Set National Standards for Health Information Systems

A recent report issued by the Institute of Medicine – titled “Best Care at Lower Cost” – calls for a dramatic transformation in health care delivery, saying “America’s health care system has become far too complex and costly to continue business as usual.” Its first recommendation (“The Digital Infrastructure”) focuses on the importance of health information systems and highlights a crucial aspect of their development that is too often overlooked – the issue of interoperability. Will the individual systems that are created be able to work together efficiently?

It’s an enormously important issue for health care broadly, and it will determine how effective those systems can be on a national level. At present, health care providers across the country are creating or enhancing their health information systems. In some cases, like ours at Intermountain Healthcare, those systems have a long history; we began instituting electronic medical records 40 years ago. Others are early in the journey. But all are being developed essentially for their own internal needs. Interoperability is low on the priority list.

Five health care providers who have been in the forefront of using electronic medical records have been collaborating on the creation of a Care Connectivity Consortium to pioneer the effective connectivity of electronic patient information across their systems. Those five are Intermountain Healthcare (based in Utah), Geisinger Health System (Pennsylvania), Group Health Cooperative (Washington), Kaiser Permanente (California), and Mayo Clinic (Minnesota). But even that ground-breaking effort, in which I’m heavily involved, will result in a multi-provider network, not a national one.

While we are already learning a great deal from the collaboration, and that learning can be broadly shared, it’s a national network that we ultimately need. Only a truly national network will allow the efficient transmission of secure patient information to best serve patients in multiple ways. It will serve them when they move (changing doctors or providers, traveling temporarily or relocating permanently); it will enable best practices to be shared across the country, and it will allow the broadest research and learning to advance health care delivery.

The IOM report recommends, in part, the following: “The National Coordinator for Health Information Technology, digital technology developers, and standards organizations should ensure that the digital infrastructure captures and delivers the core data elements and interoperability needed to support better care, system improvement, and the generation of new knowledge.” Here standard-setting is the key, and a good analogy for the problem can be seen in the evolution of the railroad in Australia.

There railroads developed one by one: some for moving natural resources like coal; others for carrying more freight, and still others for transporting people. While trains and tracks did get built, the railroad system was not constructed with adequate standards. Many different scales of railroad evolved, preventing railroad cars on one track from running on others.

To overcome this obvious challenge, the railroads built new stations and invented new contraptions to move cargo from one set of train cars to another. They were clever indeed; excellent engineering, for sure. But each contraption and transfer station slowed the transportation system down, added risk of product loss and increased the cost of shipping by rail.  After many years of subpar train service and increasing costs, Australia defined a standard gauge for its train system. It was likely a huge expense to make this change, but the efficiencies gained continue to be realized today.

The parallel is obvious for America. We need national standards to ensure, as the IOM recommends, “that the digital infrastructure captures and delivers the core data elements and interoperability needed”. The federal government has rightly made a major investment in electronic medical records, having committed $35 billion from the stimulus bill to it. We must now ensure that, as the capacities of many individual providers grow, they evolve into an efficient and effective national network.

Information and information systems in healthcare have tremendous capabilities. Moving from paper-based to digital systems is a crucial step in facilitating the sharing of knowledge, but adequate standards are needed to allow for the ubiquitous sharing of data and, ultimately, enhanced knowledge. The potential is enormous, if we set the standards that will provide common tracks on which this railroad of information will run.

Marc Probst is Chief Information Officer for Intermountain Healthcare, headquartered in Salt Lake City.

11 replies »

  1. If you’re one of the people that simply wants more fans to
    increase you rank or your star status, or you simply are trying to promote you
    business, you have to stop and think: If the accounts provided to me from
    the “get more followers” website(s) are inactive, how
    will that increase my rank or help my business flourish if there’s really no one
    behind these accounts that can help spread the word on what I do, or
    give me more business in the first place.
    Get free social exchange facebook likes youtube views and twitter followers.
    tv to expand its website and include more gamer-targeted advertisements.

  2. Curly,

    Not sure if your comment was directed at me or not.

    I am no fan of current EMRs. By most accounts they are; cumbersome, slow docs and nurses down, haven’t been shown to reduce costs or improve outcomes. (However other than that and their cost they are great – sarcasm intended).

    Be very careful about taking anything Elizabeth Warren says at face value. I carefully read her paper on the relationship between health care costs and bankruptcy. In my opinion that data in NO WAY supports her conclusions. To put it more succintly, she is total a b*lls**tter.

    See: http://www.theatlantic.com/business/archive/2009/06/elizabeth-warren-and-the-terrible-horrible-no-good-very-bad-utterly-misleading-bankruptcy-study/18826/

    Now I don’t know whether Intermountain does what she says or not. I just know that she is not a reliable source.

  3. What are you guys smokin? The HIT projects in the US are failing like they did in the UK. Outcomes are no better.

    Costs are no better from HIT and may be much higher.

    Vendors are making money. Consultants are making money.

    Doctors are struggling to keep their patients safe from the adversity of poorly designed EMR and CPOE gear. Unexpected deaths are occurring.

    Last but not least, Senator Elizabeth Warren, a bankruptcy specialist, reported in a lecture that Intermountain bankrupts more patients with aggressive collection tactics than any other medical center in the country. Is that interoperable?


  4. Not sure I know ins and outs any more… particularly the “ins” 🙂

    But DICOM is a good example for how standards eventually evolve from business needs and from use and practice. It was ACR that decided to work with the manufacturers (all collaborating rather nicely) and they came up with a workable solution together, and refined it through trial and error over time. Maybe that’s why there are so many free or cheap DICOM viewers out there. No “feds” anywhere in sight….

  5. Thanks for explaining that. Obviously, I do not know the “ins and outs” of this like you do.

    I am a Radiologist and my perspective in influenced by “DICOM” which was a set of standards forced on equipment manufacturers that allowed data to come from different machines and still be usable .

    I agree that the standards should be written in terms of what we wish to accomplish rather than the specific ways in which the problem needs to be approached.

  6. The problem here is that the “feds” requirements are too onerous to start with. They are asking that initial records be captured in a certain way (specific structured elements), bundled in a certain way (CCD/CDA) and sent/received in a certain way (NwHIN/Direct).
    Instead, if the “feds” just asked that records be sent, or available elsewhere, in whatever format or shape, or just allowed the new business models to dictate that records should be transportable, the IT geniuses would have come up with thousands of ways to do this, and eventually would have standardized on a handful that worked best. Just like the railroads.

    The railroads developed because there was a business need that merchants were willing to pay for, not because someone mandated that there should be railroads and commerce, at the behest of rails makers. It’s better to have non-standard railroads than to have no railroads at all, particularly when proposed standards are theoretical in nature and never been tested in the field.

  7. Margalit,

    Not sure what you mean.

    What I mean is that any kind of reasonable EMR should allow a doc at a hospital that a patient is admitted to (or even in the ER of) to pull up the patients records from another hospital, their doctors office (if it has an EMR) or medical clinic the patient is treated at.

    Similarly, if a patient moves his/her “medical home” their records should be able to come with them.

    If these requirements are not part of the Feds requirements for EMRs then someone was either asleep or has been paid off – in my opinion. Why would anyone write specs for systems that can’t talk to each other when patients move, get new docs, see new specialists, etc. etc.

  8. It depends on what you mean by “moved” and what you mean by “medical records”. If you just mean moved and medical records in the most obvious sense, this is not what the “feds” are asking for, and that’s why “easily” is not an option.

    I wonder what would have happened to the railroad industry if instead of concentrating on moving goods from here to there by any means necessary, it would have waited for the “feds” to define the allowed gauge of the bolts to connect tracks and the exact shape, size and consistency of goods to be transported, not to mention how to build everything from locomotives to train stations and how to fuel and manipulate every single color coded lever and button.
    We’d probably still be driving oxen….

  9. It is ridiculous that anything less than full exchange of data between EHRs would even be considered or paid for by the government.

    Suppose a patient is seen/admitted to one hospital in town and later on ends up at another on the other side of town. Medical records need to be able to come across seamlessly.

    Or a patient lives in CA but moves to NY. Again medical records should be able to be easily moved from one place to another.

    That the Feds would accept anything less that complete interoperability suggests to me that campaign contributions/lobbying are at work.

  10. Part of the problem is the absence of national providers of healthcare in the United States, with the possible exception of the VA..

    The long prevailing belief among U.S. healthcare providers is that “healthcare is a local business and will always remain a local business.”

    Since reform in any industry is rarely if ever led by established interests, a national system will probably have to come from “outsiders.”

    Ron Hammerle
    Health Resources, Ltd.
    Tampa, Florida

  11. I would say do more than set standards. Write the code. The NLM revolutionized and continues to revolutionize information technology in health care. They should write a standard EHR code for both individual practitioners and enterprise wide that works as a free downloadable product from their web site. They have the basic VA code to work with and it is difficult for me to fathom why that code should be modified and sold by independent companies. The way the EHR has been handled politically is a great example about why health care costs are so high in this country. Congress is too busy inventing high margin products for the IT biz.