Ending Medicare Advantage Overpayments

I missed this piece from May of this year. Been too busy trying to maintain a blog rather than write or find excellent articles to present at Angry Bear. I apologize for the later. Me writing? you just have to live with it for now till more authors become a regular occurrence here.

This piece is authored by Micah Johnson, Donald Berwick, and Richard J. Gilfillan. May 2024. It is featured at CAP. It does cover quite a bit of territory and issues we have presented as authored by Kip Sillivan, Maggie Mahar, Doctor Berwick, Juan Cole, Nancy Alt, Andrew Sprung, etc. It has been offered in detail at Angry Bear for years. It is now (what a surprise) that Medicare Advantage costs and over charging Traditional Medicare funds is being noticed. It is a long read and there is another part to it which I will post tomorrow.

A new Center for American Progress (CAP) analysis finds that Medicare Advantage (MA) plans are overpaid by 22 to 39 percent, The savings from curbing MA overpayments could and should be used to improve the Medicare program for all beneficiaries (The 39% is not what others have talked about). Ignore the foot note numbering. And this is a C&P.

Ending Overpayment in Medicare Advantage, Center for American Progress

In 2023, a record high of slightly more than half of eligible Medicare beneficiaries were enrolled in MA plans.³ MA overpayments are a key driver of increasing MA enrollment: High payment levels attract insurers to the MA market and enable the funding of supplemental benefits that, in turn, attract beneficiaries. These high payments also enable MA plans to finance advertising campaigns to further build their market share.⁴ The combination of rapid growth in MA enrollment and overpayment to MA plans poses a serious threat to the financial sustainability of the Medicare Hospital Insurance Trust Fund.

Overpayment to MA is more than just a fiscal problem; it undermines the goals of the Medicare program itself. An accompanying Center for American Progress proposal, “Medicare 2.0,” contends that the goals of the Medicare program should be to guarantee simple and comprehensive coverage, ensure affordable out-of-pocket costs, modernize prescription drug coverage, strengthen primary care, and promote population health and health equity.⁵ MA overpayment undermines several of these goals: Inflated MA payments increase Part B premiums for beneficiaries across Medicare, contribute to greater fragmentation and complexity in health care coverage and delivery systems, divert attention and resources to medical coding tactics rather than improving clinical care and population health, and appear to exacerbate at least some health disparities.

MA overpayment also threatens the long-term viability of traditional Medicare, which serves as a critically important lifeline for tens of millions of Americans who need a program they can depend on for the rest of their lives. Traditional Medicare’s broad access to providers is especially helpful for people with complex chronic conditions. A thriving traditional Medicare program would further important policy goals, including lowering Medicare’s administrative costs, providing more resources for clinical care, and offering coverage with public accountability to protect against harmful restrictions on care. Finally, MA overpayment consumes billions of public dollars that could instead be spent improving benefits and lowering costs for all Medicare beneficiaries.

A combination of calculated insurer behaviors and policy design flaws leads Medicare to pay more when a person enrolls in MA rather than traditional Medicare. Two key overpayment contributors, as detailed in this report, include upcoding, which makes patients appear sicker than they actually are, and selection bias, which results in MA plans enrolling more profitable patients.⁶ MA quality bonuses and county adjustments unnecessarily boost payments to plans even further, as does the inflationary effect of supplemental coverage on traditional Medicare spending benchmarks, which serve as the basis for comparing MA bids.⁷ Accordingly, a new CAP analysis presented in this report estimates that MA plans are overpaid by 22 to 39 percent, with overpayments in 2024 alone estimated to total between $83 billion and $127 billion. Without reform, Medicare is at risk of overpaying MA plans between $1.3 trillion and $2 trillion over the next decade.⁸ The one MedPac chart I have has the overpayment at $89 billion.

Sources of Medicare Advantage overpayment

The federal government uses several steps to determine how much to pay MA plans. First, the Centers for Medicare and Medicaid Services (CMS) sets a “benchmark” that is meant to approximate the cost of providing coverage to an average patient under traditional Medicare. Next, each MA plan submits a “bid” that represents their expected cost for providing insurance to an average patient. If the bid is lower than the benchmark, the MA plan is paid their bid plus a portion of the difference between the bid and the benchmark—known as the “rebate”—which the MA plan can use to fund extra benefits or lower out-of-pocket costs.

Importantly, because different patients have different expected costs, payments to MA plans are adjusted based on a “risk score” that considers all the diagnosis codes entered for a patient; typically, the more diagnosis codes entered, the higher the payment. MA plans use these lump sum payments to pay for the patient’s expenses, as well as to cover the plan’s administrative costs and profit.¹⁰ In recent years, administrative expenses and profits have accounted for about 15 percent of total MA spending, compared with less than 3 percent of spending in traditional Medicare.

In theory, by benchmarking MA to traditional Medicare, the only way for MA plans to obtain a rebate, and therefore offer more benefits or lower out-of-pocket costs, is by reducing spending compared with traditional Medicare—for instance, by using narrower networks or controlling utilization.¹² In reality, however, MA benchmarks and the resulting payments are inflated well above traditional Medicare costs because of inflated risk scores, selective MA enrollment of more profitable patients, quality bonuses and county adjustments, and the influence of supplemental coverage on traditional Medicare benchmark calculations.¹³

The scale of MA overpayments

As detailed below, a new CAP analysis estimates that MA plans are overpaid by 22 to 39 percent, corresponding to between $83 billion and $127 billion in overpayments in 2024.¹⁴ The lower estimate, 22 percent, indicates how much more MA plans are paid relative to what traditional Medicare would spend for equivalent patients. These types of overpayments are primarily driven by MA plans’ intense risk-coding efforts as well as selection bias. According to estimates from the Medicare Payment Advisory Commission (MedPAC), the independent congressional agency that advises the U.S. Congress on issues affecting the Medicare program, risk coding results in 13 percent overpayment to MA plans, and selection bias results in an additional 9 percent.¹⁵

This report’s higher estimate includes the additional subsidizing effects of quality bonuses and county adjustments MA plans receive, as well as the inflationary effect that supplemental coverage has on the traditional Medicare benchmarks against which MA plan bids are considered. Accounting for these additional subsidies to MA would suggest that MA plans may actually be “overpaid” by as much as 39 percent.¹⁶ Quality bonuses and county adjustments, combined, contribute a 5 percent overpayment,¹⁷ while the inflationary effect supplemental coverage has on traditional Medicare benchmark calculations contributes an additional 12 percent.¹⁸

When considered over the long term, the magnitude of MA overpayment is immense. The Medicare Board of Trustees projects that from 2023 to 2032, total payments to MA plans will surpass $7 trillion.¹⁹ Without reform, Medicare is at risk of overpaying MA plans between $1.3 trillion and $2 trillion over the same time period.²⁰

Drivers of MA overpayment

Upcoding or risk-score gaming

CMS uses a risk adjustment process to determine how much to pay an MA plan for any given patient, as patients have varying health needs. A key factor in determining risk scores is a patient’s diagnosis codes.²¹ As a result, MA plans have an incentive to maximize the number of diagnosis codes attributed to each patient—and to push the limits in terms of which codes can be legitimately assigned to a given patient.²² The more lucrative the code, the higher the payment. In the traditional Medicare program, there is no comparable incentive to misuse diagnosis codes or to maximize the entry of valid codes; the process of finding and entering codes can be tedious and burdensome and distract from patient care.

This contrast is clearest when a patient moves from traditional Medicare to MA. Upon this transition, an MA plan can submit many more diagnosis codes to demonstrate how “sick” a patient is.²³ Medicare then pays more for that patient because of these extra diagnosis codes, even though the patient is not any sicker, nor are their expected medical costs any higher than while they were covered by traditional Medicare. This “upcoding” in MA plans is a key source of overpayment: MedPAC estimates that MA risk scores will be inflated by approximately 20 percent in 2024.²⁴ To partially account for this, CMS applies a “coding intensity adjustment” of slightly less than 6 percent. However, this still leaves overpayment levels of approximately 13 percent.²⁵

Selection bias

MA plans tend to enroll beneficiaries who are healthier and more profitable than beneficiaries who choose traditional Medicare, even after accounting for risk scores.²⁶ Health plans have a significant and obvious financial incentive to enroll beneficiaries who are healthier than average at the same risk score. As a result, plans employ a variety of business strategies to attract healthier beneficiaries. One classic example is through offering gym benefits, which tend to be of value to people who are relatively healthier, thus attracting to MA those who have fewer health needs.²⁷

Another strategy is to limit network access to comprehensive cancer centers, which could make MA less attractive to patients with cancer and complex medical needs.²⁸ MA plans can also deploy their significant marketing efforts to selectively target more profitable patients.²⁹ From a patient perspective, traditional Medicare guarantees access to a broad network of providers, while MA plans come with more restricted networks.³⁰ People with fewer health needs are more likely to accept MA’s restrictions on care.

The result of these practices is that MA enrolls patients with lower-than-average spending at any given risk score, leading to further overpayment to MA plans, which then get paid as if their members are sicker than they are. Further evidence of this comes from a KFF analysis that found patients who switched from traditional Medicare to MA spent less in the prior year than those patients who remained in traditional Medicare—emphasizing that patients who move from traditional Medicare to MA tend to be healthier.³¹

It is worth noting that selection bias also happens in the opposite direction: Patients who switch from MA to traditional Medicare have higher risk-adjusted spending than those patients who stay in MA.³² For instance, a patient in MA who receives a serious diagnosis such as cancer might face considerable difficulty finding an in-network cancer center or dealing with prior authorization requirements for recommended scans. Therefore, they may choose to switch to traditional Medicare, which provides better access to care.

MedPAC estimates that in 2024, favorable selection will cause MA plans to be overpaid by approximately 9 percent.³³

Quality bonuses and county adjustments

The MA quality bonus program is intended to incentivize and reward high quality among MA plans. As detailed in a recent Urban Institute report, however, the net impact of the program is overly generous payments to MA plans absent any clear impact on quality.³⁴ Furthermore, as designed, the MA quality bonus program exacerbates health inequities. The program assigns each MA contract a rating of 1 to 5 stars based on performance across dozens of measures. Notably, MA plans with higher star ratings—and, as a result, higher bonus payments—appear to have larger racial disparities than lower-rated plans.³⁵ Plans in metropolitan areas with high MA enrollment and low traditional Medicare spending are also eligible to receive lucrative MA “double bonus” payments, but those payments have been found to worsen racial disparities without improving plan quality.³⁶ One study found the net effect of double bonuses was to increase payments for Black beneficiaries by $60 per year, compared with $91 for white beneficiaries.³⁷

In addition, the MA quality bonus program is not budget neutral, unlike other Medicare quality incentive programs. The program functions through an add-on payment that boosts spending to awarded MA plans above traditional Medicare levels. MA plans with higher star ratings receive a 5 percent bonus to their payment benchmark, and in those counties where a double bonus is available, plans receive a bonus of 10 percent to their benchmark.³⁸ In 2024, spending on the MA quality bonus program is estimated to reach $15 billion.³⁹

In addition to the MA quality bonus program, Congress has legislated that MA payment benchmarks be increased in counties with lower traditional Medicare spending. To calculate county benchmark adjustments, counties are split into quartiles; those counties with the lowest traditional Medicare spending receive 15 percent increases to their benchmarks, and counties with the highest levels of traditional Medicare spending have their benchmarks decreased by 5 percent.⁴⁰ The original motivation for this payment adjustment was to protect against MA plans not entering counties with historically low traditional Medicare spending—and therefore lower payment benchmarks against which MA plans would have to bid.⁴¹ However, as more than half of all Medicare beneficiaries are now enrolled in MA—and as that share continues to grow—this extra incentive is arguably no longer needed. The effect of the payment adjustments as they exist today is unnecessary inflation of MA payments.⁴²

MedPAC estimates indicate that in 2024, quality bonuses and county adjustments will inflate benchmarks by approximately 8 percent.⁴³

The two overpayment estimates presented in this report classify this increase in different ways. The lower estimate does not include quality bonuses and county adjustments as a direct source of overpayment, instead considering these payments to offset savings that would otherwise accrue due to MA bids being below the traditional Medicare benchmark.⁴⁴ The higher estimate, in contrast, classifies payments resulting from quality bonuses and county adjustments as a form of overpayments. Because plans receive a rebate that is a portion of the difference between their bid and the benchmark—typically 65 percent—this report’s higher estimate concludes that inflated benchmarks resulting from quality bonuses and county adjustments lead to overpayments of approximately 5 percent.⁴⁵

The effects of supplemental coverage on traditional Medicare benchmarks

When MA plans submit their bids to CMS to be compared against traditional Medicare benchmarks, they do so based on their expected cost of covering the traditional Medicare benefit package—services covered by Parts A and B. However, traditional Medicare benchmarks are not set based on Medicare’s spending on beneficiaries with Medicare Parts A and B alone; they also account for enrollees’ supplemental coverage, which facilitates increased use of Parts A and B covered services. That is because roughly 90 percent of traditional Medicare beneficiaries carry additional supplemental coverage, including Medicaid or private “Medigap” plans that reduce out-of-pocket cost exposure for Parts A and B covered services.⁴⁶ In effect, that means MA bids are benchmarked to total traditional Medicare spending, which includes standard Parts A and B coverage and any supplemental coverage a beneficiary has.⁴⁷

Supplemental coverage lowers beneficiaries’ out-of-pocket cost exposure, facilitating increased use of care that is baked into the benchmarks used to set MA payments. Literature historically has referred to this increased use of care as “induced utilization,” though it is perhaps better described as alleviating the depressed utilization that patients experience when out-of-pocket cost exposure is high.

If traditional Medicare alone is to be considered the baseline, the effects of supplemental coverage in benchmark calculations can be considered a source of overpayment to MA relative to traditional Medicare. A recent estimate suggests that the effects of supplemental coverage increase traditional Medicare benchmarks by approximately 18 percent, which implies approximately 12 percent in overpayments.