Yes, Doctors are Required to Use Words

EDITOR’S NOTE: This article is in rebuttal to an article written for RACmonitor by David Glaser, Esq. on June 23, 2021.

A recent article in RACmonitor, Are Doctors Required to Use Words Rather than Codes?, addresses a question published in Coding Clinic during the fourth quarter of 2015 on whether there is “an official policy or guideline requiring providers to record a written diagnosis in lieu of an ICD-10-CM code number.”

The article’s author dismisses the Coding Clinic response and concludes that, since he is unaware of any rule to the contrary, “the code is enough; no words are necessary.” At the risk of taking what might have been meant as a lighthearted post too seriously, which is reflective of the sort of coder I am, I would like to encourage a deeper look and a more serious consideration of what would happen if we took that stance to its logical conclusion.

There is No Coding without Documentation

Coding Clinic says yes, doctors are required to write words, and they go on to remind coders that ICD-10-CM is a statistical classification. The role of physician documentation is inherent in the use of the code set because the process of coding is the translation of physician documentation into codes. Before we even get to the first convention in Section I, the preamble to the ICD-10-CM Official Guidelines for Coding and Reporting reads:

“Adherence to these guidelines when assigning ICD-10-CM diagnosis codes is required under the Health Insurance Portability and Accountability Act (HIPAA). The diagnosis codes (Tabular List and Alphabetic Index) have been adopted under HIPAA for all healthcare settings. A joint effort between the healthcare provider and the coder is essential to achieve complete and accurate documentation, code assignment, and reporting of diagnoses and procedures. These guidelines have been developed to assist both the healthcare provider and the coder in identifying those diagnoses that are to be reported. The importance of consistent, complete documentation in the medical record cannot be overemphasized. Without such documentation accurate coding cannot be achieved (my emphasis italicized).”

This part of the preamble tells us three important things. First, it cites the law that makes ICD-10-CM our official code set for diagnoses, which makes adherence to the Official Guidelines mandatory (see 45 CFR part 162, subpart J, §162.1002(c)(2)).Second, it explicitly tells us that not only is physician documentation required, but it must be consistent and complete documentation. Third, there are two parties required for using the code set: providers and coders. Providers are responsible for the documentation, and coders are responsible for translating those words into codes.

Several truths need to be spoken, and they all hark back to the nature of the classification. A code is not the same as a diagnosis; it is the translation of a physician’s diagnostic statement. The title or descriptor for an ICD code is not the same as a diagnostic statement. Codes are not “shorthand” for their descriptors. Instead, a code title is a kind of shorthand for the fullness of what each code can represent.

Coding is a Language…with Rules

The code book does have words, in alphabetical order, but it is not a simple list, with a one-to-one correspondence between entry and code. It is an index, with main terms and sub-terms. Once the coder locates a number, the next required step (see I.B.1 in the ICD-10-CM Guidelines) is to take that number over to the Tabular Index and read and apply the instructional notes, inclusion terms, and excludes notes to verify that the code is correct (and if not, start over again). The classification system is a constantly evolving language with rules, conventions, and definitions. In a physician’s diagnostic statement, what does “with” or “and” mean? What is a principal diagnosis? What qualifies as an additional diagnosis? Which codes can never be reported together? Which codes must be reported together, and in what order? If doctors have the time and interest to become fluent in this language, that would be remarkable, but it would not in any way relieve them of the responsibility to supply the documentation.

We Need Coders

Even if doctors were aware of every coding nuance and which rules apply on which dates of service, what happens if they misinterpret or forget? How would we know? Without documentation, there is no way of verifying or even questioning the accuracy of the code selection, since “code accuracy” only has meaning in relation to physician documentation. Imagine one physician basing a life-or-death medical decision on another doctor’s codes, rather than his or her diagnostic statement. Imagine that decision leading to an adverse event that ends up in a courtroom, and instead of documentation, all we have is the physician’s list of codes. Or imagine a payor wanting to know if the payment they made for a case was fair and accurate, and instead of documentation, all that’s available is a list of codes that are identical to what was submitted on the claim. No one would be okay with that. If we see where the “code is enough” path takes us, it gets ridiculous quickly.

In addition to a deep understanding of the rules and mechanics of using the code sets, medical coding requires a solid foundation in anatomy and physiology, medical terminology, pharmacology, and disease processes, along with the skill to interpret documentation that is often less than optimal. From my vantage point, coding is a seriously undervalued profession. I could go on at length about why I think that is, but a big factor is obviously a lack of understanding of what coding actually entails. Coders are entrusted with turning medical encounters into data. If our society values accurate healthcare data – for analyzing outcomes, establishing protocols, identifying disparities, allocating resources, reimbursing providers, and more – we should appreciate and support coders, and give them the time and tools they need to do the best job they can.