History of Health Information Exchange

This lecture focuses on Health Information Exchange by outlining the history of U.S. based attempts at getting health information exchanged among providers.  By understanding the history presented, learners will be able to better understand some of the current challenges and ways to overcome them.

The fundamental need underlying the justification for health information exchange is nicely depicted in this scenario.  Imagine you are vacationing across the country.  Say you’re from New York and you’re in California and are rushed to the hospital following an unfortunate car accident.  The doctor who treats you in California would greatly benefit from knowing more about your underlying health condition, prior diseases you may have, or chronic diseases that you have, what allergies to medications you have, what current drugs you’re on. Even if all of this information was securely stored in an electronic health record system in your primary care physician’s office in New York, there still needs to be a mechanism in which the California-based physician can access this information in order to provide you the highest quality of care.

This problem is compounded by the fact that patients typically have data stored in more than one location, not just their primary care physician’s office, but they may be cared for by a variety of other physician specialists. Every time they get a prescription they may have data stored at one or more pharmacies. They perhaps had experiences at different hospitals, including emergency departments that ultimately result in medical records being produced, and over a lifetime much data accumulates at a variety of different places all of which are stored in silos. Now, individual providers caring for the same patient even in a given neighborhood, perhaps even across the street from each other, as we’ll see in just a second  cannot easily access each other’s data.  Currently they rely on faxing or mailing of pertinent information (usually when it’s directly requested), which makes it difficult to access needed information in “real time,” like in the example of the vacationer from New York to California.

Above is a map view of two competing pharmacies in the same neighborhood.  As can be seen in the Google Maps direction, the two pharmacies are several hundred feet apart from each other - practically across the street from one another.

Currently, there is no infrastructure in place to transfer a prescription from one pharmacy to the other.  That is, if you’re the patient, and you want to transfer a prescription you have at one of these pharmacies- to the other- there is no mechanism to “beam” that prescription, if you will, from one place to the other.  Certainly, it is possible within a pharmacy that belongs to a national chain to get a prescription transferred to another pharmacy of  that same chain, but it is not currently possible to transfer between pharmacies in different chains.  The same is true for all other healthcare organizations -- whether pharmacies, or doctors’ offices or hospitals. The unfortunate truth is, is that data is not easily exchanged from one place to the other, even hospitals that belong to the same national chains, in most cases, cannot transfer information about patients from one location to another.

So in terms of a definition, the National Alliance for Health Information Technology defines Health Information Exchange, or HIE, as the process of sharing patient-level electronic health information, such as lab results, or medication lists,  between different organizations, such as hospitals, or physician offices, or pharmacies. Obviously much of the exchange will  occur locally, that is within a given community, because most people seek most of their care close to where they live.

Nevertheless, like we had with the scenario described previously of the vacationer from New York visiting California, there are many instances that will require HIE across community or state boundaries.

The current level of health information exchange in our country is not very automated. In fact when providers exchange information about  patients that they’re co-caring for it typically involves sending things in the mail or by fax. In some cases there may be a courier who delivers information about a lab specimen to and from the  doctor’s office to the laboratory.  In other cases the patients themselves are the ones  bringing x-rays from one doctor to another or perhaps hand delivering or dropping off a prescription that was written to them by a physician, and are now presenting it to the pharmacist.

Ultimately, the goal (which is not yet realized) would be to have a system in place that facilitates the inter-connectivity of providers in communities and across communities so that patient information can be shared when appropriate.  The exchange of data on such a network would improve the quality of care by making information available in real time- where it’s needed; as well as improve efficiencies associated with the duplication of efforts when you don’t have a test result and as a result have to reorder it so that you can figure out what to do as a clinician, and also reduce the problem of inaccessible information which then results in sub-optimal decision making.

The National Health Information Network which, as mentioned, is not yet realized, but in early stages of development, is a collection of standards, protocols, legal agreements, specifications, and services that enables the secure exchange of health information over the Internet. The NHIN is a key component of the nationwide HIT strategy which provides a common platform for Health Information Exchange.

The NHIN would also allow for connectivity to federal agencies that have responsibilities related to health care, such as the Centers for Disease Control and Prevention, or the Indian Health Service, or the Veterans Administration. Likewise, the National Cancer Institute or the Department of Defense also have the need to exchange information among patients that  they care for or among researchers that deal with patients that one of these agencies care for.

Now that we’ve got some of the basic concepts out of the way, let’s focus some more on the history of health information exchanges. Now a very excellent article by Vest and Gamm was recently published in the Journal of the American Medical Informatics Association that nicely covers a lot of the history that I’m going to talk about in the next few slides.

In 1990, the Hartford Foundation gave out several grants to a variety of different cities and locales for the purposes of building what they called, “community health management information systems.”  This was one of the first attempts at exchanging information about health electronically.

Seven locations received these grants.

The systems that were developed by the grantees were centralized data repositories that housed patient information including demographics, clinical data, and insurance eligibility information.  The main purpose of the system, according to Vest and Gamm, was for assessment purposes and to make it easier to bill for patient care by having eligibility information verified right away- preventing the need for the onerous paperwork that’s typically exchanged when denial of payments occurs when someone was treated for something they were not covered for.

None of these systems ultimately survived because in 1990, technology- including internet technology, were at a very prototypical stage.  This was the first time also that  someone attempted to  integrate otherwise disparate health information- which as we have learned over the years is much easier said than done. Other major problems with the system included  privacy concerns and  in fact in some areas where these community health management information systems were created there  was a real grass root effort  mostly led by  physicians that successfully lobbied the legislator in some states to do away with this health information exchange. And lastly because these were grant funded, there was a lot of problems for these systems to transition off grant funding support. In other words when the grant money ran out,  it became very difficult  to continue the operations of these systems because it was not clear exactly who should pay for it and what the return on investment is for someone who did.

By the mid-1990s another form of HIE came onto the scene; this time in what was referred to as Community Health Information Networks or CHINs. Whereas the efforts in 1990, which were mostly collaborative endeavors by community stakeholders, CHINs were mainly commercial endeavors.  These endeavors were primarily aimed at reducing costs by sharing data.

CHINs tried to overcome some of the pitfalls learned in 1990, for example, that may have led to the demise of the previous attempts at HIE.

For example, given the fierce political opposition by some groups at making community health level data available publically, CHINs employed decentralized data structures that made it less likely to violate some privacy concerns.

There was an estimated 75 to several hundred CHINs established or at least  planned, all of which, if not almost all of which, failed.  According to Vest and Gamm the main reason for this failure was several fold; first there was  a lack of a focus on community stakeholders which ultimately left at the table  just a bunch of  competitors which proved for a difficult model for success. In addition, vendors who were building these systems in the hopes of collecting fees seemed to pit the interests of hospitals against those of community physicians, which created some dismay.

And finally, there was also no clear return on investment established which was a similar problem as you may recall that the Community Health Management Information Systems had that preceded the CHINs.

In 1999, the Institute of Medicine releases the first of  a series of reports that ignites a national focus on patient safety and quality of care...

The first report, entitled To Err Is Human, suggested that as many as 98,000 people die each year as a result of preventable medical errors; a follow-up report in 2001 talked about how Health Information Technologies, including HIE, can be used to reduce errors, and improve efficiency and effectiveness of our health care system.

This new national focus on quality has implications for how the history of HIE continue to unfold...

In the early 2000’s a new type of HIE organization began to “pop up.”  These organizations, called Regional Health Information Organizations (or RHIOs) were local, neutral organizations bringing providers in a community together for the purposes of HIE.  RHIOs are usually made up of representatives from a variety of provider organizations in a given area. And consistent with the new national focus on quality (as ignited by the Institute of Medicine reports), RHIOs were mainly focused on improving quality in their communities.

In order to facilitate HIE, RHIOs needed to get everyone at the table to overcome challenges unique to their market.  Some of these challenges may have been political, or competitive, perhaps financial or regulatory; in some cases legal.  Note that by this point many of the technical barriers, including immature internet technologies and the like have been overcome and internet connectivity and computer hardware had gotten to the point where a lot of things that HIE requires technically was much easier to do.  So a RHIO must bring together stakeholders, manage agreements among them, manage security and access to information, and manage identities and log-in credentials, so everyone accessing the local system has access to the data when they’re needed with the correct authentication and security protocols, etc.

RHIOs continue to develop to this day, but it’s important to mention that in 2004 the Department of Health and Human Services begins developing the National Health Information Network which replaces an earlier concept known as the National Health Information Infrastructure, or NHII.  The current  phase of the new NHIN included developing prototype architectures, and the second phase involved developing specifications and services, and developing working constructs of the new National Health Information Network.

In 2006 the Agency for Healthcare Research and Quality releases a report that pretty much concludes that lots of variability exists from one state to the other with respect to RHIO activities, and state level HIE activities.  While some states had advanced HIE  systems in place, able to already exchange data across competing providers other states had little or no HIE activity.  Moreover state led activities varied from one state to the other as well.  Given the many benefits that HIE brings especially to health care payers, such as health insurance companies, state Medicaid programs began leading efforts in their states at developing HIE, under the assumption that benefits would accrue to their own budgets if they had a leadership role in convening stakeholders interested in health information exchange.

2006 was a milestone year for HIE because a prominent RHIO, who was a pioneer in many ways, failed and closed down 8 years after it began its operations.

Some of the reasons it failed was reminiscent of why previous HIE attempts failed.  According to Miller and Miller in a 2007 article in Health Affairs, the Santa Barbara RHIO was grant-funded and as a result had strong leadership by the grant foundation and grantees, but this strong leadership led to a lack of major involvement from community stakeholders who were anticipated to move into a more prominent leadership role  as the RHIO evolved, and ultimately didn’t.

There were also some vendor limitations associated with the fact  that a relatively small vendor was  trying to develop a product from scratch to handle the HIE needs that were planned for, and this particular vendor wasn’t able to keep up with the demands.

Lastly, there was also a lack of a compelling value proposition which as we heard previously was the case with earlier attempts at HIE.

One of the most valuable lessons of the failure of this RHIO is the importance of building trust and achieving and constantly updating business agreements among RHIO stakeholders.

The failure of this RHIO also spotlighted the need for health payers to be involved because the more clear return on investment to this group is able to be imagined and realized when HIE is achieved.

2009 was also a milestone year for HIE.  With the passage of the Health Information Technology for Economic and Clinical Health Act, or HITECH Act, this legislation marked the broadest federal support for HIE to date.  This broad support was on several fronts.

First, the HITECH act incentivizes the use of EHR adoption among providers who adopt certified systems and are able to demonstrate that they use these EHR systems in a “meaningful” way.  Importantly, one of the criteria for “meaningful” involves the ability to exchange health information with other providers.  In others words for doctors and hospitals and other providers who want to get bonuses from Medicare or Medicaid their EHRs need to be able to exchange information with other providers.

Second, the HITECH Act also addresses privacy issues by strengthening privacy, security, and confidentiality constructs that were previously outlined in other legislation such as the Health Insurance Portability and Accountability Act, or HIPAA. Thus you see more attempts at squelching some of the previous reasons that HIE failed in the past, including concerns about privacy.

In 2010 as HIE systems were moving forward this re-raised concerns related to privacy.  With the National Health Information Network being developed, there was concern that it will be easy for health information to be transmitted, or may even make it inevitable, that government agencies, certain government agencies, will have access to private information.  Some of these agencies, including the Department of Justice, or   Department of Homeland Security , or the CIA should probably not have access to private health information, and in an attempt to learn or apply lessons learned from previous HIE historical  events  the Office of the National Coordinator for Health IT, or ONC, assured stakeholders that the NHIN that it was developing will not exchange information with these government agencies.

HIE has been an elusive, but worthy goal in the US for several decades.  Each concerted attempt at achieving HIE reminds us of the challenges associated with such a lofty goal.  The good news is, that we have learned a lot about the pitfalls to avoid, and each attempt at achieving true HIE incorporates the lessons learned from the previous attempts.

Currently, political, financial, and social forces are aligning in favor of HIE.  Incentives provided in the HITECH act may be the impetus that brings true national HIE to fruition.

Return to Previous Page

Return to Index