[HUM, ALHC, UNH, CVS, ELV, AGL] Expectations for CY26 Medicare Advantage Advance Notice
KEY TAKEAWAYS: We expect the proposed YoY change in the CY26 benchmarks for Medicare Advantage (MA) plans is likely to be relatively flat, with a range between -1% and +1% that excludes coding trend, but the finalized YoY change is likely to end up in a more positive territory in a neighborhood of 2-4%.
With the changeover of administrations taking place two weeks before the anticipated due date for CMS’s CY26 MA Advance Notice (AN) and the Office of Management and Budget’s (OMB) review of that proposed rate notice having begun, we highlight the extent to which Biden’s CMS pursues a cut to MA payments, and the levers that the Trump CMS can use to ameliorate the any cuts to the CY26 final MA rates that are to be announced by April 7. We note that these CMS releases are separate from the agency’s technical regulation changes for MA and Part D, in which coverage of obesity treatments as well as tweaks to marketing and medical-loss ratio were proposed.
While the Biden administration could take a page from the Trump administration in its first term and issue the AN in the waning six weeks of its term, especially since OMB vetting has already begun, we think CMS is more likely follow past protocol in its issuances of regulation. Statute requires that the AN be issued at least 60 days before the MA rate announcement’s due date, which is the first Monday in April (April 7). That would bring a due date for the AN to February 5 or in the days just before.
While technically the Trump administration will have begun by the time this AN is released, we suspect that it will largely be the work of Biden’s CMS, due to the tight turnaround, OMB review already underway, and desire to secure political benefits by providing relief in the final.
Key Substance of the CY26 Advance Notice
We suspect that the proposed YoY MA benchmark change will likely range between -1% and +1%, excluding coding trend. As in prior years, CMS is expected to telegraph a higher benchmark amount because it will be accounting for coding trend, pushing the proposed rates higher than what insurers deem as accurate.FFS Growth Rate. We think the range for the projected CY26 fee-for-service (FFS) growth rate is 2-3.5%. CMS is likely to utilize the methodologies and timeframes previously used to calculate the per-capita FFS growth rates (Medicare claims data from enrollees with Part A or Part B during a rolling five-year dataset with the final year reflecting the claims experience and cash activity for the first three quarters). For CY26 AN, that means CMS will use data from 2020-24, with the final year’s input referring to the claims experience and cash activity for Jan. 1-Sept. 30, 2024. This dataset will likely incorporate the higher medical utilization trends insurers and hospitals reported during late 2023 and most of 2024. It will also likely include prior-
year adjustments, and ongoing growth in outpatient medical and behavioral care at the expense of more costly inpatient treatments and in dual eligibles selecting MA instead of FFS. The elimination of inpatient medical education costs in MA is likely to account for as well.
Inpatient Medical Education Adjustment. We think CMS will proceed with its policy of applying a technical adjustment to FFS growth to account for the agency’s phasing out of MA-related inpatient medical education costs. This policy depresses the FFS growth rate (for CY24, the impact was -0.86% and for CY25, -0.53%). CMS’s plan has been to eliminate these costs over three years, with CY26 being the final year. For CY24 rates, CMS applied 33% of the adjustment and, for CY25 rates, applied 52% of the adjustment for CY25, instead of the 67% that was scheduled Since CMS did not specifically state that it is extending the phase-in period, we expect it will stick with the three years and seek to propose to implement the final 48% of the adjustment for CY26; plan B would likely be extending the phase-in and increasing the adjustment further.
V28 Risk Adjustment Model. We expect CMS to implement the final year of its three-year phase-in of v28 risk adjustment model, which started with CY24. For CY25, the agency estimated that the v28 implementation would result in a 4.44% hit to the benchmarks, though it was partially offset by ~2% from its revision of the normalization factor methodology, which we doubt would be further altered in the AN. CMS did not project the impact of v28 changes in CY24. We suspect the CY26 benchmarks would be lowered as well.
Other Policy Risks. We are skeptical the Biden CMS would use the AN to take a hatchet to ongoing excessive spending concerns and propose to utilize longstanding authorities not employed to date such as: 1) imposing a higher coding intensity cut; 2) excluding diagnoses derived from in-home risk assessments and chart reviews in risk adjustment scores; and 3) using two-years of diagnostic data in risk adjustment. Rather, it will likely defer to the policies being phased in to address overpayments.
Levers for Amelioration in CY26 Final Rate Announcement
While there has been some noise about legislative changes the Trump administration and Congress could pursue to put the finger on the scale and favor MA vis-à-vis FFS and derive federal savings, Trump’s CMS can take some actions on its own to shore up fears about plan exits, loss of care access and supplemental benefits, and higher premiums by finalizing some improvements from current policy and the anticipated AN. As such, we think the final CY26 YoY change in the MA benchmarks ends up settling between 2% and 4%.
The tools CMS has at its disposal, which could lead to higher MA benchmark calculation, include:
- Changes to the FFS growth methodology and dataset like: 1) narrowing the FFS growth dataset to a shorter period than the current rolling 5-year timeframe (for example the three-year dataset used in the ratebook for the ACO Reach demonstration model); 2) broadening the incurred claims experience and cash activity of the final year in the dataset to reflect changes that are happening in the fourth quarter of that year; and 3) counting claims data from enrollees with both Parts A and B instead of enrollees with either Part A or B. However, the statutory authority to do this is unclear and some claim passing legislation is required,
- Pause or extension of the v28 risk adjustment model implementation. CMS could claim that they are evaluating the prior policy changes to assess the impact of the model on utilization, care access, benefit design, certain types of patients, and offering of supplemental benefits. There is precedent where the first Trump administration tweaked and slowed down previous risk adjustment changes. We think it is less likely that CMS cancels adoption of this revised risk adjustment model and reverts to the prior model for CY26.
- Reverse, pause or extend the phase-in of the technical adjustment to eliminate the MA-related inpatient medical costs.
- Alter the revised normalization methodology to exclude the FFS risk scores during the COVID-19 pandemic, which stakeholders previously argue distorted the average risk score.