Interoperability vs Health Information Exchange: Setting the Record Straight

One thing that I’ve learned in the government is that words matter, and sometimes, particularly in complicated or technical areas, it can be hard to use words that are precise and accurate. Sometimes we use the same term to apply to a broad range of concepts or ideas because it can be complicated to describe all the nuances captured by that work. Sometimes the words we use start out as verbs and then turn into nouns.  Sometimes the names (and meanings) evolve over time. And sometimes the same words are used in different areas to mean entirely different things.

For example, the words “Interoperability” and “Health Information Exchange”  are two words that we often use interchangeably, but I think it’s important to realize that they are not the same thing.  Search results for the term “interoperability” will get you millions of hits, and generally include concepts of standardization, integration, cooperation, and technical specifications. (Perhaps in subsequent blogs we can discuss the different flavors of interoperability, but I digress).

Defining Interoperability

For definitions of technical interoperability, my favorite definition comes from the Institute for Electrical and Electronics Engineering.  IEEE (eye-triple E) is the largest professional organization of geeks out there. And they have been dealing with the issue of interoperability for a very long time.

The IEEE Standard Computer Dictionary defines interoperability as “the ability of two or more systems or components to exchange information and to use the information that has been exchanged.” See IEEE Standard Computer Dictionary: A Compilation of IEEE Standard Computer Glossaries (New York, NY: 1990).

That means that there are two parts to the definition of interoperability:

  1. The ability of two or more systems to exchange information
  2. The ability of those systems to use the information that has been exchanged

This means that health information exchange is different than health information interoperability. Exchange is a necessary for interoperability, but it is not sufficient by itself to achieve health information interoperability.

Defining the Difference between Health Information Exchange and Interoperability

Because exchange is a prerequisite for interoperability, here at ONC we’ve focused a lot of attention on it. Our early work with the NwHIN pilots emphasized secure, query-based exchange (using something called Web Services) in which one system asked (or queried) another system for information. The DIRECT project defined a secure, email-based exchange system in which one system pushed information to another system. And both web-services and email-based exchange are part of the 2014 edition of the certification rule that we issued last August.

But we must always remember that exchange is only part of the puzzle. If I send an email from one computer to another computer, I have exchanged information between those two systems. But if I write my message in French, (and you can only speak English), there is no way for you to automatically use the information that has been exchanged without risking losing something in translation. Similarly, if I use DIRECT to send a scanned office visit report from one EHR to another EHR, I may have exchanged information, but I won’t be able to seamlessly use the information in the new system to alert the provider automatically of a new drug allergy, for example. So to get to health information interoperability, we need more than just transport standards:  we must also use standards for vocabularies and terminologies (to help standardize the meaning of the words that we use), standards for structure (so computers know how to break a message into the appropriate information chunks), and potentially other kinds of standards (but more on that later).

Health information exchange is important – it is a vital part of modernizing our health care system. But health information exchange is not the same as health information interoperability. Transport standards are important to achieving health information exchange, but they will not get us to true interoperability unless we continue our work developing the full set of standards needed to support interoperability.

Stay tuned for more discussion and if you can’t wait, let us know your thoughts by commenting below.

Next week: The international scope of standards work.

Read other blogs by Dr. Fridsma on standards development and harmonization, coordination of federal and private efforts toward interoperability and health information exchange, and health IT innovations.


  1. Laura Ripp says:

    Thank you, Dr. Fridsma, for your article, “Interoperability vs Health Information Exchange: Setting the Record Straight.” Understanding and acting on the distinction and relationship between interoperability and exchange are key to advancing the meaningful use of technology in the health industry. Perhaps you can continue the theme with an exploration of “exchange”, the verb, as you describe and “exchange” the noun, as in a health information exchange utility or HIE–an organized system of technologies and policies that support a range of information management services such as secure storage, record location, transformation and delivery of health data for defined communities of users.

    • Jason Hughes says:

      Great informative article. Although I believe Interoperability still has a long way to go – but it is certainly the way to reduce healthcare costs and improve productivity, not to forget better outcomes.

  2. Dan Haley says:

    Thanks Doug – a useful and important distinction.
    Another related distinction that we at athenahealth have found ourselves using with increasing frequency as we talk to policymakers about the current state of interoperability and exchange is the subtle distinction between “interoperability” and “interoperation.” As we’ve said in many forums, including at ONC, we are concerned that the ratcheted-back standards of MU2 and now MU3 concentrate too much on the former – which describes a capability and could therefore be achieved passively – and too little on the latter, which describes an activity. Moving forward, EHR vendors ought to be required to achieve and practice “interoperation,” and not just with a government-operated dummy test recipient or within a contained, proprietary system. The unfortunate fact of the marketplace right now is that a lot of dominant actors have a very real financial incentive to contain actual interoperation within proprietary information silos, thereby undermining the baseline rationale of the MU program by locking in information (and by extension, providers). Thanks again for the post.

  3. Mighty Casey says:

    Wish you’d posted this last month, it would have prevented me from (unintentionally) kicking off a minor political storm in my part of the forest over a mistaken assumption that “interop” and “health information exchange” could be used synonymously.

    I do get the difference, particularly after reading your breakdown of it. However, outside of the committee rooms where NwHIN is working on creating HIE standards, many of us see interop as the purpose of the exercise. We get that the exchange has to happen before data can be used. But it’s frustrating – for patients and clinicians – to still have to wait for easy records exchange and data-based decision making that can happen in real time.

    Dan Haley points out something that has bugged me since the EHR push started with the passage of the ACA: vendors have had NO requirement to make their system play nice with any other system. Instead of data in a silo, we’ve taken that data and put it into 800+ new, smaller silos. But it’s still siloed, until we can get to interoperability.

    And as EHR vendors tend to view the data as their path to revenue, vendor lock-in and the lack of interop requirements from the outset mean we’ll be fighting from the rear for the foreseeable future.

    I am once again reminded why I have lingering antipathy for innovation tied to politics …

    • Dr. Linsay Way says:

      Lack of interoperability was always the problem with electronically ordering tests from outside labs, and now we’re experiencing the same thing with EHR exchanges. You would have thought they’d have seen that coming, but I digress…

      I disagree that we’ve gone from data in a silo to data in 800 smaller silos. A more accurate illustration would be going from 100,000 tiny silos to 800 small silos. True, right now many providers can only exchange records with a limited number of other providers using the same EHR system, but how soon we forget the days of deciphering paper record-keeping systems from other providers.

      When providers start demanding of their vendors the interoperability and easy record exchange that was the promise of EHR, then we’ll start seeing changes. It’s a matter of getting vendors on board, which is nothing compared to the challenge of getting hundreds of thousands of providers on board with meaningful use EHR.

  4. Kate says:

    Interesting string of comments. We do need to share (exchange the data) between entities before we can achieve interoperability.
    Funding Grants have been made to get this sharing going…I wonder how many states are using this funding to get the sharing to start happening. Or, are they using it to carve out a new way to raise money from the ultimate benefits the interoperability will offer?
    Sadly, I fear too many places are more concerned with the income stream to be acheived AFTER the state wide and regional exchanges are set up to offer benefits which lower the cost of health care delivery.

    We need to get back to basics…get the sharing of information and the building of interoperability going first, then later once the benefits are proven, establish the revenue goals.

  5. Adams says:

    Thanks for your information Dr. Doug Fridsma, 🙂

  6. Jamie Holland says:

    I too thought the terms were synonymous with one another. Thank you for clarifying and sharing your expertise.

  7. Omar Shaim says:

    Thank you Dr. Fridsma. One point I see worth noting into this mix is the health IT industry standard for information exchange protocols like HL7. HIE by itself is a layer that relies-on/assumes the use of HL7 to exchange information between two clinical systems for example. This means that HIE + HL7 do satisfy the IEEE definition since the information is both exchanged and used by both systems in the context of a patient related incident.

  8. Emeka Igwe says:

    Omar Shaim, in his comment above, raised an important point that all concerned need to consider as a way out of the whole back and forth issues around HIE and interoperability – the HIE/HL7 concept.

    The the HIE/HL7 approach will possibly provide some answers. HL7 will provide that standard structure or format for all participating information systems both local and outside facilities under which HIE will be functioning so that information exchanged from various will come with the same format of code sets, terminology, and semantics.

    Since HL7 is a standard Unit, not a software organization, information sharing and participation will flow freely, involving the IT, Users, vendors and administrators as members of the various levels of HL7 committees.

    If health information and data from different sources are exchanged using the same standard format, interoperability – exchange and usability – will be feasible if not fully achieved.

Leave a ReplyComment Policy