Is There a Data Utopia? CMS’ 2022 Advance Notice

The shift from RAPS to EDS for Medicare Advantage is a massive undertaking. With the current COVID-19 PHE taking place at the same time, CMS released Parts 1 & 2 early to give plans plenty of time to prepare. However, CMS has telegraphed this move for years and most of the Advance Notice for 2022 has already been covered throughout 2020 and 2021.

That being said, the latest announcement isn’t merely a rehash and there are still important points to cover within the 2022 Advance Notice. According to CMS, the shift from RAPS to EDS won’t affect risk score based payments. CMS has estimated the uniform transition from two submission systems to one won’t require significant adjustment. Additionally, they state that using 100% encounter data to calculate MA risk scores is projected to have a neutral impact of 0%. That’s a pretty big assumption, and it’s important to compare CMS’ projection with what health plans are seeing on the ground.

Diagram Source

In a perfect world with perfect data, there’d be no errors and CMS’ estimations would be correct. However, perfect data doesn’t exist and CMS’ 0% adjustment implies that, in the transition from RAPS to EDS, payers have managed to implement their EDS data without flaw. On the subject of the EDS Transition, CMS’ Advance Notice Part 1 states:

The CY 2022 impact on MA risk scores of the transition to a greater percent of the risk score being calculated with encounter data and FFS claims is 0.00%. In the CY 2021 Advance Notice, CMS projected the differential between the RAPS-based risk score and the encounter data-based risk score, calculated using the risk adjustment models proposed, to be 0.00%. Since the relative impact was 0.00% beginning in CY 2021 and CMS is proposing to calculate 100% of risk scores based on encounter data, the impact of the transition to a risk score based entirely on diagnoses from encounter data in CY 2022 is 0.00%. 

As shown in the Risk Score Breakdown below, the shift from v23 (RAPS) to v24 (EDS) models leads to higher overall risk scores for EDS than RAPS. That’s because the EDS model has higher demographic scores and lower HCC scores, but more opportunities to add HCCs and added Payment Condition Count (PCC) risk scores to make up any difference.

However, CMS’ decision is based on an average across the Medicare population. Plans would be wise to do a deep dive of their own data to see if CMS’ projections hold true for them. Because, from what we’ve seen, even a minor error rate can have real repercussions.

Living in the real world of data isn’t always as nice and pretty, especially for MA plans with high duals populations. The EDS model is beneficial for MA Plans and risk scores on average, but for plans with higher numbers of duals members, risk scores tend to trend lower due to the adjustments made to the the v24 model in which the risk scores specifically for community duals members were lowered substantially more in comparison to previous versions.

Thus, from a RAF perspective, while it’s true that there will be little impact on MA plans populations overall, plans’ duals populations will be impacted the most by the move to full EDS. As the yearly models have moved from a blend to 100% EDS, risk scores for duals members have significantly decreased. That’s a problem for every plan with any significant crossover.

Therefore, as the shift from RAPS to EDS continues, reevaluating imperfect data can provide opportunities. That’s because a low error rate around 3% can still have a significant financial impact on a plan’s bottom line. A financial impact that is forever lost, and thus the neutrality of the switch over to EDS is negated if data is not being corrected and getting through submissions.

The majority of the benefits conferred by the v24 model (increased mappings, more HCCs, and PCC) are only impactful if all of that extra data collected is put to good use. As covered in our recent webinar, “Strategies to Ensure Accurate Coding and Submission of Upstream Encounter Data,” management of upstream data can make or break this transition and stand in the way of your pursuit of perfect data.

Utopian data is great for the average MA risk score, but plans with high populations of duals members will need to re-evaluate because reality isn’t perfect. Even for plans that lack dual membership, the reality of imperfect data can lead to revenue losses. In the real world, health plans need to pay extra attention to where and how their data is coming in, or else they will see a further decrease. However, with the extra time afforded by CMS to identify and correct EDS issues, the dream of utopian data can continue.

For overburdened payers and providers, Episource helps close gaps in healthcare by marrying expert guidance with an end-to-end risk adjustment platform. Learn more about our solutions at Episource.com.