Dick Taylor, M.D.: ICD-10 will not do what you think it will

Dick Taylor, M.D., is a software engineer, pediatrician and chief medical officer for the Frisco, Texas, healthcare information technology consulting firm MedSys Group. He is also an outspoken critic of ICD-10. Dr. Taylor shared his perspective on why ICD-10 implementation will likely not improve healthcare data as hoped and why the one-year implementation delay is good news.

Health CXO: Do you see any value in adopting ICD-10?

Dr. Taylor: No, not that I can see. I see no value in ICD-10 over ICD-9. I don’t see a whole lot of value in ICD-9 and I think with ICD-10, while it adds a great deal of complexity, it is no better suited for the kinds of things people want to do with clinical data than ICD-9 was, and I think it will probably end up being worse.

Health CXO: If ICD-10 has a greater specificity of codes, how will that not help clinicians and researchers do more with clinical data?

Dr. Taylor: You have to look at how it is used. The question is not whether there are lots and lots of codes. ICD-10 is indeed much more granular than ICD-9. But I was having a conversation with an opthalmomogist friend of mine the other day and he pointed out that he bills between 50 and 59 ICD-9 codes because he deals with a specialized organ system. In ICD-10 he is also probably going to bill 50 to 60 codes. So what ICD-10 has done for him is given him rather than 1,000 codes he doesn’t bill, now he has 10,000 codes he doesn’t bill.

The researchers that want to use that data are going to be confounded by the fact that unless it makes a payment difference, then the data that is being produced is going to come from exactly the same process and exactly the same people that have been generating unsatisfactory data for claims forever. And the nature of claims data is such that it is aimed at payment. It is not aimed at informing a downstream researcher about what the patient might or might not have. And even more so, it is not aimed at informing them about the conditions that surround that diagnosis.

We have terminologies that have been very specific and very granular for a long time. The idea that we would need another terminology of this sort reflects a misunderstanding of the problem.

The problem that we have in collecting this information is that the process we have, particularly in fee-for-service medicine, are not designed to produce that type of information. And they are not going to produce it any better by making it more complicated.

Health CXO: If it is so poorly aligned with clinical and research medicine, why are so many organizations pushing for a swift adoption of ICD-10?

Dr. Taylor: You have to look at which organization and for what purpose. The Coalition for ICD-10 is probably the most outspoken group and it is comprised primarily of payors, a variety of healthcare IT organizations, and medical informatics organizations. Strikingly absent would be provider organizations, such as the AMA [American Medical Association] and the nursing organizations.

If you talk to practicing physicians and to organized medicine, you get the same answer. They don’t want it. If you look to at the letter they sent to Marilyn Tavenner, the fact is that they are saying two things. One is “we want ICD-10” and they are really weak on that in the letter. They are not arguing the benefits of ICD-10 beyond the general intent that it will make research better and it will make medicine better. But they don’t say how. What they are really saying is that people have spent a lot of money on ICD-10 and if they don’t know when they are going to implement, they may stop getting ready, which means that they will then have to spend a lot more money when it is time to get ready.

That uncertainty is a significant problem, and I actually agree with that. I think that if a gentleman came out and said we are never going to require ICD-10, that would also take care of all of the uncertainty and save all the money.

For the organizations that have spent money getting ready for ICD-10, we have to decide how much of that money was a spent cost, meaning how much of that money was no longer valuable to the organization. That is actually the point I think most organizations should take the longest look at anyway. To me, the primary value of ICD-10 implementation has nothing to do with the ICD-10 coding system itself.

I think the organizations that are advocating for ICD-10 are using that as a proxy for what they really want, which is better documentation, more modern systems, and more precise definitions and communications about what a patient really has. I think the mistake they are making is tying that to a billing and claiming coding system that is not designed for the purpose and cannot be made to serve it well.

Health CXO: Can the current ICD-10 preparation process be repurposed to better ends?

Dr. Taylor: Not only can it be, but it should be. There are three major things. The first thing is that organizations that have been really out front have been using this as the lever to convince their providers to document more effectively. With ICD-10, if your coders are going to be able to pre-ICD-10 code, they are going to need to have more documentation in the clinical record. The record has to reflect the specific thing that is going to make a difference for the ICD-10 code. But if you look at it, what you are documenting as a physician in the record is nothing more or less than what you should have been documenting already. If you look at the trauma codes, you’ve gone from leg fracture to mid-shaft leg fracture non-unlimited non-displaced. You know, if you as a provider cannot bother to say in the record where it is broken and how badly it is broken, you probably have a problem that needs to be worked on.

The organizations I’ve seen who are really using ICD-10 in a good way are primarily using it as a way to say, let’s improve our documentation in general. And there is nothing wrong with that. As a matter of fact, that is a terrific thing to do whether or not ICD-10 is out there.

Now you may say that when there is no longer an existential threat hanging over our heads forcing us to do this we will not do it. I would argue that from a professional responsibility standpoint, you shouldn’t need a threat to do what is right. Truly, physicians can change behavior. Long ago, it was considered appropriate for physicians to do things like throw objects at operating room nurses, and we’ve pretty much gotten away from those things. We need to do the same things with bad notes and sloppy documentation and we need to do so for the same reason, because it is the right thing to do.

The second thing organizations have been doing is upgrading obsolete systems and replacing systems that are no longer capable of carrying the code or are ICD-10 incompatible with systems that are. They also are upgrading to current versions of systems.

Again, this is really good behavior. There is nothing wrong with upgrading systems. If an organization needed the ICD-10 threat to do the upgrade, then the ICD-10 threat had a really good advantage.

Now the nice thing about that is that most organizations and most large systems were looking at lead times that were much longer than a year, so many of those ICD-10 preparations are done or nearly done. And organizations should complete those because it is a good thing to not be relying forever on badly obsolete software systems.

The third thing is something that got pushed to the back burner with ICD-10 but it is part and parcel of what I just talked about. Organizations need to do strategic planning around IT and up until March, the strategic plan was very simple. You had the existential threat of ICD-10 hanging over your head and you also had the incentives and soon-to-be penalties of Meaningful Use. So for most organizations, the strategic plan for last three years has been Meaningful Use Stage 1, Meaningful Use Stage 2, and ICD-10. Those blotted out the landscape as far as other strategic priorities and that is not actually strategic planning. That is a reactionary pose. But it is also the one a lot of organizations had to adopt.

This is the time to take a step back and say, “Wow, maybe we can’t rely on the Federal government to set our strategic priorities for us. Maybe we should be doing actual strategic planning and looking at IT as an investment and an asset instead of just an expense we have to incur or the Feds will put us out of business.”

Given the breathing space of at least a year and hopefully longer for ICD-10, organizations can now step back and ask themselves “What was automation supposed to do for us? It wasn’t supposed to cost us more money and make it harder to practice medicine. Automation was actually supposed to make us more efficient, reduce our costs and improve our customer service, just as it has in so many other industries.

It kind of got written sideways over the past few years into we just have to do this to survive. Maybe it is time to step back and say that all that energy we were putting into trying to manage that October 1^st transition date — because that was going to be a disaster — rather than putting the energy into trying to survive, how about if we try to thrive. How about we say, “We are smart people, we are doing a really good thing. Let’s try to figure out as an organization how to make this work out really well for us.”

That is how this hopefully very long delay in ICD-10 implementation can help organizations get out of this “a-gun-to-our-head” mode and into a mode where we say we practice medicine, we are proud of what we do and we want to do it as well as we can — and that includes using IT effectively because in 2014, organizations use IT and they use it well. That is what I’m hoping they are thinking and what I hope comes out of this.