Before we begin, I think it’s important to mention that while not every facet of every update from CMS impacts every organization the same way, we are firm believers in an all encompassing risk adjustment strategy. Doing so results in a strong ROI, as well as stabilizing and protecting revenue through complete, accurate risk adjustable condition capture and therefore benchmarking.
These updates are written in the same spirit, to provide that broader understanding to allow for greater success for your organizations. With that said, let’s move forward.
Coding Pattern Adjustment and MAO Revenue Growth
The revenue growth for Medicare Advantage Organizations (MAO) in PY 2022 will be 4.08%, which is almost double the expected rate of 2.82% from the Advance Notice, and an even greater increase from the 1.66% in PY 2021.
The coding pattern adjustment is set to the statutory minimum of 5.9%, like last year. CMS noted, “While we consider our long standing approach appropriate for 2022, the question of whether the methodology to calculate MA capitation rates for future years (2023 and/or beyond) should be revised in ratesetting or rulemaking to include FFS spending data only for beneficiaries enrolled in both Part A and Part B is one we may further consider, particularly in a situation where we consider possible future changes to the MA coding pattern adjustment and whether such a change in that policy, if any, should be considered in tandem and implemented on the same timeline. However, we decline to adopt such a revision at this time but may consider soliciting further comment on those matters in the future.”
RAPS and EDPS Transition
As mentioned, the transition from a risk adjustment processing system (RAPS) to the encounter data processing system (EDPS) is heading towards its conclusion. Encounter dates for 2021, for payment year 2022, on a final submission deadline of January 2023, are to be submitted under the V24 Medicare Advantage model exclusively via EDPS. All prior encounter-date submissions will still use the V22 and V24 MA models and can leverage both EDPS and RAPS.
This transition has been in-progress for some time now, and we plan to support RAPS and EDPS for a while, particularly within Lumanent Retrospective Review, as internal audit projects may require both formats.
The larger context: The ramifications of risk adjustment are often overlooked in this transition, with the focus often on technical necessities. Today, risk adjustment values are calibrated based on fee for service data and not on the actual cost of care for the risk adjusted populations they’re applied to under value-based care. This transition to EDPS will allow for new, population specific models that are more accurately calibrated, and thus expected to more closely predict the costs and needs of populations within the applied models. The goal is fairer, more accurate payments to risk bearing organizations. While implementation of a new model exclusively based on EDPS-based diagnosis data is still some years away, the benefits to organizations bearing risk, and the patients themselves are clear: future models built through EDPS can more readily begin to recognize, codify, and adjust for the needs of the populations themselves, all of which are large enough to be statistically relevant. As a result of that specificity, there will also likely be reduced attempts to “game” the system itself, leading to more accurate payments.
Synopsis of Model Updates
In each case, where applicable, Health Fidelity has made appropriate adjustments to our solutions and our NLP engine for the PY 2022 model, as outlined at a high level below.
As predicted in the Advance Notice, the CMS-HCC model will transition to only the Alternative Payment Condition Count (APCC) model (a.k.a. “v24”, or the “2020 CMS-HCC” model), i.e. there will no longer be a blend with the 2017 CMS-HCC model for risk score calculation. Note: We will dive into this more thoroughly in the next section.
As predicted in the Advance Notice, the RxHCC model will be recalibrated based on 2017 diagnosis data projecting 2018 expenditures. The PY 2021 model had been based on a 2014/2015 risk category coefficient calibration.
Risk Score Calculation & Submissions for PY 2022
As proposed in the Advance Notice, CMS will calculate the risk score for PY 2022 using only the APCC model:
- 100% of weight from: Encounter data + FFS data
- The risk score is calculated using the AAPC model with a normalization factor of 1.118.
As proposed in the Advance Notice, the RxHCC model is being recalibrated based on 2017 diagnosis data projecting 2018 expenditures. The current model (first devised for PY 2020) was based on a 2014/2015 calibration. As with any recalibration, the coefficients of the risk categories in the model will be updated. For now, though:
- 100% of weight from: Encounter data + FFS data
- The risk score is calculated using the same RxHCC model categories and model hierarchy as the 2020 RxHCC model version from PY 2021. The normalization factor is set to 1.043.
As a final thought, it’s important to echo the point. Strong risk adjustment leads to positive outcomes, for revenue and patients alike. With that, we want to offer our free ebook, Pulling Both Levers: A Four Year Analysis of Medicare Cost and Risk Adjustment. Regardless of your organization’s specific model, population, or risk adjustment strategy, the principles and data outlined are an excellent resource.