2025 Medicare Advantage Advance Notice: Small Changes, Missed Opportunities
Pretty much a rewrite to provide simpler reading and a better understanding for readers. My rewrite did not make it much shorter. It is taken from a recent 2025 Health Affairs article. The main thrust of this article being commercial Medicare Advantage insurance companies taking advantage of government payments for healthcare to Medicare patients. The other part being CMS doing more to resolve the issues. MA Coding for patients is set the year before the care is given.
Ask questions and I will attempt to answer.
~~~~~~~~
January 31, 2024, has the Centers for Medicare and Medicaid Services (CMS) releasing the Calendar Year (CY) 2025 Advance Notice for Methodological Changes for Medicare Advantage, Capitation Rates, and Payment Policies (the Advance Notice). The agency was accepting comments on the Advance Notice through March 1, 2024. It will publish the final rate announcement by April 3, 2024. This annual regulatory process presents a key vantage on CMS’s priorities for improving the Medicare Advantage (MA) program. It also generates a great deal of attention from MA plans due to its implications for their future payment rates.
Just so you know the differences between Traditional Medicare and Medicare Advantage. Capitation is used by MA plans. CMS pays a fixed amount to providers based on the number of patients they have, see using the coding for the patients. Capitation is determined before care is give. Fee-for-service (FFS) healthcare is based on the procedures or services providers perform at the time of service. Both these systems are used in the U.S. healthcare system.
As reported, there are no drastic changes proposed for the MA program in this year’s Advance Notice. Such was widely anticipated. The Biden administration may have wanted to avoid the appearance of Medicare cuts in an election year. The Advance Notice includes some important provisions, while leaving out other potential reforms having a lasting impact on MA. Health Affairs is providing a high-level explanation of what CMS has proposed. It is highlighting potential modifications before the final rule is released (April 3).
The Net Payment Impact On MA Plans Is Limited
The most important estimate in the latest Advance Notice is an expected change in revenue for MA plans. For CY 2025, CMS estimates that there will be a 3.7 percent average expected increase in revenue for MA plans. This is driven by increases in MA risk score (which MA can overestimate) trends over time and increases in the effective growth rate. It reflects expected increases in the national fee-for-service Medicare per capita costs used to set MA benchmarks. Payment factors based on MA plan star ratings and the Hierarchical Condition Category (HCC) risk-adjustment model will decline in 2025. However, these factors are offset by the increases in other payment factors. It has been reported these provisions amount to a 0.2 percent cut in benchmark payments for MA plans. However, total revenues are still projected to increase.
Meanwhile, MA plans and lobbying organizations portray these proposed payment updates as major cuts. This is not true. To put the 2025 MA rate changes in perspective, the 3.7 percent increase in revenue is larger than last year’s 1 percent predicted increase. It is smaller than the previous years’ increases exceeding 7 percent annually. The extremely high margins (due to over coding by MA plans) will barely be torched by the Advance Notice provisions.
As Health Affairs points out for Medicare Advantage,
“No Major Changes to Risk Adjustments with Key Reforms Left on The Table.”
As I briefly mentioned above, MA plans may manipulate risk adjustment to earn higher payments by making beneficiaries appear to be sicker through intensive, or exaggerated, risk coding. Clinically unsupported risk coding is a form of health care fraud and a major source of overpayments in Medicare. This is nothing new as Kip Sullivan has reviewed such manipulation repeatedly in his dialogs shown here on Angry Bear.
In 2023, the Medicare Payment Advisory Commission (MedPAC) estimated that MA plans’ excess coding intensity resulted in MA risk scores that were about 10.8 percent higher than scores for similar traditional Medicare beneficiaries. The result being $17 billion in excess payments to MA plans in 2021 and an estimated $23 billion in 2023.
Addressing the problem of intensive risk coding in MA, Congress first authorized and mandated the action for CMS to apply a coding pattern adjustment to account for MA plans’ intensive efforts to document and inflate risk coding. The Affordable Care Act and the American Taxpayer Relief Act established a schedule of minimum adjustments starting at 4.9 percent in 2014 and standing at 5.9 percent today. Nevertheless, CMS to date has not exceeded the statutory floor of 5.9 percent. In 2023, applying the 5.9 percent coding pattern adjustment meant that MA risk scores were still about 4.9 percent higher than they would have been in traditional Medicare.
It begs the question of why can’t Medicare Advantage plans reach the same costs as Traditional Medicare? The Overhead and profit expectations are high.
In the 2025 Advance Notice, the year-to-year percentage change the MA coding pattern adjustment is 0 percent from the 2024 Rate Announcement. This means CMS will maintain the coding pattern adjustment at the statutory minimum of 5.9 percent for 2025. By not increasing the coding pattern adjustment, CMS is leaving substantial risk-code inflation unaddressed and leaves billions of dollars in MA overpayments on the table.
In the 186-page Advance Notice, CMS offers no justification for its decision. These levels can be justified based a CMS’s internal analysis as required by statute. However and “again,” these analyses have never been released to the public. Health Affairs strongly believes CMS should publish these analyses and their methods. The public can validate if the approach is successfully accounting for MA plan coding intensity. The one thing I will say here, “if there is a knowledge public which can validate CMS actions.”
This year, CMS proposed to continue to phase in changes to risk adjustment announced in last year’s CY 2024 advance notice. These changes removed some codes from the HCC model, widely believed to erroneously inflate risk scores. CMS’s proposed changes for CY 2024 were reasonable given the extensive literature documenting MA payments as being inflated due to variations in coding intensity and potential over coding of MA beneficiaries.
The removals resulted in a massive outcry from the commercial plans. MA Plans countered with a misleading 2023 Super Bowl ad about impending cuts. When the 2024 rates were finalized, CMS backed off and elected to phase-in the new model over time.
While a step in the right direction, there are other ways for CMS to get differential coding under control which is not being considered.
CMS should consider increasing the coding pattern adjustment overall. CMS could implement a customized adjuster discounting plan risk scores in proportion to coding intensity. The authors believe CMS has the authority to implement a plan-specific coding adjustment. CMS could consider reducing the ability for plans to add extensive coding through chart reviews or health risk assessments. The authors work, MedPAC, and others finds the plans differentially use these methods which can lead to unsupported risk coding. Countering this, CMS should consider limiting the use of health risk assessments and chart reviews in an effort to reduce overpayment to plans nationally.
Fee-For-Service Normalization Factor Took a Bite Out of MA Payments
For CMS to set MA payment in a given year, it must predict expected expenditures and expected risk in Traditional Medicare. As MA risk scores are assessed relative to traditional Medicare risk. To understand MA risk, and to pay accordingly, CMS must accurately profile traditional Medicare risk.
Because CMS doesn’t know what Traditional Medicare risk will be in the following calendar year, it uses a strategy to predict risk. The predicted change to Traditional Medicare risk is known as the fee-for-service normalization factor. The normalization factor is important because it is used to adjust the coefficient values for HCCs. Suppose CMS expects the risk of the traditional Medicare population to increase by 5 percent. In that case, the fee-for-service normalization factor will be set at 5 percent and HCC coefficients will be down-weighted by 5 percent.
In recent years, CMS has used various strategies to estimate the fee-for-service normalization factor. The COVID-19 pandemic has made this more challenging because it depressed use (and risk capture) in traditional Medicare, giving the false impression of lower severity among traditional Medicare beneficiaries. In the Advance Notice, CMS proposes a new approach to calculate the normalization factor to account for pandemic-related disruptions.
To summarize this approach, CMS estimates a prediction model using five-year-level observations between 2019 and 2023. The dependent variable was Traditional Medicare risk. Independent variables were a linear time trend and a dummy variable indicating whether an observation occurred after the COVID-19 pandemic. CMS then plugged the estimated coefficients into the model, advanced the time trend by one year, and predicted a normalization factor of 1.045 for calendar year 2025.
It is unknown whether traditional Medicare risk scores will increase by 4.5 percent in 2025. We agree with CMS that it is essential to account for the COVID-19 pandemic when predicting traditional Medicare risk. In the future, CMS should consider using beneficiary-level information from the likely population of Traditional Medicare beneficiaries in a given calendar year. CMS could use estimates of historical changes in risk; based on demographic, clinical, and market factors, including MA penetration. This could predict changes in risk at the beneficiary level, before aggregating nationally. By taking advantage of rich beneficiary information as opposed to national averages, this approach would be more robust and less likely to require annual tweaks.
In addition, recent work on favorable selection suggests the traditional Medicare population is unobservably healthier than the MA population (over coding?), leading to large overpayments to MA plans. Estimates of favorable selection in Medicare Advantage could be used to increase the fee-for-service normalization factor further. What has been happening is many of the ill come back to Traditional Medicare due to lack of care or “?” with MA plans.
CMS Should Do More
Overall, the changes in this year’s Advance Notice appear to be reasonable and stop short of a wider range of reforms that CMS would have ample justification to make. CMS should do more to ensure that the MA program is functioning properly and delivering the value that it promises. Insurers may object to payment changes that trim their profits, but we have heard that before.
After last year’s Advance Notice, payers predicted that the more substantial changes would destroy the program and curb its growth Such predications have not come to pass. As the Medicare Advantage program continues to grow, it will be essential to ensure that it is delivering on its goals for Medicare beneficiaries. This Advance Notice includes some improvements to the program. In the end, more opportunities for meaningful policy change remain.
2025 Medicare Advantage Advance Notice: Small Changes, Missed Opportunities
CMS Releases Proposed Payment Updates for 2025 Medicare Advantage and Part D Programs