Mark Dunnagan - Interoperability Pt. 2: Open Data Exchange
Manage episode 405063332 series 3335700
In this episode we continue our interoperability conversation with CHESS Vice President of Health Informatics, Mark Dunnagan. Last time, we focused on the importance of shared data in value based care and the need to overcome any barriers. Today we talk about the logistics of interoperability and the modernization of data exchange.
Mark, last time we left off talking about data exchange There always seems to be ongoing conversations in this topic about APIs. Do you feel like more improvement in APIs could be a potential solution?
I do I use the metaphor of a quiver of arrows quite often when describing you know interoperability. I think you know it's my job as you know the head of a team that that must figure out how to get data and get it in a timely fashion and in a way that fulfills our contractual obligations and our obligations to the patient. I think APIs is one more arrow in the quiver. You know it gives us a programmatic way to access you know large volumes of complex data, but it's not necessarily the only way. You know when we sign on a health system let's say to one of our ACOs, you know I can pretty much rest assured that they're using one of a small number of vendors and you know those vendors are fully capable of producing certain constructs that that my team can consume. Same with most payers. Although you know, the outputs may differ certainly. But as I work my way down the chain, particularly in working with ambulatory clinics and what not, you know, I gosh last time I checked there are over 200 EMRs here in my home state of North Carolina. Each one of those with a slightly different interpretation of certain standards. Not all of them have viable API interfaces, you know, not all of them have the same way of communicating with them. So, I have to be open to old school HL 7, which is kind of the equivalent of opening up a channel and typing over it. I have to be open to flat file exchange. I have to be open to various forms of XML, JSON, and it truly depends on what that endpoint can offer. So again, APIs are extremely valuable but they're not the only tool that a team like mine has to has to be able to wield to be interoperable to be successful in the exchange of healthcare data.
Interesting. So as someone who's spent a career in the data and informatics space, can you share how these analytical tools help control the cost of healthcare?
There's many answers to this. I would say again I'll draw back to what we do which is value based services. You know I need to know when something happens and I need to be able to inform our performance improvement teams and so that they can communicate with the providers. I need to inform the care managers when something of interest when someone is checked into a hospital, someone has sought, you know, specialty care outside of network, when someone has been discharged, they need to know that and I need to inform them, you know, not only that it's happened, but give them enough descriptive information that they can intervene appropriately. I would go further to say that I need to glean enough good information, rather my team has to be able to accumulate and collate enough information to get ahead of what might be coming. You know, we're making some very powerful strides, you know, not only in, you know, intelligently stratifying our population to kind of know who to intervene with first, but also in quantifying rising risk and rising cost. Who do we think based on what we're seeing happen now? What do we think's going to happen to them tomorrow? And can we get ahead of that in time to affect that? Can we keep them out of the hospital? Do we know there's a costly intervention or fall coming, and can we intervene or get them some community based services in time? So, you know it's a large part of what we do and and again something that at least on the value side we have to contemplate every day.
Do you think that this whole system would be better if payers had access to provider data and providers had access to the payer data?
It's a sticking question. I think the altruistic mark thinks that everybody should know everything. And yet when you start to bring in, you know, let's call it business motivations, the inevitable pursuit of profit. You know, healthcare organizations, though altruistic in nature, still have to compete to survive. They're still have to make money. So, they're still going to be hesitant to share information where they think they might lose clients. And I think there's still the thought that you know if you share everything about someone particularly genetic testing or very sensitive conditions that payers might use that to decide who to include in their roles who not to. I'll be honest with you it's a gray area. I think again altruistically I do think everyone should share everything but where profitability human nature and what not come into play. I would say that it's a gray area where I would yield to our chief compliance officer for, you know, truly the right answer because I'm not sure I have it.
Well Mark, we've talked a lot about our current state of interoperability. Let's talk about how data exchange has been modernized in the last few years and I'd love to hear your thoughts about, where we were, where we are now and where we're inevitably going. What does the future hold for us?
Well, I'm going to have an interesting answer to this because I if for someone that that was on the bleeding edge for so long. You know I I do believe in the in the value of the CCDAs do believe in the evolution to fire FHIR not FIRE, you know and API driven exchanges. I do believe that's where we need to be. But I get a lot of value in kind of the commoditized simplistic exchanges of the past. You know we still maintain and I'm and I'll proudly state this you know, meat and potatoes forms of flat file exchanges, you know they are easily set up, easily replicated and you know again fairly commoditize skills where it's you know straightforward to staff up with respect to those types of feeds. I would say we are very open to some of the more modern constructs and API driven exchanges and things like this, but they're not always in demand because in order for them to truly work you have to have skills on both sides of that conversation. So, I would say again to the to the quiver of arrows and you have to be open to all you know I look forward to you know a fairly self-maintaining build at once and use it a million times over API driven exchange. But that's not necessarily that's not my reality And you know particularly with respect to you know our move or CHESS's move into supporting Medicaid here in North Carolina, it's not the reality. You know the payer based exchanges are all flat files and what I'm trying to integrate with a small mom and pop shop that's just trying to do the right thing by you know some of the less fortunate populations they serve, they don't necessarily have the best of EHRs you know and I have to get to that data in the best way that I can. So again, you know if we were all able to speak the same language and all have the facilities and the depth of bench to achieve that which is the tip of the spear of modern sustainable interoperability, I would say that's fantastic. But the reality is, is that it's all over the map. So, while we seek that fixed point on the horizon there, there's I have to handle everything in between.
(chortle) You’re right, that is surprising, but it does make a lot of sense. Mark, tell me, what is the ultimate goal of interoperability?
I mean, for me it's that it's that altruistic exchange about everything to do with a patient's health. I would love a world where you know whole person care is a reality and where you can openly exchange everything from potentially Mark struggles with substance abuse to Mark's behavioral health challenges to you know Mark's being overweight and challenges that, you know there's a there's a million different things that you know in and of itself are sensitive within certain silos that I would love to be openly exchanged and know that that is not to the detriment of Mark's future insurability, to him being able to sign up for a new provider should he be relocated to, you know, a number that that a skilled nursing facility 50 years from now or 30 years from now would take me in knowing full well that you know things that have happened to me during my lifetime. So, you know I hope for whole person care. I, I and I hope for the open exchange, but you know, again there are some realities to the modern world that are probably going to continue to be barriers for some time yet.
Well, Mark, is there anything that I have neglected to ask that is important to this conversation today that you'd like to share?
Interoperability means so many different things to many different people. It's effectively a 4 letter word. I think, you know we've kind of skimmed the surface of what interoperability means to me. I mean we could talk like this for hours on end, for days on end. But you know I would just close with you know it's an important concept and again with the conversation of the metaphor, the metaphor of a conversation rather, that it it's part of everything that we do. And as you know value based care is here to stay and particular to value we need to know and then communicate out to care teams and to providers. We need to know what happens to our patients as quickly and as efficiently as possible and in the same way we need to communicate that data, our insights, our predictions based on the same in order in order to see that those patients are cared for appropriately. I mean value based care at least from my simplistic view is the path to containing healthcare spend. It's the path to kind of squeezing what you can out of existing workflows, encouraging new workflows and you know stem the rising tide of cost. And you know we need the open exchange of data or at least as open as it can be in order to do our job. So I hope that's a good answer.
It’s outstanding. Mark Dunnagan, thank you for joining us today on the Move to Value podcast.
It's been a pleasure. Thanks Thomas.
62 episodes