• In health IT, standards matter

    Sarah Kliff has a nice piece up on her own interaction with the electronic medical record at her doctor’s office (emphasis mine):

    From the moment I walked in the door, the clinic felt like a model of what these laws are meant to encourage. In reception, I entered my information into a laptop. In the exam room, my medical history, blood pressure, temperature, and details of my earache went straight into a computer, too. I left with a prescription for ear pain medication, as well as a user name and password, so I could access my record at home and, if need be, make additional appointments.

    But the clinic also seemed stuck in a lot of what’s frustrating about health care. With the medical assistant, I went through the listing of my surgeries, allergic reactions, family history. The process of rattling off my medical history – which I’ve done dozens of times before, for dozens of other doctors – took longer than the actual appointment. I didn’t even do very well at it; I couldn’t remember the whooping cough medication I was allergic too, so I guessed. But since the computers here aren’t connected to the computers for other doctors I’ve seen, there was no better way about it.

    And there’s the rub.

    Start with the fact that something like 4 out of 5 physicians in the US are practicing without an EMR right now. Add in the fact that doctors don’t like to change their habits much. Throw in the truth that even though financial incentives are in place, they are totally inadequate to cover the total cost of getting up to speed on meaningful use. It’s difficult to get optimistic about us truly getting going in five years.

    But none of that compares to my frustration about the lack of mandated standards in the United States. The problem isn’t just that practices dont’ have electronic records. It’s that even when they do, the records can’t talk to each other.

    Let’s say Sarah’s clinic gets totally going with full bore EMR. But when she goes to the ED, they have an EMR from a different company, and no way for the two systems to communicate easily. They discharge her home, but no data is transferred to her primary care doc. Later, when she follows up, her doctor sends her to a laboratory for tests. They, too, have a different system, and it stores data in different fields than her doctors’. Turns out she needs a simple procedure at an ambulatory surgery center, with a fourth way of doing things.

    Then her primary care clinic’s EMR vendor goes out of business. They need to buy a new one. But there’s no good way to export all of the data and import it into the new system, so they have to pay someone to manually enter all the data from one to the other.

    This isn’t all by chance. It’s actually in companies’ best interests that their systems don’t talk to others easily. That way, when you need to expand, you need to keep adding on pieces from them and not others. But if that’s the case, where are the awesome cost-savings? Where are the efficiencies? Where are the reductions in medical errors?

    People can pay lip service to HL7, and Snomed, and LOINC, but those still allow far more ambiguity than you might think. Until we can agree on a standard data format, and require by law that every system can export data and import them at the touch of a button, the reality of Health IT will be far short of its promise. It’s technical, and it’s not sexy, but this is the whole enchilada, in my opinion. I wish the government would get more interested and vocal about it.

    Share
    Comments closed
     
    • Very good points about the lack of connectivity and “interoperability” between the different EMR systems.

      Another problem is that it is not in the provider’s (especially hospital/health system) interest in making data easily available. As providers come together in larger entities they want to keep their patients within the “walls” of their system and lock in referrals. So, in addition to a standard we need incentives for organizations to play.

      • That’s an excellent point, and also critical in understanding why it’s so difficult getting regional health information exchanges going.

        • There is some progress on this front, with several very large regional players — from Kaiser to Mayo — working on interaction between their many and various EMR systems.

          However, the fix is staring us in the face and is easy-peasy: require that all billing to Medicare and Medicaid — and encourage private insurers to join in — make automatic electronic access to all the patient records possible for review, if desired. Once that unified system exists, the job will be done, since providers can just hitchhike along.

          If this were implemented, it would also greatly simplify and unify billing, driving down billing costs for providers by enough to pay for the systems fairly quickly. Another free bonus is potential improvement in detection and prevention of fraud. Plus this does not require any change in patient consent, since patients already sign away the right to allow payers to examine their records right now.

    • If a patient has a stress test at a competing hospital six miles away, I cannot obtain the results in a timely fashion. Many of my patients cannot remember important details in their history, things like having had a valve transplant (less than half know which valve, I have been keeping track). We need EMRs that talk with each other and let us talk to them. We need to find some way to past the economic competition between hospitals to prioritize patient care over hospital profits.

      We also need to have apps that we can put on our phones so we can access the EMR data. If I can get on the internet via my phone, I should be able to look up their lab results.

      Steve

    • I loved everything here but the conclusion (!) and I would encourage you to think of standard as emerging not from policy but from dynamic, evolutionary processes. Take web standards, for example: http://www.w3.org/standards/. The web is built on literally hundreds of standards, constantly in revision by many different standards bodies. The web could not exist without this shared understanding. However, the web depends both on standards AND the continual breaking, building and branching of standards. Innovation online is a process of replacing old standards with new.
      If physicians had a strong and direct incentive to coordinate care, standard bodies would almost surely emerge to facilitate. But the incentives for care coordination, especially across ‘systems,’ are blunt/nonexistent and tech companies can therefore get away with propriety standards. I understand the impulse to want ONE standard (http://xkcd.com/927/), but I don’t think a master government-mandated standard is the right approach.

    • Perhaps publicly available quality of care measures could be modified to include–if they don’t already—-measures that reflect the provider’s ability to send and receive timely, complete information to and from other providers.

      An additional incentive would be to allow failure to take reasonable steps to provide or obtain information to be included as evidence in malpractice suits. To be fair, the bar for meeting the standard for reasonable efforts would need to be set very low in the beginning and raised gradually.

    • Anyone who has put time in as a software developer knows, as Jacob pointed out with XKCD link, that computer standards are a weird beast. Having that One Standard sounds really awesome and inspiring but it is so excruciatingly rare. HTML is a standard but has such varied interpretations, which is why IE != Firefox != Chrome != Opera != Safari != …..

      It’s best to accept multiple standards and move forward than to postpone while waiting for the One True Standard. Instead of relying on law to mandate a standard, it would be better that the law mandate all export formats be open and free of all costs, royalties, etc. and let the market use & determine the winning format(s).

    • Lets nationalize (eminent domain) CP codes. Compensate the AMA fairly, but free the code!