example. ONC doesn’t really have the authority, nor would we just go out and say, “Oh there’s a set of technologies that we think need standardization, so let’s go and start certifying those, or start making those more standards-based.”
We would follow an agency like CMS (Centers for Medicare & Medicaid Services), should they make a decision, for example, that they are going to keep the payments that they started to allow during COVID. That’s when ONC would come into play and say, “Okay, now we support CMS, who’s the business owner in creating a set of standards that then get put into rules.” Maybe that turns into certifications as well, and we go down that path.
HL: What key technology challenges are hospitals and health systems facing related to telehealth, and how will ONC help?
Tripathi: I suspect that the key issues that they’re facing are workflow and integration challenges, which is to say that no one wants to be in two different systems. A patient may or may not care. If you’re on the patient side, you get a link, and you just click the link. You don’t know that it’s within the provider’s EHR system, or whether it’s a different system, as long as that physician is on the other end when you click the link, that’s all you really care about.
But from the provider side, what they want is the ability to, for example, be able to document in the medical record when they’re having this encounter with you, and not be flipping back and forth, essentially. They’d like for it to be recorded as an actual encounter in the EHR. If you think about how that might work a year from now, you come and you say, we had a video visit a year ago. Unless that was integrated in their system, they literally would have had to type in, “video visit,” and create an encounter that’s like a video visit, and then you have to go through all of that. If it’s integrated into your system, you just click “video visit.” It automatically knows what’s going on and puts in that information, and then allows you to document the medical record as you would normally. And then it’s all integrated, just like a telephone encounter for like a regular encounter. So it’s mostly those workflow issues, I suspect.
Also from a [privacy] perspective, always wanting to know that it meets the HIPAA security rule. There was a temporary suspension during the public health emergency, so that people could use regular commercial solutions, like FaceTime and others, but if this goes forward, I think there will be a question of [whether there] are there standards or [other] things that are necessary. If this goes forward as something that’s more enduring, and the public health emergency is lifted, we may go back to say, “Well, those solutions do need to meet the HIPAA security rule, which will be an OCR (Office for Civil Rights) and CMS decision.” Then again, ONC might come into play at that point to say, “How would one determine that?”
HL: How can the patient ID problem be solved once and for all? Patient matching efforts seem endless.
Tripathi: Oy vey. The patient ID question is a really interesting one. As we know, the Congress has, since 1996, prevented the use of federal dollars for a universal patient identifier that was identified in the original HIPAA law. And there seems to be possible movement in the Congress, perhaps toward getting rid of that ban, but that ban is in place. I think we can, at least for now, say that is not going to be something that comes from the federal government. But it doesn’t prevent a voluntary identifier, for example, being used in the market, if it got wide adoption. It doesn’t prevent or ban a universal identifier that’s provided by the private sector, What it bans is federal money being used toward a universal identifier. So that could be one pathway to getting to one.
There are a couple of companies out there that already claim that they have a unique identifier for every individual in the country already, because they do clearinghouse functions, or they do credit history kind of data. They’re already making that claim. So there are some kinds of solutions out there right now. They’re not widely adopted for perhaps a variety of reasons.
One of the things that we need to understand is that a universal identifier, even if it got magically created tomorrow—because we’ve now had to spend, since 1996, all of those years developing other ways to do patient matching—it isn’t as if that would be a magic wand that solves all matching problems. If you’re a hospital system, for example, you’ve already invested in technology and processes to do that matching. And now, if you think about that, I might use seven, eight different data categories to match [a patient]. I would use your first name, your last name, your middle name, your address, your birthdate, and maybe your cell phone number. So I’ve already got a technology that kind of does that, and it does it pretty well. It doesn’t do it perfectly. It does it pretty well. I’ve invested as much as I think makes sense.
Now the universal identifier comes along. The first thing that happens across the industry is that everyone has to spend more money to incorporate that universal identifier in their system. So every vendor system now has to incorporate it.It all has to be promulgated, and they have to adjust their processes now to accommodate that universal identifier. Then the question is, “How much does that cost the industry? And where are we going to start to see the benefits?” That’s why I think that’s a little bit of a challenge as well. That’s not a magic overnight solution to patient matching problems, and we still could face adoption problems, even if one magically appeared. That doesn’t mean that it wouldn’t be helpful. I think it would absolutely be helpful. But there are challenges because we’ve had to come up with other ways of doing it. Those other ways are not perfect, but are adequate solutions, in many cases, as imperfect as they are.