“Correct coding” means agreeing to agree

Aug. 22, 2018 / By Rhonda Butler

I have said it before—I am not a fan of the term “correct coding” because it can so easily be misconstrued as being about individual interpretations of correct and incorrect, right and wrong. That opens the door for discussions that too often devolve into individual interpretations of what words mean—typically a roundabout way of saying “I’m right and you’re wrong.”

We all enjoy being right, right? Of course we do. But sometimes being right misses the point, particularly in medical records coding. “Correct” means all kinds of stuff in different contexts—consider the different universe of meaning between “correct answer” on a spelling test and “correct thought” in Maoist China. In the coding arena, “correct” means something else altogether.

To understand “correct” in “correct coding” it’s useful to remind ourselves what coding is all about: It is the collective effort of producing consistent data. Consistent data is the ultimate goal of coding and it is only achieved if everyone follows the official guidelines and coding conventions.

Coding a medical record creates a summary version of a patient encounter that captures and prioritizes the essential diagnostic conditions and treatments. Coded medical records can then be compared to similarly coded records, with some confidence that the patients those records represent are also similar. ICD-10 is primarily about organizing like conditions and treatment together so that you can query records coded in ICD-10 and get useful information—for example, with a single query you can answer the question, “In our facility in the last 90 days, how many patients with a diagnosis of respiratory failure were treated with mechanical ventilation for more than 96 hours?”

Producing consistent data requires constant, collective effort, to code according to the latest official instruction. We must all agree to agree as much as possible so that similarly coded patient records represent similar patients, and differently coded patient records do not mask important patient similarities.

The ethics of the profession require that coders follow the official sources of coding instruction. There are three official sources of coding instruction. These official sources perform different but complementary roles in creating consistent coded data. They are:

1. Definitions and coding instruction—Inherent to each of the ICD-10 classifications, ICD-10-CM and ICD-10-PCS are instructions pertaining to the codes themselves, and definitions as to what words mean inside the self-contained world of the classification.
2. Official guidelines—Two separate documents, one for ICD-10-CM and one for ICD-10-PCS, that lay out the high-level ground rules for using the classification to assign codes to medical records (which codes take priority, relationships between related conditions, which codes are not reported in the presence of other codes, and so on).
3. Coding Clinic—Offers advice for specific coding scenarios that coders have asked about, and for which the coder must provide the documentation from the patient record (no hypotheticals allowed). Coding Clinic establishes individual coding conventions for real-world, full-context clinical scenarios, where both diagnosis and procedure codes as well as data needs are considered.

To produce consistent data, all coding professionals need to follow all three sources of official coding instruction to the best of their ability. This behavior is what the “correct” in “correct coding” means. If coding instruction from an official source is clear, correct coding means we follow it. Whether we agree that the information satisfies our understanding of what is philosophically “the right answer” is beside the point.

Similar to federal regulation, official coding instruction exists to be followed, until the next set of instructions is issued. If there are changes in an updated set of instructions, then the new instructions exist to be followed. A source designated as official sets the standard, for the time that that set of instructions is in force. Questions of right and wrong, better and worse, are not relevant in this context. If the coding instruction is clear, we follow it. If the coding instruction is unclear, or if one source of instruction appears to conflict with another one, we ask for clarification. In cases where coding is unclear or appears to conflict, it is absolutely appropriate to discuss the situation among peers and come up with an interim consensus on how a scenario should be coded, so that the data will be consistent until clarification from an official source is made public.

This whole topic can get tricky because coders are by temperament very detail-oriented. And the more experience coders have, the stronger their opinions about coding are likely to be. But personal opinion does not supersede clear official instruction. Official sources of coding instruction are correct, by definition.

Being human, it is hard not to get caught up in wanting to be right. We especially enjoy that little neurological effervescence—like champagne in the brain—we get from being right in front of our peers. For a coding professional, being clever enough to look for things you don’t agree with in the three official sources of coding instruction is a mixed blessing. Bound as you are by the profession’s code of ethics, you are obligated to continue to follow the official sources despite your personal disagreement. As a member of the profession, you should also ask for clarification from whichever organization has lead responsibility for the instruction—NCHS/CDC, CMS or AHA. If instead you publicly announce that you are right and the official instruction is wrong, and worse yet, try to get others to go along with you and ignore the official instruction, it just makes a difficult situation worse by introducing more inconsistency into the coded data.

Correct coding does not refer to proprietary knowledge possessed by coding superheroes in brightly colored skinsuits. Correct coding belongs to all of us. It is the dedicated daily practice of trained professionals agreeing to follow official sources of coding instruction to the best of their ability. Correct coding is not about plumbing the depths of this or that definition or guideline to find deep, underlying nuggets of meaning buried in the language. Correct coding is not philosophy and it is not comparative literature.

Coding is reading comprehension, attention to detail, and discipline. Correct coding is reading comprehension, attention to detail and discipline, without letting the ego get in the way. It takes a secure sense of self-respect and a calm understanding of one’s place in the world to be a first-rate coder. The best coders are not coding personalities—they are superlative scribes, taking notes in a specialized note-taking system called ICD-10, or CPT, or what have you.

Correct coding and consistent data are inseparable. They are the result of collective consensus—setting individual egos and opinions aside for the common goal, so that similar records are similarly coded.

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