Insurance industry gearing up again for more Medicare Advantage overpayments

New Report: CMS' Changes to Medicare Advantage Undermine Care for Beneficiaries Managing Chronic Conditions

America’s Health Insurance Plans (AHIP), January 22, 2016

With 17 million seniors and individuals with disabilities depending on the Medicare Advantage program, a report from Avalere Health raises new concerns about CMS' policies that undermine health plans' efforts to care for beneficiaries managing multiple chronic conditions. After assessing the accuracy of CMS' current risk adjustment model and the cost of care for chronic health conditions, the Avalere analysis found that the model under-predicts costs for individuals with multiple chronic conditions by $2.6 billion on an annual basis. These findings come just weeks before CMS releases its annual proposed payment notice and call letter for Medicare Advantage and Part D plans, which may include further changes to the program and seniors' benefits.

In the spring of 2015, CMS finalized changes to the risk adjustment system, which directly targeted chronic disease prevention programs. This latest Avalere analysis demonstrates that these changes significantly limit health plans' early intervention efforts and seniors' benefits.

"Further cuts to Medicare Advantage and seniors' benefits are fundamentally at odds with the goal of delivering better care and better value for beneficiaries," AHIP President and CEO Marilyn Tavenner said. "Rather than relying on an antiquated fee-for-service approach as the model for care delivery, CMS should focus on strengthening Medicare Advantage and the innovative programs that improve seniors' health.”

Last year, more than 340 members of Congress, lead by Sen. Chuck Schumer (D-NY), Sen. Mike Crapo (R-ID), Rep. Patrick Murphy (FL-18), and Rep. Brett Guthrie (KY-02), urged CMS to protect seniors' coverage and provide stability to the program. Ahead of the upcoming February rate notice, more than 2 million seniors from AHIP's Coalition for Medicare Choices have mobilized, urging Washington to defend the Medicare Advantage program from further payment cuts.


Analysis of the Accuracy of the CMS-Hierarchical Condition Category Model

Avalere Health, January 2016

From the Executive Summary

Since 2000, the Centers for Medicare & Medicaid Services (CMS) has adjusted Medicare Advantage (MA) capitated payments for demographic characteristics and health status (also known as “risk adjustment”). In 2004 CMS adopted the Hierarchical Condition Category (HCC) risk adjustment model, which includes a series of patient diagnoses that impact healthcare spending. In 2014, CMS introduced a new version of the model that removed certain conditions, added others, and made additional modifications (hereafter referred to as the “2014 model”).

In this project, Avalere assessed the accuracy of the 2014 model for beneficiaries in the traditional Medicare program with certain common chronic conditions by using Medicare fee-for-service (FFS) claims data to compare predicted healthcare costs with actual healthcare costs.

In summary, we estimate that the 2014 model under-predicts costs for individuals with multiple chronic conditions by $2.6 billion on an annual basis (see Table 1). As a result, because the model is “zero sum”—that is, the values for each condition are relative to the average cost across all individuals—under-prediction for individuals with multiple chronic conditions is balanced by over-prediction of costs for individuals with no chronic conditions.

On October 28, 2015, CMS announced proposed changes to the MA risk adjustment model that the agency believes will improve its predictive power for low-income beneficiaries. Specifically, CMS intends to further refine the model by accounting for both dual eligible/low-income subsidy (LIS) eligible and disabled status.

(W)e intentionally used a different disease classification system from the HCC model groupings in order to independently assess how well the model predicts costs for specific chronic diseases.

Table 1. Predictive Accuracy for Beneficiaries with Multiple Chronic Conditions

  Total Estimated MA Over/Under- Prediction of Expenditures ($ millions)

    Multiple Chronic Conditions (All)  $ (2,613.7)
    Multiple Chronic Conditions (Dual/LIS-eligibles)  $ (401.8)

The purpose of risk adjustment is to anticipate systematic differences in costs for groups of individuals so that plans are reasonably compensated for the financial risks they bear. If plans are not accurately compensated for taking on the risk associated with a particular group, it creates a misalignment between payments and costs for higher cost beneficiaries, and under-compensates plans that enroll many chronically ill members. For any particular individual, the model may over- or under-predict actual costs, in some cases by a wide margin; every dollar of under-predicted cost is balanced by a dollar of over-predicted cost.

From the Detailed Findings

Analysis for Multiple Chronic Conditions

We reviewed how well the model predicts expenditures for individuals with multiple chronic conditions in order to determine how well payments to MA plans would be risk adjusted for the clinical severity of their patient populations under the 2014 model. As shown in Table 3, we find that the model under-predicts costs by approximately $2.6 billion for individuals with three or more chronic conditions. We also find that the model over-predicts disease burden for individuals without chronic conditions.

Table 3. Predictive Accuracy for Members with Chronic Conditions; All Members and Dual/LIS-Eligibles

  Total Estimated MA Over/Under- Predictions ($ millions)

    All Members

      Multiple (3+) Chronic Conditions  $ (2,613.7)

      Few (1-2) Chronic Conditions  $ 936.2

      No Chronic Conditions  $ 1,677.50


      Multiple (3+) Chronic Conditions  $ (401.8)

      Few (1-2) Chronic Conditions  $ 599.8

      No Chronic Conditions  $ 582.9

From the Conclusion

We reviewed the accuracy of the new CMS-HCC model at predicting costs for individuals with multiple chronic conditions, and paid particular attention to how well the model predicts costs for high cost individuals. We find that the CMS-HCC model substantially under-predicts costs for individuals with multiple chronic conditions, under-predicts costs for several specific chronic conditions, and does not accurately predict costs for high cost individuals within each chronic condition.

These findings suggest the model may need improvements and modifications in order to appropriately pay for high cost members and individuals with multiple and certain single chronic conditions. In other words, the model may not be adequately compensating health plans for treating these individuals.



By Don McCanne, M.D.

For the past four years, the private insurance industry, led by their lobby organization - AHIP, has been successful in offsetting the reductions in overpayments that have been made to the private Medicare Advantage plans - reductions that are required by the Affordable Care Act. AHIP has now commissioned Avalere to produce a study that purportedly shows that they will need higher capitation payments than the CMS’s risk adjustment program would allow. The release of this study is the first step in their campaign to, once again, offset the decreases required by ACA.

A program to authorize private Medicare Advantage plans (originally Medicare + Choice plans) was authorized by Congress as a move to eventually completely privatize Medicare once the private plans were able to show that they could deliver higher quality at lower costs. Early on the concept was proven a fraud when the plans successfully marketed their plans selectively to healthy Medicare beneficiaries, while being compensated at levels equivalent to the costs of those in the traditional fee-for-service (FFS) Medicare program who had greater health problems.

In response, CMS developed a risk adjustment program that would pay more when the Medicare Advantage plans enrolled beneficiaries with greater health care needs, based on their diagnoses. The private plans then responded by upcoding the diagnoses of their enrollees, making them appear much sicker than they were. They even went to the point of sending out teams to make detective house calls so that they could add more diagnoses that were not being itemized by the providers.

In 2004, CMS adopted the Hierarchical Condition Category (CMS-HCC) risk adjustment model, which does adjust payments upward for those with greater needs, but it still fails to prevent about four-fifths of the excessive payments.

With pressure from AHIP, and with the support of Congress, CMS used various innovative methods to boost the payment rates for these private plans. This year, they seem to be headed towards a claim that they are being paid much less for high cost patients than the actual costs entailed.

Look at Table 1 in the Executive Summary (the only part that legislators read). Based on current CMS risk adjustment methods, they predict that the calculated costs for beneficiaries with multiple chronic conditions will fall short of actual costs by $2,614 million. They are now campaigning to have those costs added to their reimbursement rates for 2017.

But look at Table 3 which is found in “Detailed Findings” (which most will not read). It is the same as Table 1, but expanded to include the predicted calculations for those who have few or no chronic conditions. It shows that the CMS calculations predict $2,614 million in excess estimates of costs.

For the dual eligible/low-income subsidy group (Dual/LIS), Table 1 shows that the costs for those with multiple chronic conditions would be underestimated by $402 million. But for the Dual/LIS with few or no conditions, Table 3 shows that the predicted costs would be calculated to be $1,183 million over the actual costs. The report indicates that Dual/LIS patients have greater costs, so they plead to be compensated for this $402 million underestimate. They remain silent on the $1,183 overestimate for the healthier sector.

Also the estimates are based on patients in the traditional FFS Medicare program, a less healthy population than those in the Medicare Advantage plans. Since the Medicare Advantage plans continue to be successful in recruiting healthier patients, the overestimates by which they would be reimbursed would be even greater.

The politics are ugly. Marilyn Tavenner, as head of CMS, participated in the conspiracy to use devious innovations to overpay the Medicare Advantage plans. She is now president and CEO of AHIP and will use her cozy relationship with members of Congress to be sure that they put more pressure on CMS to once again jack up the rates for Medicare Advantage plans, in conflict with the intent of the ACA legislation.

This is a nefarious effort that is part of the conspiracy to completely privatize Medicare. As Marilyn Tavenner said in the AHIP news release, “Rather than relying on an antiquated fee-for-service approach as the model for care delivery, CMS should focus on strengthening Medicare Advantage and the innovative programs that improve seniors' health.” This statement is not ambiguous. It is a blatant call for total privatization of Medicare.

The last thing we want is a privatized Medicare Advantage for all who can afford it.