Well connected in Indiana

Marc Overhage fills EHM in on the benefits of a statewide health information exchange.

“Doing more could be a negative factor, because it confuses everybody and they're not sure where to put their effort”
-Marc Overhage

Exchanging patient information across multiple provides overa a statewide rea is an incredible challenge on many levels. Marc Overhage, President and CEO of the Indiana Health Information Exchange (IHIE), says one of the main initial challenges for him was developing the value proposition; in other words, demonstrating that people do move between providers. He explains that many people didn't believe this actually happened when the IHIE was first started in the 1990s.

"They had this view that we're a large healthcare system and people come to our hospital and people come to our laboratory, and they don't go elsewhere. So that was the first challenge, at least demonstrating that there was a problem to be solved. The second was - not so much today but again in the 1990s - the technological challenges. We were some of the first folks to try and tackle the problems of patient matching, for example.

"Because we do not have a common patient identifier, as people move between providers, you have to have a way to link together their information. The third is provider matching, because doctors and providers in the United States also do not have a common identifier. As a primary care provider in Indianapolis, I have 47 unique identifiers, all of which need to be matched if you want to bring information together for quality improvement or simply to get a result delivered to a provider.

"The next thing on the list is privacy and security, and there we did a number of things, including technological approaches, but more importantly process and trust building.

"At the phase that we're in now, the challenges are around creating value. It is costly to bring this information together in a standardized format. In other words, every hospital, every laboratory calls a serum sodium something different. You have to normalize that in order to make the data truly useful. In order to support that, you need to find, in our view anyway, a variety of ways to help drive value out of that data and so create those sustainable business models, which is our big current challenge."

Good advice

When asked what counsel he would give to others contemplating setting up a statewide health information exchange network, Overhage's first response is to tell them not to build it from scratch if you have a choice. He continues: "The second thing is that you have to build incrementally and be patient. The only way you establish trust is by working out each use of the data in a very careful, thoughtful way, because the one thing you don't want is for anybody to be surprised about how their patient's data ends up being used.

"That's one of the key things, and related to that is being very patient, because healthcare organizations are relatively slow-moving beasts and you have to let things play out. There's just no quick way to move that down the road, as much as you'd like to."

Despite the slow speed at which things can move in the healthcare sector, there have been cases where organizations have managed to set up a health information exchange fairly quickly. Overhage gives as an example the MidSouth eHealth Alliance in Memphis, Tennessee, which took some of IHIE's agreements and technological approaches and went from nothing to operational in 18 months.

"They were far down the road with trust, and they'd done a somewhat limited set of things to start with," Overhage explains. "So it can be relatively fast. However, even  for established organizations it could take years, not because it's technologically challenging but because there are a lot of other things on their lists of tasks to get done and they've got other priorities, and maybe leadership turns over and it's not a quick process."

The IHIE recently joined with HealthLINC in Bloomington and HealthBridge in Cincinnati to form the nation's first live, multi-region clinical information exchange. According to Overhage, one of the main challenges in this project was dealing with the technological differences that existed between the three organizations involved.

"The second thing is the fact that while we have focused on the state of Indiana and HealthBridge has focused on Cincinnati, there's still a big overlap of several hundred thousand people in the southwestern part of the state who may go to Cincinnati to see a specialist even though they live in Indiana. That's just the way healthcare works.

"But the most important thing is that this is the first real live instantiation of the vision of a nationwide health information network: when a patient is receiving care in one market and has data in another market, the data move with the patient. This is for real, with real patients getting real care; not a demonstration project, not a show and tell."

National goal

There has been a lot of buzz around the current administration's goal of computerizing all patient health records within five years. Overhage, however, is quick to point out that this was an idea originally instituted by the Bush government as long ago as April 2004, when the Office of the National Coordinator was created and subsequently set a goal for the majority of Americans to have an interoperable electronic heath record by 2014.

Regardless of its origin, questions have been raised in some quarters about the feasibility of this goal. Overhage believes it depends on how you define 'electronic health record': "If you look at where most patients information is generated, it comes from laboratories, pharmacies and hospitals. There are transcription systems where a physician dictates a note and they're turned into a document. A great proportion of patients' data is already computerized.

"What's lacking are, number one, structured data from many physicians' practices - for example, what your blood pressure was when you went to see your doctor - and then the other issue is that this information is all in separate silos. A patient's data might live in six different systems, with radiology data at three different radiology centers, and he or she may have been to two different hospitals. With pharmacies, obviously there are competing pharmacy chains that might be spread across 12 locations.

"So while all that data is structured electronically, it's not linked together. That's why I think health information exchange is so critical, because it's how you pull those silos together, and in fact that's what we do. The vast majority of citizens in the state of Indiana already have an electronic health record that's fairly complete; there are things that are missing that you'd like to have, but it's starting to be pretty useful when you have lab, radiology, medications, hospital records and physician notes. It's not perfect, but I can do a much better job of taking care of the patient with that record in hand.

"In terms of the national goal, if it means every physician is going to be using an EMR by 2014, I don't believe that will happen. But if you say every citizen will have an electronic health record, that is feasible if we focus our energy right."

True identity

In contrast to many others working in the filed, Overhage does not believe that the lack of a single patient identifier is a hindrance to the development of a national electronic health record system. In fact, he goes so far as to say that even if one existed, it wouldn't help.

"In countries like the United Kingdom and New Zealand, where most people do have identifiers, it hasn't solved any of the fundamental problems," he asserts. "There are still data entry errors - roughly five percent of the health data numbers that are recorded in general practitioners' offices in the UK are wrong.

"And then you have the usual challenges of people who don't have one. The utility or value of that identifier is modest at best. There are very good statistical solutions. In other words, if I know your name and your date of birth and your gender and your Social Security number and where you lived last month, I can do a very good job of matching up your health data over time. In fact I can do that at the 99^th percentile level and make sure I don't incorrectly match anything, and that's all without a common identifier.

"That's not to say there's no value in having an identifier. There is value, but there are also costs and risks associated with that. I would say it's absolutely not essential and is not an impediment."

Looking ahead

In terms of the future for healthcare information technology, Overhage admits there is still a lot of work to do. He cites the need to build interfaces and normalize data, while pointing out that these are not technological challenges. "Those are things we've got to do and we know how to do them. We just have to get them done.

"I do think we have to guard against over-engineering and building in too much complexity. I believe we can do what we need to in the next five or 10 years with our existing standards, with our existing technology platforms, with our existing knowledge base. We don't need new technology. We don't need new standards. We need to take what we have and do the hard work to make them real.

"In fact, doing more could be a negative factor, because it confuses everybody and they're not sure where to put their effort and they don't know if Betamax or VHS is going to win. So they either sit on their hands or they lobby for one or the other and we get all this noise and confusion when we could be getting real work done that helps patients.

"The most important thing we need to do is connect local healthcare: the hospitals and labs and pharmacies within a market or region. In my mind, the value of a nationwide health information network is in dealing with the national overlays. It's the care systems like Kaiser Permanente or the VA or the national laboratories or the national pharmacy chains that need a common way to connect to different markets.

"The last mile part that we're missing today is market by market, whether it's Cincinnati or the state of Indiana or the city of Chicago. The need to have the various healthcare enterprises connected is really where the work and the focus needs to be."

Marc Overhage is President and CEO of the Indiana Health Information Exchange.

Marc Overhage on the Indiana Health Information Exchange:

"The Indiana Health Information Exchange was created five years ago as a response to the Regenstreif Institute, a research organization that had been developing software and evaluating the value of health information exchange for about 10 years. We realized that we needed to create a vehicle for sustaining that effort, not as a research project but as a service that folks could rely on and build on.

Our mission is the usual 'motherhood and apple pie': to improve the quality, safety, and efficiency of care, to be a model that others can look at, and then to facilitate research into the areas of healthcare informatics.

"The coalition is a fairly broad one. There are representations of providers, including physicians, hospitals and public health. Payers are represented. Business entities are represented. Research and medical education are represented."

IHIE services

The Indiana Health Information Exchange (IHIE), based in Indianapolis, provides an interoperable, standards-based health information infrastructure to directly address the lack of access and coordination of clinical information that can result in errors, misdiagnoses, patient safety issues and cost inefficiencies.

By bridging the gap between paper-based and electronic-based medical offices, IHIE has created a secure network that can be used by physicians who have IT systems and those who do not have IT systems. This provides reach to even the small or rural physician practices - the setting where over 80 percent of care is delivered and the places least likely to have adopted an electronic medical records system.

By delivering clinical information at the most critical time, the point of care, IHIE's goal is to align transparency, efficiency and quality to improve patient health.

DOCS4DOCS

IHIE's DOCS4DOCS service provides health information in near real time, where and when it needs to be for patient care (to emergency departments, outpatient centers and ambulatory practices). Since 2004, more 50 million test results and other clinical information have been delivered to physicians.

Quality Health First

The Quality Health First program is made possible through IHIE's partnership with the Regenstrief Institute, Inc., through the Indiana Network for Patient Care (INPC), which powers the data used in the QHF program reports. The goal of the QHF program is that patients will experience fewer health complications and physicians will see better adherence to evidence-based medical practices.