Should ICD-10-CM Feature Codes for Every Genetic Condition?

It may become common practice to identify genetic abnormalities prior to the development of manifestations.

Any of you who know me know that I wholeheartedly embraced ICD-10-CM. I love the granularity it offers. When I teach, I try to get providers to surmount the “hASSLe” and give us Acuity, Severity, Specificity, and Linkage so we can derive the correct best code.

The March 2023 ICD-10-CM Coordination and Maintenance Committee Meeting presented me with a dilemma (Topic packet March 8, 2023 ICD-10-CM Coord & Maint Cmte Meeting). Now that we have the technology to unravel the genetic basis for conditions, are we obligated to create unique codes for each variant?

There were at least eight meeting topics addressing gene-related conditions, many of which were developmental and epileptic encephalopathies. During the meeting, the initial presentation centered on the clinical aspects of the disorders and manifestations, as well as the genetic basis. Then personnel from the National Center for Health Statistics (NCHS), a subgroup of the Centers for Disease Control and Prevention (CDC), offered coding options and solicited comments.

Illustrative of the topics eliciting this type of discussion were glutamate receptor, ionotropic (GRI), gene-related neurodevelopmental disorders. The request for new ICD-10-CM codes was brought to the CDC by the CureGRIN Foundation. This is not unusual – stakeholders who requests codes vary from individual physicians to professional societies to commercial companies to disease organizations. In fact, it is not uncommon for a family member of an afflicted patient to speak passionately to the committee, imploring agreement with creating the new code or codes.

The way it was explained for these GRI codes is that there is an overarching family of ionotropic glutamate receptors in the brain, and specific genetic mutations result in different neurological disorders. These related disorders present differently, including issues such as autism, seizure disorders, or intellectual disabilities. The incidence for each is approximately 2-5 per 100,000 births. The request was for nine distinct codes identifying each abnormality and presentation.

The two coding options were creating a new sub-subcategory for glutamate receptor-related neurodevelopmental disorders in the F section, Mental, Behavioral, and Neurodevelopmental Disorders, or creating a set of “genetic susceptibility to epilepsy and neurodevelopmental disorders, related to glutamate receptor genes,” with specific codes for each variant. This approach was repeated for other genetic disorders being addressed during the meeting.

Human beings are believed to have between 20,000 and 25,000 genes. A gene can have 100 to millions of deoxyribonucleic acid (DNA) base pairs, so there are more than 3 billion base pairs in the human genome. Different DNA mutations can cause the same phenotypic presentation, or mutations at a different place in the same gene may cause variations of a disease process. There are base substitutions, deletions, and insertions. Point mutations result from single base substitutions. The bottom line is that there is the opportunity for many, many genetic variations that can cause survivable, clinically significant disorders.

Why do interested parties want unique ICD-10-CM codes for specific conditions? It is hard to know how prevalent a condition is without being able to identify patients who have it. Prevalence guides resource allocation. Research relies on being able to identify subjects. Analyzing data and being able to determine which therapies are effective on which conditions is facilitated by having unique codes. Tracking patients through the healthcare experience and assessing outcomes like morbidity and mortality is enabled by having an entry in the code set.

What are the objections to making specific codes for every condition identified? Although there is potential for billions of codes with seven alphanumeric characters, that isn’t really actualized, because certain characters have limitations (e.g., the initial character is always and only a letter). The first letter establishes the category in which the condition is housed. For instance, E is for endocrine, nutritional, and metabolic diseases. If a genetically induced condition results in a metabolic disorder, it would belong here. If you were to run out of E codes, you can’t just roll over into another letter. Also, ICD-10-CM tries to stay close to the World Health Organization’s (WHO) ICD-10 – it can’t just go completely rogue.

I think the most problematic issue is getting providers to avail themselves of ICD-10-CM’s specificity. The researchers petitioning for the new codes would be eager to utilize them, but the average clinical practitioner might not be as excited to scroll through hundreds of offerings to find the most specific code. If we can’t get providers to give us the specificity of malnutrition, what makes us think they are going to get them to tell us “obesity due to disruption of MC4R pathway, nuclear receptor coactivator 1 (NCOA1) gene mutation?”

But if they don’t have those codes as an option, they can’t provide the specificity. This certainly thwarts research and epidemiological surveillance.

In my opinion, NCHS needs to determine its philosophical approach and apply it across the board. First, they need to decide if there should be an incidence or prevalence threshold. Is there another factor like risk of mortality, likelihood of passing a condition on to others, or utilization of public resources? Or are they just going to make new codes for any organization or individual who takes the time to submit a request?

Then, how do they do it? Is it for the overarching disorder, or for each individual genetic abnormality that can cause it? If a single base pair mutation causes a very clinically significant or common condition, it makes sense to me. But every single one?

I do not subscribe to only using a “genetic susceptibility” code to pinpoint these rare disorders. Having the susceptibility does not necessarily mean you are expressing the disorder/syndrome/condition/disease. Some diseases have variable expression. It may become common practice to identify genetic abnormalities prior to the development of manifestations. We have that technology available to us right now, like knowing you have the gene for Huntington’s disease prior to the development of symptoms. It may become customary to sequence your genes and know what might lie in store for you.

If that is going to be the approach to the specific genetic defect, I would recommend using a code identifying the manifestation of the genetic mutation: either a nonspecific condition, like “autistic disorder,” or a somewhat specific one, like “glutamate receptor-related neurodevelopmental disorder.” Then, they should add a “code also” instruction for the hyper-specific genetic susceptibility Z15- code identifying which genetic abnormality resulted in the condition. If there are manifestations other than the primary one, as above, those should be additional codes, e.g., deafness, epilepsy, intellectual disabilities, etc.

I could see Z15-, Genetic susceptibility to disease, quickly becoming overwhelmed. As an aside, I find it curious that BrCA 1 and 2 don’t have their own specific codes (they are lumped together in Z15.01, Genetic susceptibility to malignant neoplasm of breast), but we are requested to create a code: Z15.140, Genetic susceptibility to epilepsy and neurodevelopmental disorders related to pathogenic variant of SLC6A1 gene. Do they just keep adding more genetic susceptibility codes as new mutations are discovered?

I don’t really have the definitive answer to this. If you have an opinion or suggestion, now is the time to offer it. The deadlines for comments are available in the topic packet linked in the second paragraph above (depending on when the code is being considered for implementation).

Be a part of the process!

Programming note: Listen to Dr. Erica Remer every Tuesday when she cohosts Talk Ten Tuesdays with Chuck Buck at 10 Eastern.