Every patient deserves to receive exceptional care – regardless of the costs. But with over 31 million Americans enrolled on Medicare Advantage, how can providers ensure they anticipate and manage long-term healthcare expenses – especially given how different patients’ requirements are?
The answer is CMS-HCC risk adjustment – and this article provides a complete overview of how it works.
The CMS-HCC risk adjustment model is a methodology, implemented by the Centers for Medicare and Medicaid Services (CMS), intended to meet two requirements:
The cost of long-term patient care varies wildly based on individual requirements, and this creates a major hurdle for providers that work with Medicare Advantage patients. If all MA patients received the same funding, taking on higher-cost patients would lead to potential losses for providers – and create an incentive to only onboard lower-cost patients.
Risk adjustment is designed to ensure this does not occur. The CMS-HCC risk adjustment model adjusts Medicare Advantage reimbursements based on individual patients’ expected care costs and requirements. The CMS establishes a baseline cost for each county and calculates how much each individual patient will vary against that baseline – offering more funding to providers that take on patients with above-average requirements.
Using this baseline allows the expense of higher-cost patients to be offset by lower-cost patients – ensuring all can receive the treatment they need. Providers can confidently take on any MA plan, safe in the knowledge that extra services and care costs will be covered through its risk adjustment model.
The CMS-HCC model uses two categories of data to calculate Individual patient risk:
These two factors are combined to produce a risk adjustment factor (RAF) score which determines the patient’s MA reimbursement. The RAF score is benchmarked at 1; any variation from this baseline score will result in a change in funding. For example, an RAF score of 1.25 will lead to 25% higher-than-average reimbursement for the provider.
But before we explain how RAF scores work, we need to understand how medical diagnoses are recorded and submitted to the CMS.
The CMS has developed an extensive list of medical codes known as hierarchical condition categories (HCC), which map over 70,000 specific diagnoses to 115 distinct categories. For example, HCC category 19, Diabetes with Chronic Complications, has 400 different diagnoses attached to it.
Healthcare providers undertake HCC coding to record each individual patient’s relevant HCC codes; they then submit the patient demographic data and HCC codes to the CMS.
Once the CMS has your patients’ demographic data and HCC codes, they can calculate the patient’s RAF score. The RAF score is the sum of the patient’s:
The CMS calculates the total risk score of all submitted HCC codes and adds it to the baseline demographic risk score; this is the patient’s RAF score.
The CMS HCC model was first introduced in 2004 and has remained relatively stable since then. While there have been adjustments to the risk scores associated with each HCC code, the scores tended to change relatively little. A study from 2018 showed that recent updates had led to an average increase of 0.78%.
Of course, even such small changes can have a dramatic impact on Medicare reimbursements, and providers should always be aware of the shift. But it was only when a wholesale shift to a new model, known as V28, that many providers really understood how much was at stake:
All of which is expected to decrease risk scores anywhere from 2% to 16%, depending on the patient’s condition.
These changes have been phased in gradually, but they will come into full force in 2025. This has placed a new level of pressure on providers to understand and improve their HCC coding – to avoid the common mistakes that lead to lower-than-expected reimbursements.
Accurate risk adjustment requires specificity, but many providers have gaps in their patient data. The data may be stored in multiple digital systems that are not interoperable; in unstructured notes that are not easily available; or even in other healthcare organizations which the provider cannot access.
This not only negatively impacts care; it makes it harder to complete accurate HCC coding and therefore receive the full reimbursement you are owed.
Even when providers do have access to a complete medical history, they may make errors in the HCC coding. Individual providers may be required to code for diagnoses that are outside of their area of expertise, which leads to a lack of specificity – and can lead to the wrong HCC code being used.
Both issues discussed above are exacerbated by a lack of time and resources to complete HCC coding. Providers are already over-worked and want to focus on patient care; the extra effort of manual coding can lead to understandable errors or even incomplete coding.
HCC Assistant is an innovative tool that uses natural-language processing (NLP) to ingest huge quantities of patient data and make HCC coding recommendations at the point of care, with 97% accuracy. Providers can simply review and accept the codes they believe are appropriate, while rejecting any they don’t agree with.
As a result, providers save hours of manual effort while improving the accuracy of their HCC coding – and increasing their RAF scores by an average of 35%.
