The real reason doctors hate ICD-10

It has been freezing cold in much of the country for the last two months, but things have been heating up in the controversy over the implementation of ICD-10. First, a quick primer for those of you who have not been following this.

The “ICD” in ICD-10 stands for International Classification of Diseases. The “10” refers to the version of the taxonomy, which is maintained and revised periodically by the World Health Organization (WHO) and “is the standard diagnostic tool for epidemiology, health management and clinical purposes.”

Although conversion from the ICD-9 standard, which is still in use in the US, to ICD-10 is causing a major kerfuffle, it is important to note that ICD-10 has been around since 1990, and the WHO is poised to release ICD-11 in 2017. The 9th and 10th editions differ primarily in their specificity of coding, with the 10th differentiating between acute and chronic states of the same condition, left and right sided findings, initial and ongoing treatment, etc. The net result, of course, is that there are a lot more ICD-10 than ICD-9 codes to describe the full array of human disease and unfortunate mishaps, even though humans and the things that befall them have not gotten much more complicated since 1990.

The current controversy arises from the fact that the Centers for Medicare & Medicaid Services (CMS) has mandated that hospitals and physicians submit their bills using the new codes as of October 1, 2014, effectively creating a new national standard for reimbursement determinations. The timing of the changeover means that doctors and hospitals must implement this as they simultaneously struggle with new quality mandates and IT meaningful use requirements. No wonder, then that the AMA has renewed its call for a delay in implementation, citing, among other things, a study (that it funded) that estimates that it will be financially “disastrous” for physicians to implement ICD-10.

Although these are legitimate concerns, I think the objections that many physicians have to ICD-10 goes deeper than having to change some old habits of how we write our notes and drop our bills. I think it has to do with a fundamental disconnect about the role of documentation.

As students and trainees, we were taught that the medical record is a tool for patient care. That it is intended to share information with other providers; or create a narrative over time, so that a patient’s progress (or lack thereof) can be observed; or provide a repository of reference information that may serve a future, as yet unidentified, clinical need. Yes, including enough information in our records for others to summarize into ICD-10 codes based on hospital documentation, or selecting the codes ourselves for office-based encounters, serves those ends. But the problem is that most clinicians believe that they can achieve the fundamental goals of clinical documentation without the constraints and complexity of ICD-10 coding.

Here is the real problem. Just as I pointed out with EMRs, we have accepted a system that pays doctors and hospitals for “doing stuff.” Naturally, those paying the bills want to make sure that the stuff they are paying for is both appropriate and actually getting done, and have demanded that we document both. The language chosen for that exchange (we tell you what we did, and you pay for it) is an epidemiologic classification scheme that was not designed for that purpose.

Is it any wonder that doctors hate it?

Ira Nash is a cardiologist who blogs at Auscultation.

Comments are moderated before they are published. Please read the comment policy.

  • PamelaWibleMD

    Reductionist medicine’s obsession with documentation has really taken us off course from the task at hand: caring for the patient. We need time to look up from the computer and make contact with the patient – heart to heart, eye to eye, soul to soul. ICD-10 moves us closer to soulless and heartless medicine and the loss of the patient-physician relationship. Here’s a TED Talk describing how one town took back health care and salvaged the doctor-patient relationship:

  • doc99

    At this point, ICD 10 is coming whether we like it or not. The Political class has spoken, hellbent on placing yet another barrier between the doctor and his/her patient.

    • ninguem

      And other countries using ICD-10, use it in “back office” statistical work, research, etc.

      ONLY IN THE UNITED STATES, are front-line physicians ordered to use ICD-10 on a daily basis, and are punished for errors in coding.

      • Petter

        Not entirely correct. Sweden has been doing this for a while (not exactly ICD-10 but a similar code framework), the result is more administration (e.g. it pays to employ administrative ‘coders’ to extract as much reimbursement as possible), less patient focus, suboptimization with regards to patient safety, etc, etc.

        • ninguem

          First of all.
          Are the front-line practicing physicians required to use ICD-10 to document patient care?

          Not back office personnel in their health-care finance system, but the actual physician, where the physician is help responsible for knowing that ICD-10 code, and punished for good-faith errors.

          And it is not a trivial matter that it is “not exactly” the ICD-10, as many countries use a very abbreviated version of the ICD-10 that is nowhere at the level of detail anticipated in the USA rollout.

          And third
          What good has it done the Swedes?

          More administrative burden, and to what end?

  • Dr. Drake Ramoray

    More mandated work by third party payors without benefit to patients or physicians. The question should be why would I like the change to ICD-10?

  • johnfembup

    Two things that are not clear from the article – which I think otherwise provides helpful information.

    1. The ICD scheme is a diagnostic scheme. How is it a diagnostic scheme is the basis to pay for services rendered? Is CPT inadequate?

    2. If ICD-11 is due out in about 3 years, what is the point of adopting a massive diagnostic coding scheme that will be replaced so soon? Is WHO not really so “poised” to release it by then? Or is CMS really that bureaucratic? Oh, well, maybe this question answers itself.

  • ninguem

    Again I have to ask. Because I never get an answer from the “experts”.

    Is there any other country in the world that requires front-line physicians treating real patients, to know and use ICD codes to the degree required in the USA, and punish the doctors if they are used incorrectly but in good faith?

  • guest

    “Very little prevents us from choosing that approach.”
    It always perplexes me to see statements like this, which appear to imply that the financial resources devoted to healthcare are infinitely expandable. Of course they aren’t, as most people know. So what that means is that the more money we spend on paying for “back office administrators” to tend to an increasingly Byzantine system of coding, the less we will spend on the people who are actually providing the medical care, and the quality of care we provide will continue to decline.
    This is already happening in our healthcare system, as the ever-proliferating number of administrators siphons off time and money from those of us who do the work. Extra layers of bureaucracy such as ICD-10 and the increased demands it brings for more specificity in coding and documentation, just will continue to detract from the time and resources that are spent doing what it is that our patients want us to be doing, which is taking care of them.
    Oh, and the link above? I followed it and just spent about 20 minutes inputting various diagnoses that I commonly used into the search function. In all cases, the search retrieved a solid 2-3 pages of minutely different variations on the same diagnosis, many of them not valid under our current coding system. Sorting through these each time I admit or discharge a patient is going to improve my efficiency and specificity exactly how?
    Just a very wild guess here, but I’m thinking that Dr. Voran does not spend his days taking care of patients, like many administrators assuring us how simple this all really is if we would only take the “time” to focus on it. I have said this before, but it bears repeating: once you have a system in which a deep layer of middle managers dictate the workflow of the people actually doing the work, you will, inevitably, end up with an inefficient system that produces a low-quality end product.

    • David Voran, MD

      It’s not your efficiency it will improve. It’s going to improve the efficiency of administrators, researchers, policy makers and others. They’re having to do a lot of work now with the inadequate and practically irrelevant coding system we have today. Remember, we’re now in a world where it’s really not all about us physicians but the system as a whole.
      We can kick and scream as much as we want but since we’ve abrogated the fiscal relationship directly with patients we have to follow the rules imposed by those who pay us.
      There are a lot of ways to look at this but arguing about which terminology we are going to use as a whole is fruitless. The structural language is going to be evolving as we go forward in the same fashion medical knowledge keeps changing.
      It just may be if we spend a little more time selecting a specific code for a visit then those who depend on those codes (most never see the notes) then maybe the rules coming down the pipe might improve. They could get worse as well but the world moves and we either need to move along with it or step aside.
      In my own world I’d like us to spend a lot less time crunching out notes and a little bit more time reconciling and maintaining problem lists so they’d be more accurate. I also think we physicians should use SNOMED-CT and let the administrators do the cross matching.

      • guest

        At least your honesty in admitting that the system is now about “administrators, researchers and policymakers” is somewhat refreshing. But here is what is slightly dishonest: statements like “it’s not about us physicians anymore,” imply that physicians who object to these trends in healthcare are just malcontents, whining about how they aren’t being catered to any more. In fact, healthcare was never about the physicians, it was (and should have continued to be) about patients and taking good care of them.
        Those of us who are expressing concern are not whining about no longer being catered to. We are concerned about a healthcare system which is creating ever-growing barriers to providing adequate patient care. Because that “little more time selecting a specific code” comes out of the time that we have to think about things that actually matter to the patient, like what is their diagnosis and how best to treat it. So truly, medicine in many ways is really no longer about the patient.
        And if administrators were really being honest, they would be taking responsibility for directly explaining that reality to the patients, rather than pretending that the same levels of care are still going to be possible in our present system.

  • ninguem

    “…..The main difference is they don’t use it for billing so there are no repercussions for incorrect codes nor is there a person standing over their heads reconciling their documentation with their diagnostic codes…..”

    So……you’re telling me that the other countries could be extremely sloppy with the ICD-10 codes, no problem. Well, so much for uniform reporting across countries.

  • ninguem

    From the Wikipedia page on ICD-10

    The code set allows more than 14,400 different codes and permits the tracking of many new diagnoses. The codes can be expanded to over 16,000 codes by using optional sub-classifications. The detail reported by ICD can be further increased, with a simplified multi-axial approach, by using codes meant to be reported in a separate data field.

    The International version of ICD should not be confused with national Clinical Modifications of ICD that frequently include much more detail, and sometimes have separate sections for procedures. The US ICD-10 CM, for instance, has some 68,000 codes. The US also has ICD-10 PCS, a procedure code system not used by other countries that contains 76,000 codes.


    The other docs from various countries that you claim said “of course” when you asked them if they themselves apply the ICD-10 code, They are using an ICD system that has about a fifth of the codes the USA plans to use.

    From that same Wikipedia page, there is a breakdown of different countries using ICD-10. They all have their own modified version of the system. None of them are anywhere near as complicated as the system anticipated in the USA, and I mean they have maybe one-fifth of the codes we are about to impose on physicians.

    So once again, can you explain WHY the USA has chosen to impose on practicing physicians a coding system with a level of complexity no other country requires? NO OTHER COUNTRY is using an ICD-10 system anywhere near as complex as the one we are about to use. Not even close.

Most Popular