Executive Summary
Medicare Advantage (MA) now provides healthcare coverage for 35 million enrollees, accounting for more than half of all Medicare beneficiaries.1 While the program provides critical care for American seniors, private insurers have exploited its payment system for decades by using upcoding to increase profits.
Upcoding, where plan sponsors artificially inflate patients' risk scores by making them appear sicker, led to an estimated $170 billion in excess payments from the Centers for Medicare and Medicaid Services (CMS) from 2015 to 2024, with another $50 billion projected across 2025 and 2026.2 These losses strain the federal budget and raise Part B premiums for every Medicare beneficiary.
The incentive misalignment is clear: because MA plans are paid based on how sick their enrollees appear on paper, they have a financial incentive to make enrollees look sicker than they are. Although CMS and Congress have taken steps to address upcoding, further reforms are needed to address the underlying financial incentives.
Background
In March 2026, Aetna, one of the largest health insurers in the U.S. and a CVS Health subsidiary, agreed to pay $117.7 million to resolve allegations that it overcharged Medicare.3 The U.S. government accused Aetna of submitting inaccurate diagnosis codes for its Medicare Advantage (MA) enrollees, making patients appear sicker on paper to receive higher payments.
Insurers have used upcoding to increase MA profits for years. Recently, major insurers like Kaiser Permanente and Elevance Health have paid hundreds of millions of dollars to resolve similar allegations.4,5 All five of the largest MA insurers, which collectively cover nearly 70% of MA enrollees, have faced whistleblower lawsuits, government fraud allegations, or Office of Inspector General (OIG) findings of overbilling – Figure 1.1,3-13
MA plans are privately administered alternatives to traditional Medicare, where CMS pays private insurers to provide Medicare benefits.14 Compared with traditional Medicare, MA plans can be attractive because they typically offer lower cost-sharing, additional benefits such as dental and vision coverage, and low or $0 premiums.15,16
Over the years, MA has become increasingly important in providing health coverage to seniors across the country. As of 2026, 35 million people are enrolled in MA, representing more than half of all Medicare beneficiaries – Figure 2.1,17
The Risk Score Mechanism
Medicare Advantage relies on a risk score model to estimate the expected healthcare cost for each enrollee.18 The model calculates an individual’s risk score using a combination of demographic and health factors, with each factor assigned a numeric value.19 In general, older and sicker patients receive higher risk scores.a
CMS then uses these scores to determine how much it pays plan sponsors each month for each enrollee.18 Higher risk scores mean higher payments, as CMS expects plans to spend more on sicker patients. These scores are updated annually to reflect changes in enrollees’ diagnoses.
a For more details on how risk scores are calculated, refer to Appendix A.
Evidence of Upcoding
Since payments received by plan sponsors from CMS are determined by reported diagnosis codes rather than patients’ actual health conditions, plans may have an incentive to inflate risk scores by reporting additional or more severe diagnoses than patients actually have, including those not supported by medical records or related follow-up treatment. This practice is known as upcoding.20
Concerns about MA upcoding started years ago. In 2012, the U.S. Government Accountability Office (GAO) published a report estimating that MA upcoding inflated enrollee risk scores by 5% to 7% compared with traditional Medicare, translating to $3.9 billion to $5.8 billion in excess payments to plan sponsors in 2010 alone.21
As MA grew over the years, so did the impact of upcoding. From 2015 to 2024, upcoding led to roughly $170 billion in excess MA spending, with another $50 billion projected across 2025 and 2026 – Figure 3.2 For context, $50 billion is roughly half of annual federal spending on the Supplemental Nutrition Assistance Program (SNAP) and equivalent to the entire annual National Institutes of Health (NIH) budget.22,23
In addition to straining the federal budget, these overpayments also directly increase Part B premiums for all Medicare beneficiaries. The Joint Economic Committee estimated that MA overpayments raised Part B premiums by approximately $212 per enrollee in 2025, totaling $13.4 billion in additional premiums.24 Although upcoding is not the only source of these overpayments, it is a major contributor.
More recent evidence has continued to identify unsupported diagnoses submitted for risk-adjustment payments. In a 2024 audit, OIG found cases in which enrollees received diagnoses for conditions such as lung, breast, and colon cancer without corresponding treatment documented within six months before or after the diagnosis. Similar patterns were also identified for other high-risk conditions, including stroke and sepsis.25
OIG identified health risk assessments (HRAs) and HRA-linked chart reviews as major contributors to upcoding. During an HRA, a healthcare professional collects information about an enrollee’s health status and demographic characteristics, which can help identify diagnoses that MA plan sponsors report to CMS.26
Although HRAs typically occur in physicians’ offices or other healthcare facilities, they may also be conducted through telehealth or during in-home visits. Many in-home HRAs are administered by third-party vendors rather than patients’ regular providers. These visits also rely heavily on self-reported medical histories and often lack the diagnostic equipment needed to verify reported conditions, raising concerns about their validity.26 Nevertheless, diagnoses identified through in-home HRAs may still be submitted for risk-adjustment purposes, even when no follow-up care or additional evaluation is documented.27
HRA-linked chart reviews, in which MA companies retrospectively review an enrollee’s medical records to identify missing or additional diagnoses, may be even more vulnerable to misuse. Because these reviews rely heavily on past medical records, they may capture outdated or unsupported diagnoses and include them in risk-adjustment payment calculations. Together, diagnoses reported only through HRAs and HRA-linked chart reviews generated an estimated $7.5 billion in MA risk-adjusted payments for 2023.26
Reform Proposals
In 2010, CMS began addressing MA upcoding through a 3.4% coding adjustment.28 This adjustment reduces risk scores reported by MA plans to account for upcoding. As of 2026, the adjustment is 5.9%.29 This means that if an MA plan reports a risk score of 100 for an enrollee, CMS would reduce it to roughly 94 before calculating payments.
Many policy stakeholders have argued that 5.9% remains too low to fully offset the financial impact of upcoding practices. In 2026, for example, the Medicare Payment Advisory Commission (MedPAC) estimated that MA risk scores were 10% higher than they would have been under traditional fee-for-service (FFS) Medicare. As a result, the 5.9% adjustment only offsets part of the upcoding impact, and CMS still pays MA plans roughly 4% more than it would for the same enrollees under FFS Medicare – Figure 4.30 This pattern is consistent across the years, with MA payments remaining 2% to 10% above FFS Medicare levels even after the coding adjustment.
While the coding adjustment is a step in the right direction, further reforms are needed to address the economic incentives that encourage MA plan sponsors to inflate enrollee risk scores in the first place. Depending on resource constraints and political feasibility, policymakers could consider the following approaches.
Target Unsupported HRA and Chart Review Diagnoses
The proposal to target HRA and chart review diagnoses builds on core elements of the No Unreasonable Payments, Coding, or Diagnoses for the Elderly Act, known as the “No UPCODE Act.”31 Introduced in March 2025, the Act would exclude diagnoses collected through HRAs and chart reviews from MA risk adjustment.32
Instead of excluding these diagnosis entirely as proposed in the No UPCODE Act, policymakers could instead adopt a more targeted auditing approach, excluding only HRA and chart review diagnoses that lack sufficient support from hospitalization records or follow-up treatment. MA plan sponsors that repeatedly fail to provide sufficient documentation for their diagnoses would face increased compliance oversight and additional financial penalties. This approach could preserve the convenience and clinical value of HRAs and chart reviews while deterring their abuse. However, the effectiveness of this proposal depends on whether CMS has sufficient resources and audit capacity.
Encouragingly, CMS is already looking to expand its capacity to review outstanding payments from 2018 through 2024.33 Many components of this initiative, such as the use of enhanced technology to review medical records and the expansion of CMS’s medical coding workforce, would help support this proposal.
Adopt Plan-Specific Coding Adjustments
To address the underlying financial incentive for upcoding, policymakers could consider a tiered coding adjustment system. Instead of reducing every MA plan’s reported risk scores by the same 5.9%, CMS could assess each plan sponsor’s coding practices on an annual basis, applying larger adjustments when a plan shows unusually high or unexplained risk-score patterns.
Compared with the current uniform adjustment, this approach would more directly link the size of each adjustment to a plan’s coding behavior. However, it would also require a more comprehensive analysis; CMS would need to determine whether unusual risk-score patterns reflect aggressive coding practices or actual changes in enrollee health and demographics. As such, enhanced technology and stronger data infrastructure would be especially valuable in this case.
Restricting Vertical Integration
Another option to limit the impact of upcoding is for policymakers to consider restricting vertical integration between MA plan sponsors and care assessment entities. When plan sponsors can contract with or acquire companies that conduct HRAs, they gain greater control over how those assessments can be used to maximize diagnosis capture for risk-adjustment payments.
In March 2023, CVS Health completed its acquisition of Signify Health for approximately $8 billion.34 Signify Health specializes in in-home HRAs, completing more than 3.5 million in-home health evaluations in 2025 alone.35
The acquisition occurred amid ongoing Department of Justice (DOJ) allegations of upcoding, OIG findings related to overpayments, and research highlighting the risks associated with in-home HRAs. Nevertheless, many MA plan sponsors continue to work closely with affiliated care assessment entities, including CVS Health and Signify Health.
To address this conflict of interest, policymakers could restrict plan sponsors from using affiliated care assessment entities to conduct in-home HRAs and retrospective chart reviews for their own MA plans. Instead, CMS could contract directly with independent third-party assessment entities.
Conclusion
Medicare Advantage remains an important source of coverage for millions of beneficiaries in the U.S., but its current risk-adjustment structure creates financial incentives for plans to inflate risk scores. The evidence reviewed in this paper suggests that MA upcoding is a structural issue that contributes to substantial excess federal spending. Addressing these incentives will require a combination of more targeted auditing, alternative models for payment adjustments, and stronger guardrails around vertical integration. Together, these reforms could help preserve MA coverage while ensuring that federal payments more directly contribute to patient care.
Methodology
The purpose of this paper is to provide a policy analysis of the program integrity risks associated with Medicare Advantage (MA) risk adjustment and upcoding practices. All data and analysis presented in this report are drawn from publicly available sources and are cited accordingly.
Figure 1 is adapted from a 2022 New York Times analysis of fraud allegations and overbilling findings involving the largest Medicare Advantage insurers.6 The Institute team reviewed and updated the underlying cases using more recent DOJ and OIG sources, while adding a settlement column to further track how these cases progressed.
This paper does not assume that allegations equal proven Medicare fraud, nor does this paper attempt to hold any individual company legally liable. Instead, the report aims to highlight the importance of MA program integrity while summarizing where abuse or misaligned incentives may exist. Policymakers and other interested stakeholders should view this report as a contextual policy analysis intended to support informed discussions on MA upcoding risks and potential reform efforts.
Appendix A. How Medicare Advantage Risk Scores Are Calculated
MA plans rely on the Hierarchical Condition Categories (HCC) model to calculate the risk score for each enrollee.18 The HCC model assigns each disease category a code and a corresponding value. Diseases associated with higher expected healthcare costs receive higher values. For example, metastatic cancer and acute leukemia are assigned a value of ~2.6, while diabetes without complications is assigned a value of ~0.1 – Figure 5.19 These disease coefficients are additive, meaning that if the same person has multiple unrelated diagnosis, the corresponding values are added together.
Demographic characteristics, such as age and sex, provide a baseline value for each enrollee. The rationale for assigning these values is similar: people expected to require more healthcare services receive higher values. For example, a 73-year-old male would have a higher baseline value than a 67-year-old male.
Finally, each enrollee is placed into a group based on whether they live at home or in a long-term care institution, whether they qualify for Medicare because of age or disability, and whether they are also eligible for Medicaid. Based on these characteristics, the same disease category may be assigned a different value.
References
1. Freed, M., Fuglesten Biniek, J., Damico, A., Ochieng, N., & Neuman, T. (2026, June 5). Medicare Advantage in 2026: Enrollment update and key trends. KFF.
https://www.kff.org/medicare/medicare-advantage-in-2026-enrollment-update-and-key-trends/
2. Medicare Payment Advisory Commission. (2026, March). The Medicare Advantage program: Status report. In Report to the Congress: Medicare payment policy (Chapter 12).
https://www.medpac.gov/wp-content/uploads/2026/03/Mar26_Ch12_MedPAC_Report_To_Congress_SEC.pdf
3. U.S. Department of Justice. (2026, March 11). Aetna agrees to pay $117.7 million to resolve False Claims Act allegations. https://www.justice.gov/opa/pr/aetna-agrees-pay-1177-million-resolve-false-claims-act-allegations
4. U.S. Department of Justice. (2026, January 14). Kaiser Permanente affiliates pay $556M to resolve False Claims Act allegations. https://www.justice.gov/opa/pr/kaiser-permanente-affiliates-pay-556m-resolve-false-claims-act-allegations
5. Schulte, F. (2026, June 26). Medicare Advantage company pays $342M to government in midst of billing probe. KFF Health News. https://kffhealthnews.org/medicare/medicare-advantage-cms-elevance-crackdown-overcharging-payment/
6. Abelson, R., & Sanger-Katz, M. (2022, October 8). ‘The cash monster was insatiable’: How insurers exploited Medicare for billions. The New York Times.
https://www.nytimes.com/2022/10/08/upshot/medicare-advantage-fraud-allegations.html
7. U.S. Department of Justice. (2017, May 16). United States intervenes in second False Claims Act lawsuit alleging that UnitedHealth Group Inc. mischarged the Medicare Advantage and prescription drug programs. https://www.justice.gov/archives/opa/pr/united-states-intervenes-second-false-claims-act-lawsuit-alleging-unitedhealth-group-inc
8. Freidin Brown, P.A. (2018, November 19). Freidin Brown, P.A. attorneys secure $7 million result in whistleblower case. PR Newswire. https://www.prnewswire.com/news-releases/freidin-brown-pa-attorneys-secure-7-million-result-in-whistleblower-case-300753286.html
9. U.S. Department of Health and Human Services, Office of Inspector General. (2025, December). Medicare Advantage compliance audit of specific diagnosis codes Humana Health Benefit of Louisiana (Contract H1951) submitted to CMS (A-06-21-02001). https://oig.hhs.gov/documents/audit/11315/A-06-21-02001.pdf
10. U.S. Department of Health and Human Services, Office of Inspector General. (2025, June). Medicare Advantage compliance audit of specific diagnosis codes that Coventry Health and Life Insurance Company (Contract H1608) submitted to CMS (A-02-22-01020). https://oig.hhs.gov/documents/audit/10329/A-02-22-01020.pdf
11. U.S. Department of Justice. (2021, July 30). Government intervenes in False Claims Act lawsuits against Kaiser Permanente affiliates for submitting inaccurate diagnosis codes to the Medicare Advantage program. https://www.justice.gov/archives/opa/pr/government-intervenes-false-claims-act-lawsuits-against-kaiser-permanente-affiliates
12. U.S. Attorney’s Office, Southern District of New York. (2020, March 27). Manhattan U.S. Attorney files civil fraud suit against Anthem, Inc., for falsely certifying the accuracy of its diagnosis data. U.S. Department of Justice. https://www.justice.gov/usao-sdny/pr/manhattan-us-attorney-files-civil-fraud-suit-against-anthem-inc-falsely-certifying
13. U.S. Department of Health and Human Services, Office of Inspector General. (2021, May). Medicare Advantage compliance audit of specific diagnosis codes that Anthem Community Insurance Company, Inc. (Contract H3655) submitted to CMS (A-07-19-01187). https://oig.hhs.gov/documents/audit/8985/A-07-19-01187-Complete%20Report.pdf
14. Centers for Medicare & Medicaid Services. (2026). Understanding Medicare Advantage Plans. Medicare.gov. https://www.medicare.gov/publications/12026-understanding-medicare-advantage-plans.pdf
15. Peter G. Peterson Foundation. (2024, August 20). Budget basics: What is Medicare Advantage?https://www.pgpf.org/article/what-is-medicare-advantage/
16. Cubanski, J., & Damico, A. (2024, July 2). Key facts about Medicare Part D enrollment, premiums, and cost sharing in 2024. KFF. https://www.kff.org/medicare/key-facts-about-medicare-part-d-enrollment-premiums-and-cost-sharing-in-2024/
17. Centers for Medicare & Medicaid Services. (2025). Table 22: Health insurance enrollment and uninsured. In National Health Expenditure Data: Historical. https://www.cms.gov/files/zip/nhe-tables.zip
18. Centers for Medicare & Medicaid Services. (2024, December). Report to Congress: Risk adjustment in Medicare Advantage. https://www.cms.gov/files/document/report-congress-risk-adjustment-medicare-advantage-december-2024.pdf
19. Centers for Medicare & Medicaid Services. (2024). Revised CMS-HCC model relative factor tables. https://www.cms.gov/files/document/revised-cms-hcc-model-relative-factor-tablespdf
20. Office of U.S. Senator Bill Cassidy. (2025, March 14). Cassidy, President Trump’s CMS nominee agree on need to stop Medicare overpayments. https://www.cassidy.senate.gov/newsroom/press-releases/cassidy-president-trumps-cms-nominee-agree-on-need-to-stop-medicare-overpayments/
21. U.S. Government Accountability Office. (2012, January). Medicare Advantage: CMS should improve the accuracy of risk score adjustments for diagnostic coding practices (GAO-12-51). https://www.gao.gov/assets/gao-12-51.pdf
22. U.S. Department of Agriculture, Economic Research Service. (2025, July 24). Supplemental Nutrition Assistance Program (SNAP): Key statistics and research.
https://www.ers.usda.gov/topics/food-nutrition-assistance/supplemental-nutrition-assistance-program-snap/key-statistics-and-research
23. National Institutes of Health. (2025, June 13). Budget. https://www.nih.gov/about-nih/organization/budget
24. S K, S. (2026, March 10). Congressional committee says seniors paid higher premiums due to alleged Medicare overpayments. Reuters.
https://www.reuters.com/business/healthcare-pharmaceuticals/committee-says-seniors-paid-higher-premiums-due-alleged-medicare-overpayments-2026-03-10/
25. U.S. Department of Health and Human Services, Office of Inspector General. (2024). Medicare Advantage compliance audit of specific diagnosis codes that HealthAssurance Pennsylvania, Inc.(Contract H5522) submitted to CMS (A-05-22-00020). https://oig.hhs.gov/documents/audit/10006/A-05-22-00020.pdf
26. U.S. Department of Health and Human Services, Office of Inspector General. (2024, October). Medicare Advantage: Questionable use of health risk assessments continues to drive up payments to plans by billions (OEI-03-23-00380). https://oig.hhs.gov/documents/evaluation/10028/OEI-03-23- 00380.pdf
27. Medicare Payment Advisory Commission. (2016, March). Report to the Congress: Medicare payment policy. https://www.medpac.gov/wp-content/uploads/2021/10/march-2016-report-to-the-congress-medicare-payment-policy.pdf
28. Kronick, R., & Welch, W. P. (2014). Measuring coding intensity in the Medicare Advantage program. Medicare & Medicaid Research Review, 4(2). Centers for Medicare & Medicaid Services. https://www.cms.gov/mmrr/Downloads/MMRR2014_004_02_a06.pdf
29. Centers for Medicare & Medicaid Services. (2025, April). Announcement of calendar year (CY) 2026 Medicare Advantage (MA) capitation rates and Part C and Part D payment policies. https://www.cms.gov/files/document/2026-announcement.pdf
30. Medicare Payment Advisory Commission. (2026, January 16). The Medicare Advantage program: Status report. https://www.medpac.gov/wp-content/uploads/2026/01/Tab-N-MA_Status-Jan-2026.pdf?
31. Office of U.S. Senator Bill Cassidy. (2025, March 25). Cassidy, Merkley introduce bill to stop overpayments in the Medicare Advantage program.
https://www.cassidy.senate.gov/newsroom/press-releases/cassidy-merkley-introduce-bill-to-stop-overpayments-in-the-medicare-advantage-program/
32. No Unreasonable Payments, Coding, or Diagnoses for the Elderly Act (No UPCODE Act), S. 1105, 119th Cong. (2025). https://www.cassidy.senate.gov/wp-content/uploads/2025/03/No-UPCODE-Act.pdf
33. Centers for Medicare & Medicaid Services. (2025, May 21). CMS rolls out aggressive strategy to enhance and accelerate Medicare Advantage audits. https://www.cms.gov/newsroom/press-releases/cms-rolls-out-aggressive-strategy-enhance-accelerate-medicare-advantage-audits
34. CVS Health. (2023, March 29). CVS Health completes acquisition of Signify Health. https://www.cvshealth.com/news/company-news/cvs-health-completes-acquisition-of-signify-health.html
35. CVS Health. (2026, March 26). Improving health one in-home health evaluation at a time. https://www.cvshealth.com/news/condition-management/improving-health-one-in-home-health-evaluation-at-a-time.html
