Since John Irvine’s taken over as the business lead for THCB, we now have a raft of new sponsors including CDW, Silverlink and now Orion Health. Apparently the marketing folks at Orion thought that it would be a good idea for me to interview their Chief Medical Officer Chris Hobson. What I didn’t realize is that Chris is a wealth of knowledge about health care systems around the world, and in particular how EMR use became prevalent–yup that essentially 100% adopted–in New Zealand We had a very interesting conversation about that, and if you’re as interested in that conundrum as you ought to be (which is very!) you need to read this. And hopefully the marketing people wont be too upset that their CMO barely got close to the topic of what OrionHealth actually does!
Matthew Holt: This is Matthew Holt at The Health Care Blog and today I’m doing another podcast. And with me I have Chris Hobson. Chris is the Chief Medical Officer at Orion Health. I’m very happy to be talking to anyone at Orion Health because unbeknownst to me last week they have decided to become a sponsor of The Health Care Blog. As part of that arrangement, I’m very delighted to interview Chris because I interview a lot of people who do not sponsor me. [laughter] Anyway, Chris, good morning. Thank you very much for joining me. Thanks to Orion. I don’t know who was it who your organization who decided to do this as I am no longer the business rep of the Health Care Blog but delighted to talk to you.
Chris Hobson: Good morning. It’s nice to be here.
Matthew: I sense by your accent you are one of these American immigrants. Well, you’re in Canada, right?
Chris: Yes, actually I’m native from New Zealand. Orion Health, actually, we started in New Zealand. As we’ve grown from New Zealand across UK, Canada, Australia, and other parts in the US, I’ve sort of tagged along with the company.
Matthew: That’s great. Well, it’s always good for people to come from the rest of the English speaking world to tell the North Americans how to do it. I’ve been doing it for years, not that anyone is listening.
Matthew: Let’s talk a bit about that. There are a couple of things that Orion does. For those people who don’t know about Orion, and you’ll explain it better than I do, loosely you’re in the business of improving data communications and data integrations and that ends up being a lot around messaging and interoperability issue–currently a big picture problem in the US. But also elsewhere. Let me ask you to start with a couple of things. First off let’s talk a bit about what you perceive to be the big problem in the US in that sector. Because you guys are also in a lot of other countries, you mentioned New Zealand, UK, Canada and also some other European countries, give me a sense of. Is this the same problem everywhere in health systems or is the US unique?
Chris: Sure. Well, we have our own perspective on what’s wrong with the health care system, but I guess, there would be fairly few people who would disagree a major problem with health care is the fragmented nature of it. There are a lot of different people, all well intentioned, doing a lot of good things across that the patients may interact with. The problems that we see arise as the result of all this—Going on from fragmented system to provider-to-provider-to-provider. In particular the information does not move along with the patient and it is very easy for the provider to focus on a narrow area and miss the big picture for what is going on with the patient.In the US, health care is more fragmented probably than anywhere else. In the sense that there are six thousand hospitals and just a huge range of both providers and payers. And if you look outside of the US, health care is not as fragmented; however, it is still quite fragmented and the same problems do arise.So if look for instance, the classic kind of story comes where down to and I’m taking this case from Don Berwick, and so I hope that’s OK but–
Matthew: [laughter] We steal from Don Berwick all the time.
Chris: [laughter] You steal from Don Berwick all the time. That’s great. He’s a Harvard professor of pediatrics, and his wife developed an obscure neurological complaint. It took some nearly six to twelve months before she got better more or less as a result, or not as a result, of the health care system. Along the way, she saw a huge range of different professionals who were all trying, well intentionedly trying, to help. The problem from his perspective was that each time they went to see a new professional he had to remember the case history. Each time they would see another professional, they would ask, "Tell me about what’s been going on." sort of thing. Of course, he had told the story so many times and been questioned about it so often that by the time he got to about 10 days or 10 weeks of this it was very hard to be strictly correct or accurate even with the best intentions.Another case in a sense that may be described as another sort of sense from when I was working at South Auckland in New Zealand, we went out and visited a home visiting nurse. And the first thing she said as she went in to visit a patient, she said, "Everything you tell me will be kept completely confidential, and I won’t share it with anyone." She then proceeded to take the whole history about what had been going on, and the patient had diabetes and had been to see a primary care practitioner who had said, "You have diabetes." But of course, the patient didn’t like that message particularly. So, he did nothing and a few months later he still wasn’t feeling very well, and went to see another practitioner who said ‘You’ve got diabetes’, and he didn’t like that story either. But eventually he managed to end up in the emergency room in the hospital and they looked up the history and said ‘You’ve got diabetes’. And they operated on the patient and then sent him out into the community. So when the community visiting nurse went to see the patient to look at the ulcer on the leg and dressings she knew nothing of all of this. Even though she worked for the same hospital and the same surgeons had done the surgery they hadn’t communicated onto the next provider what needed to be done.Now let’s rewrite that script and go back and say. The patient goes to see the first GP and he says you’ve got diabetes and the patient doesn’t like hearing that news so he goes to see the second practitioner. When he goes to the second practitioner, this time the stories different because he says ‘hey you’ve got diabetes, and by the way that first doctor that you didn’t like for telling you that you’ve got diabetes. He was right. You’ve got diabetes and now two of us have told you and you need to take this seriously.’ Let’s imagine the patient still did nothing and ended up in the hospital. The hospital specialist will say ‘You’ve got diabetes, and by the way, all these other people who’ve been telling you the same thing, it’s about time you started to take note of it.’So that’s a short vignette on what we see as the biggest problem in health care from our perspective. I mean health care is full of problems, but the information continuity and sharing of information, and sharing it in a way that improves the quality of care and re-enforces messages to the patient and it’s consistent. We see the lack of that as a huge barrier to improving the quality of care.
Matthew: Now that strikes me as being particularly apparent in countries where there isn’t free flow of electronic data. And I’d like to just talk about this internationally for a second, because I was reminded, we look at the U.K. and New Zealand and certain other countries have much higher rates of physician computer use by primary care doctors, partly because the GP’s are all essentially employed by the government. The government said you will start using this thing, and there was actually a couple of interviews I’ve done, with Dr. Jim Morrow who’s this doctor that believes the government should mandate it here too. You’ve got that situation where there should be these primary care records. I know in England it’s changing a little bit with the change in the national spine, but in the UK those records are intended to be fairly concentrated. They’re kind of stuck in that one practice, but quite often patients don’t move between those practices as much as they do here. But here you’ve got not only the problem of patients moving between different providers, like the one you described in New Zealand. I know in the UK that’s kind of hard to do, but maybe in New Zealand that’s easier. Here it happens all the time; a Medicare patient will go to see several different specialists and none of them will know what the other one did.But also here there really isn’t that much recording of electronic information, at least in the clinical sense in the out patient setting, so my question is. You said what would be the desired outcome. It would be great if I went to the next doctor and he knew what the previous doctor had done before him, diagnosed how to treat the other disease. But here we’re trying to make all these work arounds about how to get that data. We’re trying to get it out of payer systems and we’re trying to figure out if we can get it out of the administrative system with the claim submission form from the data. We don’t typically have electronic medical records that can easily be shared or even a more basic record that can be messaged around with the patient. People are talking now about putting together personal health records that can be held by the patient; on a disk drive.What would you all sense be as the current best workaround solution to that problem in the US now? And we’ll talk a little bit later about where you think we’re going. But can you actually describe what you think we should be doing about it now?
Chris: I’ll just rewind just a little bit and then come forward again to exactly the questions you’re asking. Just to clarify slightly, so as you’re saying in the UK it was more as a government mandated thing and the GPs had to take the system, but in New Zealand there was no government mandate that said ‘you must use this system.’ There were some government carrots, over the years from time to time they chucked a little bit of money at it. Here’s some money to help you buy computers but New Zealand managed to get to virtually 100% of 3,000 GP’s, 100% of them using computers; more by, I suppose, a carrot than a stick.And I think in one of the pieces of the puzzle why that worked, why the uptake happened, was because of an Orion subsidiary company, which is HealthLink. We already had the messaging technology which made it very easy to message electronically things such as discharges, referrals, lab results, status messages, the whole nine yards, all of the radiology images and reports; all of those can be messaged around the system. So because New Zealand had a good network it became sort of like a FAX machine problem–if you are the first person with a FAX machine, yeah, you know the one. So in New Zealand we managed to put together a good network with a lot of voluntary uptake, and then it took off because it became so much of an easy way to do things.Now, OK, so roll that forward to the U.S. and say, "Well, what can we do?" And obviously there is a lot of push on physicians to take up electronic medical records. But I’m not aware of how easy it is within the U.S. system for a physician to, say, send a referral or receive a discharge or receive a laboratory result. It seems that that varies widely.
Matthew: Yeah, that’s true.
Chris: Yeah. So our sort of take on all of this is two things: One is to keep putting together that backbone network so that eventually, like with the FAX machine or the ATM machines with the banks, eventually you will get some common standards and you can go to any ATM machine and get your data and your money; it doesn’t matter where it is, it’s all interoperable. So we do think it is really important to keep promoting that message, and hopefully uptake will start to happen. Now we have another partial solution to this problem, and that is, because part of the problem with taking on an EMR is the cost.
Chris: And so the cost engine is a lot less for a GP who wants to take on an EMR. The cost to achieve is affordable. When I was in primary care practice I paid $8, 000 or whatever, I had a system, and thereafter I paid a small amount per month. That was it for years. In the US the prices are much steeper for the EMR. So one answer we have to that is we provide software with web browser access, so if the physician at least has a computer and can link to a network, then he can access in just a web browser–a very thin client, no requirement to have lots of technology on his desktop–he can actually see through that web browser, see the relevant data. So that is another way to make it a bit easier to start to get the uptake.
Matthew: Let’s just talk about that for a minute.That actually may be one of the key differentiators. And I don’t know about what the New Zealand system is. But there has kind of been this suggestion–obviously this has been going back 20 years in U.S. healthcare—should you be trying to put sort of thin-clients or pieces of technology which help physicians look at the data that is flying by from labs and from hospitals and discharges, and not worry too much about them generating their own clinical notes in their own practice, and then take that data from elsewhere and just make it available.Or should you be trying to do more of an approach that is typically the mainstream US approach–which is in large groups where you create a large EMR in the physician practice and then start trying to figure out, after you have taken the clinical notes and all the other pieces within the context of the physician workflow, you then make that available later outside.Do you come down on one side or the other on that?
Chris: Sure. The thing is, a physician has got to get value from accessing the system and he has got to get that value quickly, he achieves that value quickly, yes. So going back a few years in New Zealand, so when I was working at South Auckland we had 50% uptake of the system. And the way we got physicians using it more was we put clinical data down the same pipe as the claiming data. So they were claiming electronically, so we just sent lab results or discharge letters down the exact same mechanism. So without the physicians doing anything, more or less just making their claims, which is something they are always–even in New Zealand they worry about money. So they have got to make their claims; and hey, by the way, look, there is the patient’s lab results or the patient’s discharge, it just appears.Same with the web browser approach, if you can make it really easy then that gets you over a first hurdle. When we can implement these web browser based systems both across a region and within a hospital. So same sort of technologies: Web enabled in hospital clinical record or a web enabled record that goes across a whole region or state or whatever. So for instance working in the state of Vermont. Coming back to within the hospital, if you provide a web browser where all the physician has to do is log on and he’s got data that’s useful to him or her, that does a lot for the uptake. You can use that as a way of implementing systems, because there’s so many hurdles in getting to that full blown, fully EMR system that if you can take things little by little, step by step at a time, it makes things a lot easier and much more likely that you would succeed. So a physician knows he just gets to log on and an icon on his desktop that will give him lab results and documents and the emergency room situation and maybe some disease management or some medication histories on the patient. And he doesn’t have to do anything to get that except look at it. That way you get a first go where your chances are high. You get a lot of physician uptake and the project won’t be complex or mission critical until it’s working well. And subsequently you can start to build EMR type features where the physicians have to interact and start typing or entering orders and results.
Matthew: So what’s your sense about the flow from one to the other. I mean take the New Zealand case. How long did it take. I know from the UK system and I assume that the NZ system is similar that there’s a lot of back and forth between the primary care docs and the specialists who are hospital based about appointments and patient information flowing back and forth. So we have a similar issue here, it’s just more multifaceted. So how long did it take before most GPs were getting the lab data and discharge data and other stuff and then made that switch to actually saying: Well OK, now I need to start entering information I’m recording here, prescriptions I’m giving out, orders I’m actually making. Putting those into the system as well? How long would that take?
Chris: Well within a hospital you’re looking at maybe – I’m thinking about a hospital in New Zealand with lets say 1500 physician and nurse users. Getting from just viewing to getting them to putting data in is a ramp up process of I’d say two years.
Matthew: But in New Zealand those physicians who are hospital based are working with the hospital. They’re not actually seeing patients out of the hospital.
Chris: That’s correct. But the biggest problem that you face is that the physicians can still say no. So if you put up some technology that isn’t at least usable and reasonably physician aware, they can reject it. And there’s a number of high profile cases in every country in the world that I know of, and certainly in New Zealand and the UK, where the physicians have rejected the technology.
Matthew: I didn’t know there were high profile cases in New Zealand. In the UK there are several right at the moment, and of course there’s a bunch of others in the US that are going down in folklore.
Chris: And there are some cases like that in New Zealand too. And physicians will just decide: Look I can’t use it. It’s too hard. You’re asking too much of me. I’m busy. And I can’t use it. And in that situation, even in New Zealand where they are employed by the hospital, they can still say no. So you have to mix as much carrot and a little bit of stick as you can. But we would hope now – when I’m talking two years for a 1500 physician hospital – we would hope now that given what we know further, we could do that more quickly.
Matthew: And then going out to the GPs and the PCs out in the community. How long did it take them to start figuring out ‘Oh I can go on-line and get this data.’ So do I have to buy new systems, is that within part of the system?
Chris: Yes, so the financing of these kind of things is a big issue. So in New Zealand we took maybe six years to get to 3000 physicians. And along the way various vendors wanted to charge exorbitant amounts of money to connect to the network. And the physicians didn’t want to have to pay. And the same sort of issues. But as you get more uptake, and if people can see value, then they are more willing to come onboard. As you get volume the price comes down.
Matthew: In those six years, is that six years to get from sort of a small percentage…
Chris: From zero.
Matthew: to 100% on the network. What about the next piece, where I know on the network I am getting lab results and seeing discharge. I’m now going to start entering my own data.
Chris: Oh no. That would include all of that stuff. That’s the whole shooting match.
Matthew: All right. Good enough. Let’s talk a bit about Orion and your role here and maybe other people’s role before we came into the States and elsewhere. When you did that, they are on a network that you guys have laid out. They are seeing that you are putting your box in the labs and in the discharge part.
Chris: Yes, that’s correct.
Matthew: Are you now, in old Web 1.0 terms, ASPing the clinical software, or do they have to install it at their own site? Are you getting a mix of different vendors that are all going to the same network and physicians? How did that work out in New Zealand?
Chris: Right. In New Zealand, there are seven different EMR vendors. Given that each one of them has different versions, so maybe up to 10 different EMR entities to integrate. Now realize, on a scale, that’s a much lower scale than the US. That’s the world that we are talking about.Within the hospitals we have integrated existing applications. So certainly in the New Zealand model there is not really anything that really is ASP.
Matthew: Then there is the network that the GPs can connect to. They are now saying, ‘OK. I’m now going to put $3, 000 or $8,000.’ New Zealand dollars or American dollars –
Chris: That. A long time ago too.
Matthew: They are buying the system and off they go and they are starting to use it. You know, that pretty much 100% of them are using it, which is an incredible ratio for Americans to understand. You’re saying the integration of those different systems into the network wasn’t that hard, because there were relatively few of them.
Chris: That’s right.
Matthew: On the other hand, here we have standards bodies, and I don’t know if we’ll have problems with this. You would think that the integration part should be manageable here but it’s more the uptake that you’ve got a problem with.
Chris: Yes. I think the integration part does seem to be a problem as well. This is perhaps a personal view. So in the US system there is a huge amount of effort going into developing standards for interoperability and so on. I’m speaking personally. I think there’s nothing wrong with that effort. It needs to be done and I appreciate the importance of it. But from a personal point of view, and from a perspective of how we make things work, not just in New Zealand here, but also in Australia, some projects in the UK and Canada, we are talking about getting ahead pragmatically in making something work. I think you have to have, I can see this is a top-down and bottom-up approach. So from the top down, you have to have these big standards bodies and committees and these are devising standards and that is good work. But from the bottom up, you have to have maybe a small state – say, for example, we are working in the state of Vermont at the moment. It is relatively small. People just have to sit down and nut it out and work together and take a pragmatic approach. Often, you can make things work, without necessarily recourse to huge numbers of standards.You get something up and running. You get physician adoption. You get the improvements from the clinical perspective. At some level, you can always adopt further standards as they come along, because the standards do keep changing. So this is my personal perspective. As a clinician, I don’t see interoperability and all of that as an end in itself. I want to do something with that inter-operable network. I can exchange lab results and discharges, but I want to be able to do more than just that. I want to be able to fully manage my patient. I want to be able to improve the standard of care. And I want to use that information to improve patient safety, to increase conformance with guidelines, to ensure that patients are on the right medications and so on. So I’m a little bit impatient with the large standards bodies’ approach where it is taking forever to come up with interoperable standards.
But I am part of that effort, and I think it is really important. I highlight the approach of the certification commission for health IT where they are taking a very pragmatic approach. This is a standard that a vendor could reasonably be expected to implement by May 2007 or May 2008. I am very keen for people to take something that they know is reasonably defined and make it work.
Matthew: OK. Let me go back a second to that adoption of the EHR Clinical Workflow function by the office because actually, you just twigged something there that I know I have seen in a couple of cases. One is that I have had American doctors or other sources actually – some of the health plans — say to me, ‘Yes, I was able to know, who was on Vioxx or whatever drug it was that was recalled, and so we can email people or we can contact patients immediately.’ I was doing some work in the UK about five years ago now trying to figure out which patients were on a certain drug. I was in a GP practice. They went over and within 30 seconds of looking on his system the doctor told me, ‘Oh yes. We have so many patients with this disease. And of them, that many are on this drug. And this many are on the following drug.’ And you are thinking, we know that American paper based clinicians can’t do that. And there is a strong argument against putting up this kind of system even though Newt Gingrich says a paper prescription is malpractice and Jim Morrow was saying that you can’t be a great doctor if you aren’t practicing with one of these systems.What was it that got the New Zealand physicians over the hump from looking at the other stuff coming from these hospitals to actually inputting and using this stuff to manage their own patients? What was it that pushed them over the hump?
Chris: I think, speaking specifically about the New Zealand situation, there were a lot of keen, enthusiastic family practitioners. There was a coordinated government initiative, trying to drive these things. The government did some things like the Single Identifier, which is a whole other issue because everyone in New Zealand has a Single Identifier that so enables some of this stuff. So the government was driving it in one direction, but I think a lot of it did come from the physicians. My own personal experiences, exactly as you say, that I was a much better doctor when I had the computer in front of me. So I well remember practicing, where I had my own EMR in my own practice, that I could see a large number of patients in the day and feel pretty confident everything was pretty under control. I could go locum or working for some other physicians from time to time in a paper based practice and just find it a nightmare.So somehow the physicians have got to see and accept that they will be better doctors if they have quality EMR that is linked into the entire network. I think that realization dawned on the physicians in New Zealand 10 or 15 years ago. There are other examples in other countries. Denmark is another one that comes to light, where the physicians have also taken up these systems to 100%. So yeah, on one level, it has to come from the physicians. The light bulb has to want to change in order for this to happen, yeah.
Matthew: Yeah. Three of the countries, which are the most advanced, are New Zealand, Norway and the Netherlands. If we change the country’s name to the United States of North America we will get it done very quickly.
Chris: [laughs] Yes. That’s the one.
Matthew: Now let’s talk a little bit about the American market for interoperability, for want of a better word. It seems to me that there are two major areas here—I don’t know a lot about what Orion has been doing—one of them seems to be do you go after the internal provider problem, which is you have a big hospital which has a gazillion different systems already that needs to get them to talk to reach other. And there is a whole bunch of issues around that.
The other one is, do you go after the RHIO issue of trying to connect different provider organizations in one geographic area?I’ll give you my bias. It seems to me that we have 6, 000 hospitals, but actually that’s probably a couple of thousand ‘systems’. Some of them have just got one hospital. Some of them have got two or three or four or even more all of which have got lots of different systems that need some kind of integration. There is a whole piece around that. It seems to me that’s a huge market for a company like yours to go after.And then you have these RHIO’s that have got a less certain business future, but obviously still have an opportunity there. And there are some that look very encouraging. You’re not a huge company. Where are you guys spending your efforts in the US?
Chris: Actually, it’s at both levels. So we have examples of large prominent healthcare systems that are choosing to adopt our portal solution. And we have examples of whole states and RHIO’s. So while I am talking about the discussion that we are having here, at one level we can integrate the data between systems. Sending a message from a GP to a hospital is a message-based communication. Similarly, sending an ADT message, and admission message to a lab system to tell the lab system that Matthew Holt has just been admitted to the hospital is a message-based backbone. We have software that does that. We have software that can connect any number of different types of databases with back end and messaging type connections. If you take that idea of integration to another level and we are talking about the clinical portal, the idea of the clinical portal is, from the end-user perspective, a new thing. This is like Google, Yahoo, and MSN. These are all portals. You can set up your Google home page or your MSN home page with the weather.
Matthew: Yeah, or with stock quotes or whatever.
Chris: Sorry, yeah, you can put all these things on the front page. The way we set this up for the clinicians is they have a home page where they can pull everything together. Say if I am a microbiologist, I might want only the microbiology labs and some other particular types of data. I put those on my home page, whereas if I am a generalist, I want to pull together data from all over the show. So this sort of unifying view with the portal technology we can apply medical applications portal over say one hospital, or we can put it over the whole region.
So some good examples of this, if we can talk about Canada at the moment, we have been working in the province of Alberta. We put a portal across five hospitals in Edmonton. The city of Edmonton has five hospitals. We linked together all of the data in all of the various hospital systems. We pulled back together and made that a system known as Netcare, which enables clinicians working within any of those hospitals to access any of the data on the hospital network. Then we subsequently took that to a higher level. Using the same software we were actually able to provide data from across the whole province from a front-end perspective. We are actually working at both hospital level and a statewide or a RHIO level.
Matthew: In terms of your business, let’s take a more typical American case. So I’m a physician. I admit to maybe one or two hospitals. I have a bunch of patients, who I am managing out in the community, who probably don’t go to hospital much, but I have data of my own. I have data coming from random labs and imaging centers, which may or may not be connected to a hospital. And then there is other information flying around. Does your business in the US go after the hospital players? Does it go after the physicians? Who is your end customer? Which product are you providing?
Chris: OK. From the integration and the portal perspective, we are looking a lot at individual hospitals or hospital networks, where we are pulling together data from multiple systems. That’s certainly a key market for us. Our ability is to integrate data and present it through a single sign-on, but also with a single unified view of the data, for the physician. So that is one big piece of work that we feel our particular portal is potentially good at doing. If you are looking as you say, in a community where you have sources of data that come from all over, even if I am a physician in Vermont and I know I have some specific labs that are coming from Mayo Clinic or whatever. The approach that we take is to try and provide the broadest depth of information available in the portal. What we can do is incrementally add more and more. So you may never get to the perfect. That’s the pragmatic thing. We are offering the physician something. It will get them going. And as time goes on, we will add more data sources.
Matthew: I get the business from the hospital about making the stuff available. If you look at that market the big HIT vendors, the McKessons of the world—I think McKesson has more portals out there than anybody else—there are a lot of other players in that market, who think, ‘OK, we need to make data accessible to our business partner physicians. So we’ll put out a portal system.’ Then there is the customer and they are paying the vendor for the license. Is there a separate market where you are charging physicians for access to this other stream of data or is it always going to come via the local hospital? Who is going to pay for this is what I’m getting at?
Chris: Yeah, exactly. And further to that is the sustainability of the RHIOs.
Matthew: Well that’s another point we should get into. But let’s talk about when you are out there; somebody has to write a check. While they’re at it they can write me one. [laughs] You’re seeing hospitals. They are having problems with their own business partners. They want to connect to their physicians. They want to make sure the physician sees what’s going on there so they don’t take their patients down the street. Is there another source of dollars, or is that it?
Chris: A hospital is a major source of dollars obviously. From a hospital’s perspective putting in the portal is a cheap, reliable, easy way to significantly upgrade the quality of IT that their physicians are accessing. So generally, whether it’s our particular portal or a competitor, it’s still a relatively easy business case to make for hospital to buy a portal, because it saves so much grief in terms of changing systems. It puts a nice face on it. Then the IT department can later change what’s on the backend. An extension of that in a highly competitive environment may well be that the hospital sees the ability to make their data available to physicians in a community. So a RHIO, if you’d like, or an extension of a hospital to the surrounding physicians so that they are able to see data and access it and do things with it, maybe another part of the business case for the hospital to pay for. When you get up to regional levels, simply exchanging the data and making it available, may well not be enough to convince anyone to write a check. This is where we want to take the whole inter-operability story to a third level. We have talked about backing up integration of data. Then we have talked about confronting with the user, the physician accessing through a portal. Then, at another level, from a statewide level for instance, you have business imperatives that any particular state may see as justifying this expenditure on an inter-operability basis.This is where the real money is. So for instance, in Vermont, the governor of the state of Vermont has a strategy around disease management. He is not necessarily or particularly, at least initially thinking so much about interoperability. He’s thinking about, ‘I’ve got all these patients with diabetes, heart failure and various medical conditions that are not being properly managed.’ There is so much evidence about this. ‘I’ve got all these patients, who, through Medicare and Medicaid are using a whole lot of medications that may or may not be appropriate or may not be the best value for money. How can I do something to improve the overall standard of care delivery?’ So, in Vermont, we linked the development of the RHIO, which were largely providers working together initially with the Vermont IT leaders with the Vermont Department of health and their initiative around disease management. What we are saying is, with what we do, we can provide a much better disease management program if we have complete data about the patient. So, we use that kind of argument that you can build a business case to justify statewide or departments of health, and people at a higher level to actually invest in some technology.
Matthew: That’s pretty interesting. There is no question that everything you said is completely true. We have a massive problem with managing public budgets and private budgets being essentially wasted in the poor management of these core disease groups. We have had a real struggle to figure out what to do about it. You could argue that the RHIO if used properly, and if the data from the RHIO is accurate, and you’re able to use it to manage these patients, that would make a huge difference. On the other hand, the benefits and payoffs are somewhere down the street. It’s difficult to see if really there is going to be a significant business in doing that and providing the technology for that in the future. But what you are essentially arguing is that there is. And that is the place that you guys are going.
Chris: Right. Exactly. So, to expand it a little bit, and this is more from a clinical perspective as a Chief Medical Officer is more where I come in, we provide further applications that depend on the ability to integrate, for those applications to work. So we have a disease management application, medications reconciliation, order-entry applications, which are intended to have broad access to data in order to work properly. Talk about this for a minute and think around it for a second: diabetes. We’re talking with a diabetic patient who went to multiple different providers, then the hospital, then the community care. What we can do is, in the standard view of disease management that the GP, or primary care practitioner, sitting in his EMR, he’s thinking, "OK, I want all my diabetics to have an annual check. Maybe they should have a foot check, and they should have their A1C". So there are some well-known, standard things that any physician would say, "OK, I know when I manage my diabetics, I’ve got to do these three or four basic things: I’ve got to do their blood pressure, their cholesterol, I’ve got to have their feet and eyes checked. And I should ideally, if I’m a good doctor, I should do that annually."
But then the problem is, if the patient comes to see you, but you only see them once, and then they disappear. So one of the things I used to do was print long recall lists, and I would get this great, long list of patients printed that I needed to recall to have their foot check or their eye check. If you can do it, you find them up, and they say, "Oh, no. I’ve gone to some other doctor, " or the patient doesn’t live there anymore, and so on. If you take a regional view, rather than just that individual primary care practitioner, you can manage that whole population so much better. So what we’re saying is we can use the integrated data to drive initiatives like disease management.An example, then, would be, if a patient came to see me, I did maybe the foot check and some bloods but never saw them again. They lived across town, saw another practitioner. What we’re able to do is distribute the work that needs to be done for the patient and make a view of that guideline accessible anywhere. Another example, maybe, to try to make this a bit clearer: a child has immunizations. A child is born, and then they’re expected to have two-month, four-month, six-month series of immunizations. If the child has their two-month and four-month immunizations with one practitioner, and maybe they’re staying with Grandma in another town, she takes them along to the doctor and says, "I want you to immunize the child."If the practitioner has only got a paper record, he really doesn’t know what’s going on and what immunization is due. If he’s able to access, online, a record of that child and the immunizations they’ve already had and the immunizations they need to have, he’s able to give them the right immunization at the right time and the right place. So what we’re doing is, by providing access to data across large areas–integrated data that’s available from multiple different providers–we can also distribute the guidelines and the expected standards of care. And if you ask your practitioner, "What do you really want when the hospital sends you a discharge letter?" or "What do you really want when you get data?" at the end of the day, the physician will say, "I just want to know what I’ve got to do now. Just tell me what I’ve got to do, and I’ll do it." At least, that’s a common response. And so if the patient who’s diabetic turns up, you’ve never seen him before, but you go online and it says, "This patient needs a foot check, " then, "Fine, I’ll do the foot check." I’ve built an integrated work flow. I’ve contributed to the total picture.
Matthew: Obviously, we’ll say something about Vermont in a second, but is your sense that there is an appetite for public sector organizations in the US do that kind of overseer role? Because you’re talking about putting in the technology, allowing the private physician access to plug into it, and then also, somebody at some point is going to have to point out…I don’t know if it’s just a data exchange between the key players, or if someone’s going to say, "OK, we’ve got this section of the diabetic population which hasn’t had a foot exam, " or whatever, and go after them.
Matthew: Is that what you’re hearing in Vermont?
Chris: Yeah, I think so, definitely. Obviously, in Vermont, the state, the governor, and the Department of Health have been terrific. There’s just fantastic strength and support behind the whole initiative in linking the disease management to the Rio and making things happen. To be fair, it’s possibly easier because it’s a state of 600, 000 people. Obviously, it’s not Texas or California or New York. So our view on that: you have to choose the right size. And we used to talk about this a lot in New Zealand. It’s a great idea if get the local physicians working with the local hospital, and everyone sits around and they say, "Yeah, that’s great!" [laughs] And the next thing you know, "Oh, why don’t we bring in the ones from the west and the south?" And pretty soon, "In fact, why don’t we do the whole city?" And then, "Well, actually, let’s not stop at that. Let’s do the state and the country." And pretty soon, you’re trying to boil the Pacific Ocean…You’ve got a big problem on your hands. So we sort of see the optimal size as being whatever the political, as well as the geographic realities and social realities you have to deal with.So the disease management concepts also, really apply to any payer or funding organization, which could be the state, if the state’s paying for it, but it could also be any one of the payers. Medicare and Medicaid also fall into this group.
Matthew: Do they have the private payers on-board in Vermont?
Chris: Actually, I can’t answer that question. [laughs]
Matthew: [laughs] All right.
Chris: Bottom line: I don’t know. We’ve been working with the Department of Health.
Matthew: Right. But, in actual terms, tell me how far have you gotten so far? What’s up and running there, and where are you in the planning stage? What’s happening? And this will vary across the country, of course.
Chris: Yes. Yes, it does. In terms of Vermont, we’re looking at a project that’s really underway, from a technical perspective. From a sort of business-level perspective, this has been a number of years in the making, so they have now gotten good alignment of physicians and providers, and everybody’s happy to work together, at least in the general sense. From our perspective, we have software that we’re busy installing at the moment, and we’re looking at pilots in the next few months. So it’s pretty real.
Matthew: I see. So it’s coming up in the final stages, and then we’ll see the continuing of a roll-out plan from there.
Chris: Yes. Yes.
Matthew: I’m putting you on the spot, but what’s your guess about when the vast majority of players in Vermont will be able to interact with each other, and patients will be able to get their records? Are you’re taking a 2-year process or is it really a 10-year process. [laughs]
Chris: Yes. Sure. Understood. From the point of view of any provider being able to access the system and access data that will be fairly quick, so we’re not looking at prolonged periods of time. Because it’s web-browser enabled, adding for the users is not a big deal. We’re looking in terms of the next three months or less, in terms of actually getting everything implemented so that that’s available. So from the front-end perspective, physicians can see data and can access the system and use it for some useful things: that’s relatively quick. Now, from the point of view of making accessible all of the data across the whole state, to the data sources, that will be an aggressive effort, and it may well take more than a year or two to get all the data sources feeding in.
Matthew: And then the intention is, at least, for some other body to be looking at and helping, as far as the disease management part, whether it’s the state or whomever else, is going to be coming in over time as well.
Chris: Yeah. The disease management we’re also putting in at the same time, and we’ll be doing pilots of that in the next few months as well. Perhaps it may be easier or harder, but let’s talk of a real example. So we talked about, in the province of Alberta, Canada, the Portal implementation across five hospitals–officially, that took 12 months. Now, from project start to being useful, we’re talking between four and six months. So some people are accessing it and doing some things. To get to five hospital-wide and an official public launch, that took 12 months. And we’re looking, there, at 10, 000 users. So it’s a big deployment.In terms of the disease management that took eight months, and it was done sequentially. So first they did the regional Portal, then they did the disease management, then, in terms of making it province-wide, it’s another 12 months, more or less, to make everything work and robust and you can go look at it anywhere in Alberta and see the data. So that gives you some idea.
Matthew: So what you’re saying is you’ve got the example now. It’s all up and running in Alberta?
Chris: Yes. Absolutely.
Matthew: So you not only could, if you’re a physician in Alberta, look up a patients hospital records. You could also… in the province… I assume that somebody is able to look at the records of all diabetics.
Chris: Yes. Now there’s ifs and buts of course, but at the end of the day the disease management is accessible by any of the providers who are involved in the disease management program. Actually the way it’s worked out, we have two disease management implementations. We have a disease management implementation in Edmonton and a separate one in Calgary. So that’s the sort of if and but. However, the Calgary disease management is city wide and also a lot of the surrounding area. So we’re looking at a population of well over a million people. We’re looking at 60,000 patients on the system and over 20 different disease management programs. In Calgary they’re more focused on the disease management. In Edmonton they’re more focused on the portal, and they can talk about their 10, 000 users and their disease management programs. I think they have two different programs and again several thousand patients. So it can be done.
Matthew: And I think that that’s kind of the interesting part, and we’ve kind of been going back and forth on this, but I can see why from your perspective. I’ve kind of in my mind said that there’s this integration of systems internally and then there’s this integration between systems. But you’re saying, really from a technical perspective, they’re kind of the same problems to be overcome. And when you mine into the U.S. market, and we’ll end by going abroad in a second, but when you mine in the U.S. market, are you spending more of your corporate effort going after these RHIO programs or are you spending more of your effort looking at the internal system?
Chris: So probably we are spending more effort overall on the internal hospital systems because hospitals can be complex environments. And you want to put a portal across a range of different vendors in every situation. To some extent every situation is different. As we’ve gotten more and more experience we can start to say yes if you’ve got X, Y, and Z systems we know we can implement them and do that nice and easily. But getting to the point where you have enough experience with integrating enough vendors – that takes time. And the internal of a hospital I think you’re talking more something that is mission critical. And so this puts a lot of effort being that there’s a lot of hospitals. And it’s more on a social, political, business type thing… it’s an easier thing to do. In terms of the RHIOs, the data that you’re putting together- you’re looking at lab results, you’re looking at documents – but although it’s the same functionality and technology, it’s not as complex as it is inside the hospital, even though in both cases you’re using the portal technology.
Matthew: So let me ask you one philosophical question. And you guys are working in different countries as well; Canada, the U.K… Are there some other European countries you’re also working in as well?
Chris: At the moment we’ve found in the European countries a lot of interest in the disease management. So Italy, the Netherlands, so yeah we are active in those markets.
Matthew: I’m only raising that because there’s kind of a philosophical argument on the policy at the moment. The issue is the politics; the policy of health care is coming back around in the U.S. People are starting to begin to talk about looking into these other countries – France, Japan – and in some ways both in policy and in IT, everyone says ‘Well you can’t do that here because America’s different’. But is that true? Or are all doctor’s and patients the same everywhere? And we should be able to figure it out technologically.
Chris: Right. Well from our perspective, and especially my perspective as a clinician, it’s not that different. The diseases are the same. The patients are pretty much the same. The physicians have been trained in much of the same way. So, the best way to manage heart failure is the same in America as it is in Canada or China. So clinically they’re not different. Now administratively, in terms of billing and items like that, the American system is unfortunately very very complex. And my perspective on that is that it’s needlessly so. That such a high proportion of a physician’s time is spent handling administrative difficulties. That’s not the way it works in other countries. And that doesn’t mean… it was the idea of HIPAA wasn’t it. The idea was to simplify some of the administrative work. But that’s the major difference. And with our approach in the U.S., from my perspective and generally speaking in terms of the company, is to go after the clinical value and try to say: look, there’s a clinical benefit here, patient safety, improving the management of medications or diseases and so on. There’s probably some cost reductions because you will be managing your patients more efficiently. There’ll be reduced duplication. There’ll be less time wasting. So you will get a lot of improvements. That applies whether it’s in Canada or the US or anywhere.
Matthew: Well I’m delighted to hear your optimism come out of this. I’ve been studying American health care for a long time going – Wow, this stuff is so complex. But I think that the point that is worth making, if you’re putting in the portals, the messaging, and you get enough of the systems up and running, connecting between them actually makes it much easier, once you got all those systems with the data flowing. And I assume that you’re coming at this from both ends. And there’s so much clearly clinical value to be gained in doing this, that you just have to look at the fact that we do three times the amount of heart surgery in different states.
Matthew: Nobody knows why. And we have all these poor records, and I think that it’s true in all countries; we could do some much better in managing our mistakes. It’s good to hear an optimist. [laughing]
Chris: I sympathize because we whipped this disease management solution in New Zealand in 2001. And we’ve been sitting there and looking at the U.S. system from afar and thinking ‘it will never sell.’ The Americans will never understand us. How do you make it work when you have multiple payers? The New York Times did a discussion recently about diabetes, it was January this year, they’re talking about preventative care for diabetics being stopped because it wasn’t in the hospitals interest. It was in the hospitals interest to amputate. And you just throw your hands up in horror when you read this sort of thing, cause that is just wrong headed. But the hospital gets 100 dollars for a foot check and they get 10, 000 or 20, 000 dollars for an amputation, so let’s go for the amputation. But I think the message that we’re trying to get with the disease management makes sense at a state wide level such as the department of health in Vermont. And it makes sense for payers, if the payers can be convinced to just see that benefit at the end of the day, big cost savings, then they can be encouraged to take on the disease management model too.
Matthew: Yeah and I think that you’ve come at the right time. The policy debate is not unconnected. If you look at California or Massachusetts or the Federal debate, people are starting to realize that the way we treat diabetics doesn’t make sense, and there’s got to be a better way.
I guess the answer is, I’ll have you back in, in six or twelve months, and I’ll figure out whether it works [laughing].
Chris: Thank you for inviting me along. I really appreciate that.