If you really want to get Chuck Christian’s goat, ask him why all banks are connected through ATMs, and yet healthcare still struggles with interoperability. The simple answer? It took quite a bit of time for banks to figure out a system. Oh, and accessing data is slightly more complex than withdrawing $10. In the latest of our Fireside Chat series, Christian talks about the progress Indiana Health Information Exchange has made in facilitating data exchange among organizations, why it’s not a lack of standards, but rather the lack of a single standard, that is hurting the industry, and what healthcare can learn from industries like banking and railroads. He also discusses the need to use data to help and not hinder clinicians, and why he strongly disagrees with reports suggesting that HIEs have maxed out.
- SHIEC’s growing network
- Patient-centered data homes
- Sending & receiving ADTs through the Heartland Pilot
- Using zip codes – “I don’t know why we didn’t stumble across it sooner. It’s so simple.”
- Eyeing clinical summaries
- Interoperability’s critics: “I don’t believe a lot of what I read.”
- Multiple standards: “Pick one.”
LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO SUBSCRIBE TO OUR iTUNES PODCAST FEED
We’re using the standard transaction sets and the backbone of the eHealth Exchange to do all this. It’s the same thing most HIEs in the country are hooked to already.
The idea of the patient being in control of their healthcare still a valid idea, and this does not replace any of those; it’s just in addition to what’s already there. But I will tell you, we’re querying other exchanges to say, do you have data? And it’s been very interesting to see how many of those patients actually have care outside of their normal region.
If you’re talking about whether we’re sending transactions all over the US on a regular basis, the answer is no. We’re doing it as needed basis.
We’re trying to stitch together a quilt of HIEs so we can exchange data and cover as much as the country as we possibly we can. And that’s why you’re going to have other options, like using the direct model to bundle up a CCD or some other document, and send it through a secure email type of infrastructure. If that’s all you got, it’s a whole lot better than nothing.
We as an industry need to come together and agree these are the standards we’re going to use; that there are going to be some winners, and some losers. Let’s concentrate on those and enhance those to take care of the use cases rather than having standard sets compete with each other.
Gamble: Hi Chuck, thanks as always for taking some time to speak with us.
Christian: Glad to do it.
Gamble: It’s been a little while since we spoke, and obviously quite a lot has been going on with Indiana HIE. Let’s start with a piece that recently ran about IHIE connecting with other major HIEs. Can you talk about that?
Christian: We’ve been collaborating with the other HIEs around us, and have been for quite some time. The Michiana HIE, which is right around South Bend, and the Health Collaborative, which HealthBridge is part of in Cincinnati — we’re moving data back and forth through both of those every day. We’re also working with SHIEC, which is a trade association for HIEs in the country. SHIEC stands for the Strategic Health Information Exchange Collaborative, and it’s about 52 or 53 sustainable HIEs throughout the country — and growing every day, and we work together on common things. Even though we’re not all alike, there are similar things we need to address, and so we work together.
We have this thing called a patient-centered data home that allows the data to move wherever the patient goes. In May, we have the Indianapolis 500 here, so I’ll use that as an example. Let’s say somebody from East Tennessee came to Indianapolis for the race, and unfortunately winds up in one of our emergency rooms in Indianapolis. That ER physician would only know what the patient or their family members could give them from a history standpoint. What we can do is, we have seven HIEs in what we call the Heartland Pilot and East Tennessee happens to be one of those, so when that patient hits the emergency room at one of our hospitals in Indianapolis, we will get the ADT transaction for the admission. We’ll see that the patient is not from here by looking at their zip code. If, let’s say, we see that the zip code is an East Tennessee’s region, we’ll query the exchange in East Tennessee and ask if they know the patient, and if they do, we’ll ask for a clinical summary. And they’ll bundle up a clinical summary and send it back to us, and then we’ll present it in the patient’s record for the ER physician to look at in real-time.
Now, at the end of their visit in the emergency room, the goal is also for us to bundle up a clinical summary and send that back to their home HIE, so that their attending physician or their primary care physicians in East Tennessee will have the advent of whatever care they had while they were in Indianapolis. The concept of just using the patient’s zip code — I don’t know why we didn’t stumble across it sooner, but it is so simple. It doesn’t require a massive record locator service or anything other than us keeping up with what zip codes are in what exchanges that we need to ask for.
We’re using the standard transaction sets and the backbone of the eHealth Exchange to do all this. It’s the same thing most HIEs in the country are hooked to already. We exchange data with the Social Security Administration and the VA through that mechanism, and so it’s no great push to do it, since we’ve been onboarded to that and we’re already successfully exchanging data that way. It’s just another transaction that we’re using.
Gamble: And then you referred to it as a pilot, so it’s still in pilot phase at this point?
Christian: It’s an operation. We’re exchanging a lot of ADTs right now. A couple of the exchanges are changing their technology stacks and changing their vendors out. And so all seven of those are not connected yet, but those of us who are able to go ahead and do ADTs are doing that right now and then the clinical summaries will come very, very soon. We wanted to hold off until all seven in the exchanges are onboard, but we may not be able to do that because of delays that one or two have had, so the rest of us are just going to go ahead.
Gamble: You said that clinical summaries are the next thing. Do you foresee it just incrementally evolving to as everyone can get onboard with it?
Christian: Right. There are two other pilots going on in the country right now. There’s one centered around Oklahoma and Arkansas which is adding other HIEs, and then we have one in Colorado, Utah and Arizona. Our next step while we’re doing the regional ones is connecting those pilots and working on the technology to connect those pilots together. Where do a lot of people go ski? They ski in Colorado. They ski in Utah. If you look at the map of where patients come from, a lot of people from Northeast will go to Colorado or go to Utah to ski. Unfortunately, skiing has the tendency to cause you to also see an orthopedic surgeon to fix something. Those are the type of things we’re looking at.
This idea of exchanging data nationally is not something that’s going to happen for regular and usual patients, because healthcare happens regionally. Most of my healthcare happens right here in Indianapolis, and has for a long time. If something different comes up that can’t be handled, I’ll go to one of the larger facilities like MD Anderson or Mayo or Cleveland Clinic. But on a routine basis, other than when people are travelling, that data is on an ‘I need it now’ basis, and so we need to be able to move that data to quickly so they can get it without a lot of friction in the transaction. I don’t see any reason that we can’t use this model.
Now, one of your next questions might be whether this model should be used instead of CommonWell or Carequality, and the answer is no. I think that it’s complimentary, because if you look at the various use cases around interoperability, it’s like houses. Not every house is painted the same color, and they’re not all built the same either. There are different needs for different use cases. This is one of those use cases of a patient not being in the region where they typically have care being able to get to that data without them having to intervene and do anything.
The idea of the patient being in control of their healthcare still a valid idea, and this does not replace any of those; it’s just in addition to what’s already out there. But I will tell you, we’re querying other exchanges to say, do you have data? And it’s been very interesting to see how many of those patients actually have care outside of their normal region. In Indiana, between the exchange in South Bend and Indianapolis where those patients are going come to Indy for specialty care, we know that, and so it’s kind of a no-brainer, which is why we’ve already been working with them collaboratively for years.
Gamble: It’s another reason why it makes so much sense to collaborate and work together, because we do see so much travel, and things are bound to happen when you’re away from home.
Christian: Absolutely. It’s inevitable. You never hope it happens but it does, particularly if you’re going to be on the highway.
Gamble: Now, as far as like Michiana HIN and East Tennessee, I imagine that you’re in contact with them pretty often.
Christian: We have monthly calls, and they’re actually more frequent when we’re setting up doing the technical pieces around that, because we want to make sure we do adequate testing so we don’t have any false negatives or false positives. So we’ll do a boatload of testing before we start actually moving the live transactions.
The interesting thing is the folks at ONC are aware of the pilots that we’re doing. They actually funded the one for the Heartland, and we’re providing metrics and feedback and updates back to them on a regular basis.
Gamble: As far as like getting those numbers over to them, does that happen on a continual basis? It’s probably easier now to be able to compile that information.
Christian: I can tell you exactly how many transactions of what type that we’re sharing between each and every one of those, as the other pilots can as well. When I read in the press and some of the trade journals that interoperability is struggling, I have a tendency to say, ‘no, not really.’ If you’re talking about whether we’re sending transactions all over the United States on a regular basis, the answer is no. We’re doing it as needed basis, but there’s no real reason. These are not like banking transactions that are moving from the Federal Reserve Banks. They’re not like that. Care is provided locally and regionally.
Typically in every state, if you ask anybody that lives there, they can tell you where the major healthcare centers. For example, in Georgia it’s Atlanta, in Indiana it’s Indianapolis, and in Illinois it’s Chicago. The specialty care and the high-end care is actually provided in those hubs of care. Community hospitals do a great job of taking care of their patients — I worked at one for 25 years. But there are always patients that need to be transferred to a higher level of care or need a specialty that’s just not available and those will go to those regions.
I don’t believe a lot of what I read that interoperability is suffering or not working. Now, if you look at the map of where HIEs are, there are some white spaces, and we’re trying to stitch together a quilt of HIEs so we can exchange data and cover as much as the country as we possibly we can that way. And that’s why you’re going to have other options, like using the direct model to bundle up a CCD or some other document, and send it through a secure email type of infrastructure. If that’s all you got, it’s a whole lot better than nothing.
We need to do a better job with this. I’ve also read that there is a lack of standards; well, I respectfully disagree. There are some gaps we need to fill, but there are, in some cases, too many standards. If you look at the standards compendium that ONC puts together, you’ll see that there are several transaction types or data types, and there are multiple standards for those. Well, pick one. If there’s going to be a standard, let’s have one. There’s nobody holding a gun to any EHR vendor’s head saying you have to use this one, and so they could pick which ever one they’re good at or have had experience with, which may be absolutely contrary to what others in the industry are doing.
So now we have two different groups talking two different languages, and they can’t share with each other because of that. But if you peel back the layers of the onion, they’re still using a standard, and so it’s a standards-based transaction. People who know me think I’ve lost my mind because I don’t really think that the federal government should be telling us what to do; we’ve had enough of that with Meaningful Use. But I think that we as an industry need to come together and agree these are the standards we’re going to use; that there are going to be some winners, and some losers. Let’s concentrate on those and enhance those to take care of the use cases rather than having standard sets compete with each other.
The other thing is a couple of years ago I was at ONC and I invited a group of folks to come together to talk about creating a standard feedback loop so the industry can be aware of what standards are out there, who’s using them, and those types of things. All of the standards bodies were there and they told ONC, ‘we’re already doing this.’ They’re doing it individually, which is fine, but when you have a developer out there that doesn’t know there’s a standard for this code set because there’s no place they can go look it up, rather than do that, they’ll use their own. It could be done locally or it could be done regionally.
But now you have a system or a code or software that’s using a set of nonstandard descriptors rather than something that’s standard space. We’ve got some work to do in that. At ONC, Steve Posnack, is very much an expert, and he’s working on this. He has an industry he’s trying to work with that ONC can influence, but there’s no really overarching authority that says you must and will do this. And so I think there’s another way of doing it rather than doing the Big Brother thing. I think there is a compelling reason for the industry to do it on their own; we just need to go about that work.