I was in a planning meeting this morning – a joyful, exuberant, bi-weekly event designed to maximize performance from developers and analysts alike – and someone raised this question:
“Is there a crosswalk we can use to price Medicaid claims?”
The inquirer was referring to an ICD-10-to-ICD-9 “back” crosswalk with the expectation that we might receive claims coded using ICD-10 while our existing software will price only ICD-9 claims.
I stated that there was no such crosswalk, and the 52 percent of people in the room who know of General Equivalency Mappings (GEMs) all inhaled simultaneously, sucking half the oxygen out of the room, fully confident in their knowledge of GEMs as a viable crosswalking system. (Incidentally, is “crosswalking” really a word?)
As you would expect, I sensed the instantly hostile tone and was compelled to backpedal a bit by addressing the obvious concern my comment had raised among team members. So I pointed out three facts to the individual who originally asked the question while the remainder of the team listened in, hanging on every word in the sincere hope that I’d make yet another faux pas as good as the last!
This is what I said:
- There are crosswalks, including GEMs, but they are “one-to-many” comparisons – there are ICD-10 codes that have no ICD-9 equivalent. A developer to my right chimed in, saying “Well, yeah, but there is a ‘priority’ indicator to tell which is most likely.” I conceded that he was correct but stood my ground that it really didn’t help, since we have no way of knowing which ICD-10 codes the various government payers actually will include in their new regulations.
- Because my team’s mission is to price certain government claims as precisely as possible, we couldn’t presume to second-guess what a Medicaid program…or Medicare…or Tricare…will do in the future. For example, if we referenced an ICD-9 code in our software and it had 16 possible ICD-10 codes to which it could map, we had no basis on which to decide how many of those ICD-10 codes actually would wind up in the payer’s payment protocol (even though we could map a billed ICD-10 code back to an ICD-9 code, in most cases). In other words, we can crosswalk back to ICD-9 fairly easily, but that is no guarantee that the ICD-10 codes billed will be correct when the payers finally publish their rules.
All of this really points to my No. 1 problem with GEMs and all other ICD crosswalks:
I don’t care to make business decisions based on best guesses and probabilities. That’s not so much business as it is structured gambling, and let’s face it: in gambling games, the odds always are in favor of the house. That is, you will lose eventually; it’s only a matter of time.
Those who have decided to count on crosswalks to do their ICD-10 grouping rather than buying an ICD-10 grouper or paying someone to do it for them have my unmitigated respect, because they are truly dedicated gamblers. They have decided to risk their future livelihood and, ultimately, their success on the probability that a particular ICD-10 code is the right choice. Relying on probabilities only makes sense in two scenarios:
- When the odds are statistically significantly in your favor, or
- When you can make conscious decisions and take actions that move the odds to your favor.
Any other time, you may as well pin all your hopes and dreams on a coin toss, because using GEMs in this manner – which CMS (to its credit) has unequivocally denounced – is equally irrational.
There is an old saying from Mother England:
“Penny wise and pound foolish.”
You may save a penny thinking it’s smart, but it might cost you a pound to fix what your cheap side embraced.
Or something to that effect…
Using a crosswalk to code claims is foolish: you simply can’t win. If you are a healthcare professional, think with your brain rather than your heart (or worse, your checkbook) and hire experts to code for you – or train the experts you have.
The money you spend doing your coding the right way will be a pittance compared to the benefit you’ll realize over time. Remember, ICD-10 is going to be with us for 10 years or so, and it will be followed by ICD-11: and then you’ll know what pain really is, especially if you don’t act now to do it right the first time.
About the Author
Billy K. Richburg, M.S., FHFMA is HFMA-Certified in Accounting and Finance, Patient Accounting and Managed Care. Bill has held positions including CEO, COO, CFO, and CIO in hospitals ranging from 75 beds to over 300 beds. Bill is a Board Member of the Lone Star Chapter, HFMA, and is Director of Government Programs for the Revenue Cycle Technologies business segment of MedAssets, Inc. His office is in Plano, Texas.
Contact the author
To comment on this article please go to firstname.lastname@example.org