Walking the tightrope: Balancing ICD-10 coding rules and data needs

December 4th, 2017 / By Rhonda Butler

Coding is fundamentally a balancing act, a delicate, high-wire dance intended to get useful data from the medical record that can be widely shared in health care. It happens at all levels of the ICD-10 coding and classification process, from coding an individual record, to creating the coding rules and conventions that coders use, to updating the ICD-10-CM/PCS systems themselves.

When coding an individual record, the coder is balancing a personal understanding of that specific record and how it should be coded, against official coding advice, guidelines and instructional notes. Upstream of the coding process, developing the coding conventions and guidelines that inform coding is every bit as much a balancing act. There, the Cooperating Parties, the Coding Clinic Editorial Advisory Board and others balance the perceived need to publishing more coding advice and guidelines against the possibility that the information will be misunderstood, used in inappropriate contexts, or have other unintended consequences. The balancing act continues: Updating the classification itself through the annual C&M process is an exercise in balancing—balancing the requests for more distinctions in the classification against the resulting tendency for the classification to become more complex, and therefore harder to use.

I have walked all of these tightropes at some point in my career, in roles that included coding inpatient records, developing official PCS coding advice and guidelines, and working on the annual update to ICD-10-PCS. I especially like the tightrope metaphor as it applies to Coding Clinic advice. Having served on the Editorial Advisory Board (EAB) in my 3M capacity as contractor to CMS since 2009, I am no stranger to that high wire act. What the Coding Clinic EAB most resembles is something less colorful: A bunch of lawyers and law clerks discussing whether a proposed ruling is consistent with previous legal precedent. So, it’s refreshing to picture all twenty of us instead lined up on the same tightrope, thirty feet above a flimsy net, in spangled skinsuits and grippy slippers, each adjusting our balancing pole as the tightrope sways. Fabulous!

It is my privilege and pleasure to share a tightrope with this group of people. Like any high-level set of principles, there is room in ICD-10 coding for interpretation and differing opinions, and sometimes we struggle to draft coding advice that feels like it will stand. But what makes this group so good to work with is that we share an understanding of what it is we are balancing: The need to make coding advice “as simple as possible, but no simpler” (to quote Mr. Einstein), against the need to make meaningful distinctions in the coded data.

Official coding advice needs to be as simple as possible, so coders across the country code according to the advice and we end up with more reliable data, but not so simple that it produces one-size-fits-all coding, eliminating useful distinctions that could have been captured in the coded data. Here’s an example. The EAB reviewed a question about female sterilization procedures involving the fallopian tubes, and the result was published in the 3rd quarter 2015 issue of Coding Clinic. The fallopian tube sterilization was done by ligating the fallopian tube and then excising a section of it. The coder asked, “What is/are the correct root operation(s) for this procedure?” Coding Clinic’s answer contained this sentence: “There are several distinct procedures performed on the fallopian tubes for sterilization, including ligation alone, fulguration, and ligation followed by excision. These are coded to the root operations ‘Occlusion, Destruction, and Excision’ respectively.” So, applying the advice to this specific case of ligation followed by excision of a section of tube, Coding Clinic answered that the root operation Excision would be coded in this case (3Q 2015 p.31).

As a group, we felt that the advice was pretty straightforward, and successfully balanced the need for clear instruction with the need for useful coded data. Here’s where it got interesting: Coders sent mail to the AHA challenging the advice, saying that since the overarching objective of the procedure in all cases of tubal sterilization was to occlude the fallopian tube, then all three techniques should be coded to the root operation Occlusion.

The EAB includes among its MD representatives an OB/GYN, and the Coding Clinic advice was drafted with the needs of that specialty in mind. Coding a single PCS root operation for all tubal sterilizations would be oversimplification—ignoring useful distinctions that could be captured in the data if the most specific root operation available were used. Being able to distinguish between the surgical techniques of ligation, fulguration and ligation followed by excision, three techniques neatly differentiated by the three PCS root operations Occlusion, Destruction and Excision, allow data users to compare efficacy and complication rates among the three techniques. If all tubal sterilization procedures were coded using the root operation Occlusion, this information would be lost.

This textbook example illustrates the challenge of the balancing act: Up on our own particular high wire, it is hard not to focus on our own feet and that single next step. What we would see if we looked up, and out, and around, is that there are high wire acts all the way to the horizon.

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

Are your CDI and coding teams speaking with one voice to physicians?