The CMS-HCC V28 Model

In early 2003, the Centers for Medicare and Medicaid Services (CMS) announced the adoption of a new Hierarchical Condition Category (HCC) risk model. Encounters with dates of service up to December 31, 2022 were subject to risk score calculation based on Version 24 (V24) of the CMS-HCC model. Risk scores for encounters with dates of service on or after January 1, 2023 are now, in part, subject to calculation based on the new version (V28).

Given the significant change in going from V24 to V28, this caused a lot of anxiety and turmoil in risk adjustment circles. I specifically, and Secondwave generally have received many questions regarding the shift in the risk model and I thought it might be useful to share a few facts about the new risk model and its potential impact on various entities operating within risk adjustment.

What’s new?

1. The most significant change is the exclusion of 2000+ ICD-10 codes form V28 compared to V24. These diagnosis codes were observed by CMS to be reported much more often on Medicare Advantage (MA) members, who are subject to risk adjustment, compared to Fee for Service (FFS) Medicare members who are not subject to risk adjustment.
2. By comparison, less than 270 ICD-10 codes were added to V28 that were not part of V24. The majority of these codes pertain to acute illnesses and conditions specific to infants.
3. The number and order of HCC categories was changed. V28 has 115 categories, while V24 had 86. The numbers assigned to each category were also changed, for example, diabetes with chronic complications was HCC18 in V24 and is now HCC37 in V28.
4. The weight (Risk Adjustment Factor or RAF) for each category was changed. For some categories the weight increased (cirrhosis of liver went from 0.363 in V24 to 0.447 in V28) and for other categories the weight decreased (morbid obesity went from 0.250 in V24 to 0.186 in V28).
5. Some hierarchies were changed or eliminated, for example, diabetes with acute complications and diabetes with chronic complications now risk adjust at the same weight as diabetes without complications.

What’s the result?

1. Given the significantly higher number of diagnosis codes removed from the model compared to the number of codes added (close to tenfold), there is no doubt that there will be a negative impact to the bottom line of all entities operating in risk adjustment. That is just a simple fact that we all must come to terms with.
2. The negative impact will be softened, to some extent, by the increase in weight for several categories that include prevalent, clinically significant diagnoses.
3. CMS agreeing to phase in the switch over the course of 3 years, rather than abruptly, will also allow entities to adjust and prepare for the impact of the change. For payment year (PY) 2024, risk scores will be weighted at 67% based on V24 and 33% based on V28. That ratio will reverse for PY 2025 to 33% V24 and 67% V28. For PY 2026, 100% of risk scores will be weighted based on V28.
4. Organizations who focused their risk adjustment activities on diagnoses that risk adjust, rather than those with clinical relevance, will see the highest negative impact to their bottom line as a result of the shift to the new model. Entities that directed providers over the years to alter their documentation habits in order to assign diagnosis codes that risk adjust over those that the patient actually has, and those that focused on capturing diagnosis codes that inaccurately increased patients’ perceived severity of illness will see the most significant drop in their population’s risk scores.

What can be done about it?

1. There is a lot of swirl in risk adjustment circles about strategies to “mitigate” the effects of the shift from V24 to V28 and the negative impacts to members’ risk scores. Simply put, there is nothing that can be done to mitigate or reverse the negative impact. Organizations should focus their efforts on embracing the change and building their financial models and predictions based on the gradual shift to the new risk model over the course of the next year or two.
2. Risk adjustment entities should also shift their focus back to clinical relevance when it comes to provider education. The messaging to providers should remain to address every active diagnosis with the patient as part of a face-to-face encounter at least once during each calendar year, and to document the diagnosis to the highest level of specificity along with an assessment and plan.

I, as a clinician, welcome the switch and feel like it’s way overdue. We needed a course correction that would shift providers’ time and effort from documenting diagnoses without clinical relevance just because they risk adjust, back to managing conditions that have real life implications to patients’ health and well-being.

At Secondwave, our prospective program is built around improving providers’ performance in risk adjustment and quality, but at the same time, any information we present to providers is vetted by clinicians to confirm its clinical validity and relevance. We never subscribed to the trend of documenting diagnoses purely for the sake of risk adjustment and inflating a member’s risk score. For that reason, although we and our clients will be negatively affected by the switch to V28 like everyone else, the impact will be softer than to those entities who relied heavily on the diagnoses that were removed from the risk model to sustain their populations’ elevated risk scores.

Osamah El-Aroud MD, CRC
Chief Medical Officer