Full steam ahead: CMS finalizes new condition count model and risk score blend for Medicare Advantage plans

The Centers for Medicare & Medicaid Services (CMS) recently finalized its methodological changes to the Medicare Advantage (MA) risk adjustment model for the 2020 calendar year in the newly released MA and Part D Rate Announcement and Final Call Letter. Lesley Brown, Cotiviti vice president of product management for Risk Adjustment, breaks down the major model updates and risk score calculation changes that MA plans should know.

CMS goes with the alternative payment condition count model

CMS made a somewhat surprising move by announcing that the payment year (PY) 2020 hierarchical condition category (HCC) risk adjustment model will leverage the alternative payment condition count (APCC) model for blended risk score calculations. As required by the 21st Century Cures Act, this model focuses on the number of conditions an individual beneficiary may have, with the model adjusting as the number of conditions increases, and includes additional HCCs for dementia and pressure ulcers as noted in the table below.

HCC number HCC label PY 2020 APCC model rule
51 Dementia with Complications If this HCC is listed, then HCC 52 will drop
52 Dementia Without Complication Not applicable
157 Pressure Ulcer of Skin with Necrosis If this HCC is listed, then HCCs 158, 159, and 161 (Chronic Ulcer of Skin, Except Pressure) will drop
158 Pressure Ulcer of Skin with Full Thickness Loss If this HCC is listed, then HCCs 159 and 161 will drop
159 Pressure Ulcer of Skin with Partial Thickness Skin Loss If this HCC is listed, then HCC 161 will drop

This alternative model was proposed in the Advance Notice in December 2018 after industry feedback had suggested that the original payment condition count (PCC) model would not improve prediction in the CMS-HCC model for high-need beneficiaries with multiple chronic conditions. It was also accompanied by additional analysis by CMS to identify potential chronic non-payment HCCs that were under-predicted. The added HCCs for dementia and pressure ulcers met the conditions for inclusion in the model, as they are well-specified, predict medical spend, are definitively diagnosed, and indicate significant disease burden. The majority of stakeholders expressed support for this additional research undertaken by CMS and supported both the APCC model (versus the PCC model) and the addition of these new HCCs, indicating that these inclusions would result in more accurate predictions.

Cotiviti’s initial analysis suggests that implementing the APCC model for PY 2020 will work toward reversing the potential negative revenue impact of the PY 2019 model changes, which we previously outlined in this white paper. However, the increased normalization factor for PY 2020 (up from 1.041 to 1.075 for the RAPS/2017 model and from 1.038 to 1.069 for the APCC/EDS model) may minimize some of the risk score increase MA plans would otherwise realize from the APCC model changes.

The new risk score blend

CMS’s continued intent is to put these risk adjustment model changes into place over the next three payment years, ending in 2022. For PY 2020, CMS announced that risk scores will be calculated as follows:

  • 50 percent using diagnoses from encounter data, fee-for-service claims, and RAPS inpatient records under the APCC model
  • 50 percent using all RAPS records and fee-for-service claims under the 2017 CMS-HCC model

This continued transition comes despite several stakeholders recommending that CMS ease up on the transition to encounter data for PY 2020, suggesting a slower path of 33 percent calculated using the APCC model in PY 2020, 67 percent in PY 2021, and then 100 percent in PY 2022. For example, America’s Health Insurance Plans (AHIP) and the Blue Cross Blue Shield Association (BCBSA) both recently submitted letters to CMS supporting the use of encounter data, but requesting that the agency slow the transition until operational problems and uncertainties have been resolved.

Time to take action

Now that they have certainty about the new payment model, MA plans can take the following steps to help ensure their risk adjustment revenue is appropriately optimized in 2020:

  • Continue to talk with and train your provider network regarding appropriate documentation and coding, especially for these latest risk score model additions for pressure ulcers and dementia
  • Minimize the 50/50 blended risk score impact by ensuring your submissions teams are identifying and assessing any risk score disparities between RAPS and EDS
  • Ensure that your in-home assessment questionnaire includes appropriate coverage for the newly added risk-adjustable HCCs for both PY 2019 and PY 2020

Interested in learning how the 2020 Final Call Letter also impacts the Star Ratings program? Read our recent blog post and learn what you need to know.

In a frequently changing regulatory environment, Cotiviti's Risk Adjustment solutions provide deep subject-matter expertise to ensure risk-associated revenue is optimized while maintaining appropriate compliance. Learn more about our end-to-end prospective and retrospective risk adjustment services.

Read the solution overview



Lesley Brown
Lesley drives the product strategy for Cotiviti’s end-to-end Risk Adjustment solutions, ensuring that we continue to deliver innovative and compliant solutions for our customers. She has more than 30 years’ experience in product and project management in the healthcare and pharmaceutical industries. Before joining Cotiviti, she served as senior vice president of product management at Halfpenny Technologies, a leading provider of clinical data exchange solutions for healthcare providers. She also previously served as senior vice president of care management products for TriZetto, where she was accountable for the quality, delivery, and support of the software solutions in the its care management portfolio, and she held multiple positions at Alere Health. Lesley is the co-author of 10 U.S. patents and has been published in 16 peer-reviewed scientific and clinical journals.

Write a Response