Interoperability Isn’t for Humans (yet) | Part I of III

A conversation about the challenges and the future of interoperability in healthcare.

Those of us in healthcare and health IT have been hearing about interoperability for over a decade. EHR vendors originally made great strides in digitizing health data and cutting down on the actual paper in hospital paperwork but have since struggled to innovate both with regard to bringing value to the data they already have and to elevating the process of viewing and capturing the data at the onset.

While HDAI is not an EHR vendor, and we aren’t claiming to solve the problem of interoperability in healthcare, we are able to circumvent certain challenges these companies and systems face. The following is a summarized transcript of a conversation with David Clain, VP of Corporate Strategy, about the state of interoperability in healthcare – and where we might go from here.

***

Q: Hi David, tell me a little bit about your background and how it intersects with EHRs and the issue of interoperability.

David: I started my career at The Advisory Board Company, where I spent about three years advising health system finance departments and CFOs, primarily, in a range of strategic and operational issues. And that included on the operational side, a lot of work on revenue cycle, which includes coding and intersects with EHRs and practice management systems and data warehouses.

And then on the strategic side, I was helping them think through new payment models. This was around the time the Medicare Shared Savings Program was introduced and bundled payments were introduced by Medicare and by some commercial payers and in both cases, to really evaluate those contracts, you need an understanding of the population you’re serving, the quality of care that you’re providing, and the cost of that care. Then you can start to evaluate these contracts from a financial perspective and think about whether there are opportunities to improve the quality of care you’re providing. All of this analysis depends on having access to data that lives largely in EHRs.

“The fundamental challenge is that interoperability is really a machine-centric idea {…} and it really, as far as I’ve seen, hasn’t done much to meaningfully improve the quality of practice for physicians, and certainly not patient care.”

Q: Got it – so you’re on a mining mission. And everyone had the same EHR and it ran smoothly?

David: Right – and then we wouldn’t be having this conversation. What we saw in large health systems is that they generally have multiple different EHRs, as they’ve likely formed through acquisitions of smaller hospitals and smaller physician groups that are all on different systems. And so, essentially what they have to do to get a relatively full picture of the work that they and their clinicians are doing is to build these data warehouses that pull from all of those different systems and then analyze that. Effectively, all EHRs are doing is serving as data repositories, with minimal analytic capabilities and decision support services—and those that do exist are restricted to the visits recorded in those EHRs, so providers don’t have almost zero clarity on what happens to patients when they’re outside their own four walls.

Q: So these multi-million dollar software systems are basically functioning like giant electronic filing cabinets?

David: Essentially, yes. Filing cabinets with locks on every drawer, and you don’t have all the keys. And when you manage to open the drawers, some of the folders are missing, and some of the documents are in different languages.

“If a doctor can’t actually learn the important things they need to learn about their patients from [the] data, what’s the point?”

Q: Yikes. And then you went on to work for one!

David: Yes, I went on to athenahealth, which is an ambulatory EHR and practice management vendor with more than 100,000 providers on its network. This was after the passage of Meaningful Use and the Affordable Care Act and a bunch of other pieces of legislation that included provisions to encourage EHRs to share data with each other, to be interoperable.  I will say that Athena, I think in part because it’s a smaller player in the EHR market, has really advocated for interoperability. They pushed strongly to inpatient EHRs to share data with them to pair with their outpatient data. And what we saw at athena time and time again is that despite the fact that they are legally required to it’s just incredibly hard to get these systems to share data in any way at all, and especially in a way that is actually useable for doctors. That has gotten better in the past few years, without a doubt—but it’s by no means a solved problem.

“Despite the fact that they are legally required to it’s just incredibly hard to get these systems to share data in any way at all, and especially in a way that is actually useable for doctors.”

Q: You’d mentioned in a previous conversation that right now, even when interoperability works, it’s really for machines and not humans. Can you talk more about that?

David: Sure. For example, if you are outside their network, what you get from one EHR vendor in particular is just a list of potentially hundreds of clinical documents, many of which, probably most of which, have no real relevance to what it is you as a clinician would want to know about the patient. It more or less replicates what a fax machine does. Some EHRs would have a clinical document for every phone call that a patient made and talked to a receptionist, PDFs with lab results that weren’t structured or interpretable. You’d have to go through as a doctor and read hundreds of these documents and figure out what to with that information, which of course doesn’t happen – they don’t have time to do that.

Q: So far, we’ve mentioned filing cabinets and fax machines in a conversation about these huge software/tech companies that were supposedly responsibly for bringing the digital evolution to healthcare. That’s a frustrating disconnect.

David: Right, it’s incredible. An orthopedic surgeon recently told me “EHRs are better than paper—barely.”  But given the cost of implementing an EHR and the stakes for patients, that’s an awfully low bar.

I think the fundamental challenge is that interoperability is really a machine-centric idea. It’s about connecting to APIs and it really, as far as I’ve seen, hasn’t done much to meaningfully improve the quality of practice for physicians, and certainly not patient care. It doesn’t generate patient insights or even flag important information. And interoperability as we are currently defining it allows EHR vendors to check the box by implementing some standard methods and formats for exchanging data, but if a doctor can’t actually learn the important things they need to learn about their patients from that data, what’s the point?

“What you get from one EHR vendor in particular is just a list of potentially hundreds of clinical documents, many of which, probably most of which, have no real relevance to what it is you as a clinician would want to know about the patient. It more or less replicates what a fax machine does.”

Q: Thanks, David! I look forward to Part II where we focus on EHR vendor’s incentives (or lack thereof) to share their data. ‘Til next time!

David: Talk to you then.