ICD-10: The end, the beginning

Out with the old, etc. I hope 2015 saw the end of some ICD-10 habits we need to break—and 2016 is the beginning of some things we need to get on with.

First, we need to break the habit of mapping from ICD-9 to ICD-10. Seriously, we need to chuck the 9 to 10 mapping. If I could get the 9 to 10 GEMs file taken off the CMS website, I would. Mapping 10 to 9 is necessary until we have enough ICD-10 claims data for analysis. But no one should be looking to ICD-9 codes for wisdom. The number of inquiries I have seen that use the 9 to 10 GEMs as the final authority on how the world should work is concerning. To those of you who continue to live in the world of ICD-9 and map forward, you need to stop. Learn ICD-10, and use it. Please. We need to give up the “let no new thing arise” mindset. Yes, this first year after implementation is a break-in period and we don’t want to change too many other things at the same time. But some want to severely limit the first code update after the thaw, and this is a bad idea. The ICD-10 code update freeze was supposed to be a two-year thing and it turned into a four-year thing. We need to avoid letting this become an every-year thing.

So, we need to trade in our pre-implementation habits for post-implementation habits. Now that we are done with all the stalling and waiting, and also now that we benefited from all the dedication and preparation for October 1st, we need to just get on with it. No more cheerleading and sound bite-friendly pronouncements about the things we can do with ICD-10. Now that we have ICD-10, we need to do them.

Many methodologies used for payment (MS-DRGs and other groupers) and many quality measures were of necessity designed around the idiosyncrasies of ICD-9. Shoving the square peg of ICD-10 into the round hole of ICD-9 is what we call “replication.” It is a necessity for the short term only. As soon as possible we need to take the next step and get on with designing methodologies that make the most of ICD-10. We need to do this well or ICD-10 risks being left behind as the industry experiments with other payment models and other ways to measure quality.

At the same time that we are taking next steps and making changes, we need to look at what we do, and why we do it. Why do we code medical records? It is a pretty basic question, with a pretty simple answer. Coding is a means to an end. What is the end? Condensing the wealth of detail in the medical record to a form that is manageable, actionable, and transferable. But while we spent all this time fighting through three ICD-10 delays, the new world has emerged—of EMRs, EHRs, HIEs, and a return to capitation (repackaged as value-based payment). Where does coding fit into this new world? I hope to write about this in future blogs.

And finally, we need to be ready to adapt to what may be an Uber-scale disruption of our industry. If Big Data can deliver for health care what it has done for so many other areas of our lives, the potential for creative disruption is huge, and sooner than we imagine we may find ourselves nostalgic for a “non-event” like ICD-10. This last thought scares me, but in a good way. I am all about doing what works, and health care needs to live fully and effectively in the Internet age, whatever form that takes.

Cheers to you all—and bring it on!

Rhonda Butler is a senior clinical research analyst with 3M Health Information Systems.