PNHP

Medicare Advantage: What Physicians Need to Know

By Ed Weisbart, M.D.; Kip Sullivan, J.D. and Mark Krasnoff, M.D.
Missouri Medicine, May/June 2026

When she was 82 years old, Pat started to experience episodes of chest pain and shortness of breath. Her internist in Florida referred her to a local cardiologist, who diagnosed her with hypertrophic cardiomyopathy (HCM). Her brother had died at age 85 from the same diagnosis.

At the time, there were no clinical centers in Florida specializing in HCM. Her family in St. Louis found that Washington University had a team with this focus and flew her up to Missouri. She underwent alcohol ablation of the hypertrophic area of her septum, spent a few days recovering as an inpatient without complications, and soon returned to Florida. Over the next decade she had good quality of life and ultimately passed away at age 94.

Despite traveling out of state, she never had to go through a prior authorization process, never had a copay or deductible, and never received a surprise medical bill. Pat had Traditional Medicare and paid a monthly premium for a Plan F Medigap supplemental policy. This story may have been quite different had she instead signed up for a Medicare Advantage plan. Far too many people don’t really understand the implications of this choice, or even that they’re making a choice.

Origin of Medicare Advantage

There was a time when doctors and patients didn’t need to understand Medicare Advantage (MA) because it didn’t exist. When Medicare was implemented in 1966, it was a relatively simple program. Original Medicare, now often called Traditional Medicare (TM), consisted of Part A, which paid for hospital services, and Part B, which paid for physician services. There was no Part C, the program that today finances the MA program, and no Part D, which pays insurance companies to cover prescription drugs.

Americans age 65 or older are automatically enrolled in Part A if they or their spouse has worked and paid Medicare taxes for at least ten years. Enrollment in Part B is not automatic; beneficiaries can enroll in Part B if they pay a premium.

The relative simplicity of Medicare was due to three features: (1) direct payment of providers (no insurance companies or other risk-bearing corporations sat between Medicare and providers); (2) beneficiaries’ freedom to choose their doctor and hospital; and (3) monitoring for fraud and abuse, but little if any interference by the administering agency with the practice of medicine. Those three features of original Medicare have been changed beyond recognition over the last half-century.

The transformation of these three features was driven by increased health care spending in the US in both the private and public sectors.

Congress took the first step to address rising Medicare expenditures in 1972.1 The Social Security Amendments of 1972 (P.L. 92–603) authorized the Social Security Administration to contract with a new form of insurance called the “Health Maintenance Organization” (HMO). The HMO label was invented in February 1970 by a Minnesota physician named Paul Ellwood at a meeting with three members of the Nixon administration. Ellwood argued that health care inflation was due to doctors ordering unnecessary services, and this was happening routinely because of the way doctors were paid: the fee-for-service (FFS) method. Given that diagnosis, it is no surprise that Ellwood thought the solution was to turn the FFS incentive upside down by paying doctors through “capitation,” shifting insurance risk onto doctors. Capitated doctors ran the risk of losing money if they continued to order the allegedly unnecessary services.

For capitation to have the most impact on medical decisions, patients had to be limited in their choice of doctors. Limiting patient access to one capitated doctor, or a group of them, would be possible only if the limitation were written into a patient’s insurance contract. HMOs were insurance companies that were paid a premium like any other insurance company (for some reason these premiums also came to be called “capitation” payments) that limited a Medicare beneficiary’s choice of doctor and paid doctors by capitation instead of fee-for-service.

Ellwood and the early HMO proponents never addressed an obvious question: Is it possible for HMOs to reduce unnecessary services by an amount large enough to offset HMO overhead (administrative costs plus profit) without harming patients? Three decades later, Bruce Vladeck, HCFA’s administrator from 1993 to 1997, and Barbara Cooper posed the challenge facing Medicare HMOs as follows: “[C]apitated plans – with administrative expenditures in the range of 8-25 percent – have to incur medical expenditures 10-25 percent less … just to break even. In most markets, capitated plans cannot attract enrollees unless they offer additional benefits, which also costs money… .”2

The Medicare Advantage Trap

Pat’s health outcomes may have been quite different if she had chosen to avoid paying the premiums for Medigap and instead enrolled in Medicare Advantage, where narrow networks, prior authorizations, and out-of-pocket expenses frequently delay and often deny access to the healthcare chosen by patients with their physicians.

If her care had been denied or delayed, and she had then decided to disenroll from her MA plan, she would have faced a trap. She would have needed to buy a Medigap plan to protect herself from the large out-of-pocket costs to which TM beneficiaries are otherwise exposed. People in Traditional Medicare without a Medigap plan are exposed to unlimited personal financial risk. In 2026, Part A requires a $1,716 deductible for each hospitalization period and Part B requires a 20% coinsurance for all outpatient services,3 with no upper limit. Although the most common Medigap plans nearly eliminate such personal expenses, for many people Medigap isn’t a realistic option. This has been called “The Gap Trap.”4

Many people incorrectly assume that the ACA’s protections against insurers pricing based on preexisting conditions applies to all health insurance. Similar protections apply to Medigap during the first six months after enrollment in Medicare Part B or the first twelve months in MA. However, after six to twelve months, with some state variations, Medigap insurers are allowed to individually underwrite applicants and decline to sell someone a policy which effectively destroys the guarantee of being able to purchase Medigap at the same price as the rest of the community.

Had Pat been in an MA plan instead of TM with Medigap, by the time she would have wanted to leave MA, return to TM, and purchase Medigap she would have been saddled with the kind of diagnosis (hypertrophic cardiomyopathy) that would have ensured a Medigap policy to be priced beyond reach – if any insurer were willing to sell her a policy at any price.

Medicare Advantage’s protections against initial outof-pocket costs for relatively health enrollees, along with the highly promoted promises of additional benefits, accounts for its rapidly growing capture of the Medicare population. However, these features of MA come at a significant cost to the Medicare program, patients, and ultimately taxpayers.

Causes of Medicare Advantage Overpayments

By the 1990s, it was clear HMOs typically were not capable of reducing medical expenditures sufficient to offset their overhead costs without harming patients. Private health insurance companies can make money off of Medicare only if they are paid far more to insure a beneficiary than the beneficiary would have cost had they remained in TM. Publicly funded MA insurers depend on the delay and often outright denial of medically necessary services. In the 1980s and 1990s, multiple studies appeared demonstrating that Medicare HMOs were enrolling beneficiaries who were on average much healthier and therefore less costly to insure than beneficiaries who remained in TM (a phenomenon known as “favorable selection”), but Medicare was paying the HMOs as if they enrolled beneficiaries in average health.5 Keeping the introductory clause implies sicker MA patients no longer disenroll and switch to TM (they do), and MA plans no longer benefit when sick enrollees return to TM (they do).The Health Care Financing Administration (HCFA), the agency created in 1977 to administer Medicare and Medicaid, did not ignore the overpayments caused by favorable selection. The staff and HCFA’s consultants attempted to measure the “risk” that HMO enrollees would need health care, compare it to the risk of beneficiaries who stayed in Traditional Medicare, and modify HMO payments accordingly, a process known as “risk adjustment.” But the measures HCFA used for risk measurement —age, sex, nursing home status, and enrollment in Medicaid—were so crude that risk adjustment was essentially worthless. In a 1994 article for Health Affairs, economist Joseph Newhouse characterized “risk adjustment technology” as “primitive.”6 Analysts generally agreed the problem with HCFA’s risk adjuster was that it did not include direct measures of health, namely patient diagnoses.

In the Balanced Budget Act of 1997, Congress allowed insurance companies of all stripes (not just HMOs) to participate in Medicare and instructed HCFA to add diagnoses to their risk adjuster. HCFA, renamed the Centers for Medicare and Medicaid Services (CMS) in 2001, phased in a diagnosis-based risk adjuster between 2000 and 2006. But the new risk adjuster, known as the Hierarchical Condition Categories (HCC) model, added very little predictive value to the old demographic-based model.

To make matters worse, the introduction of the HCC created the opportunity for MA plans to increase their overpayments by “upcoding.” Upcoding refers to adding diagnoses to patient medical records that are accurate but irrelevant to the true cost of care, or just plain false.

Today’s MA Plans Make Money Off Both Favorable Selection and Upcoding

The Medicare Payment Advisory Commission (MedPAC), an advisory body created by Congress, estimates in its annual reports to Congress the total cost of the overpayments to Medicare Advantage plans. Their latest estimate is that the plans were paid 20 percent more for each enrollee than the enrollee would have cost had they remained in TM, roughly half due to favorable selection uncorrected by CMS’s risk adjuster and half caused by upcoding, and that these overpayments added $84 billion to Medicare expenditures.7 Unfortunately, overpayments to Medicare Advantage plans are boosted as well by other features of CMS’s method of calculating capitation payments to the plans in addition to favorable selection and upcoding. In a paper published in October 2022 by Physicians for a National Health Program (the three of us were among the authors), we estimated that when all causes of overpayments are taken into account the overpayments may be as high as 35 percent, or about $150 billion in 2025.8 Note that these estimates do not take into account the additional administrative costs Medicare Advantage plans impose on clinicians and hospitals.

Are We Getting Our Money’s Worth?

People unfamiliar with MA might reasonably think that the vast overpayments to the MA plans means they offer insurance superior to that offered by TM. The evidence does not support that inference.

Determining the quality of MA compared with that of TM is more complex than determining the relative costs of the two programs. Cost relies on a single measure, namely, dollars spent. Quality, on the other hand, must be assessed with multiple measures, including breadth of coverage (services covered and exposure to out-of-pocket payments), quality of care, and availability of providers. This multiplicityof-measures problem is aggravated by incomplete and inaccurate data from the plans. Consequently, all three of those categories of quality measurement are very difficult to evaluate.

The least difficult of the measures is breadth of coverage. It is the measure most aggressively promoted by MA plans in their marketing. Federal law and regulations permit MA plans to use a portion of their overpayments to cover services beyond those covered by TM such as pharmacy, dental, vision, and hearing services. In a chapter in its June 2025 report to Congress entitled “Supplemental benefits in Medicare Advantage,” MedPAC reviewed the sparse research on supplemental benefits and concluded, “Altogether, our review of numerous data sources pertaining to MA supplemental benefits reveals a fundamental lack of transparency about how often enrollees use the benefits and plans’ spending for the benefits.”9 One of the few studies that sought to determine the extent to which supplemental services are used by MA enrollees concluded that a large fraction of MA enrollees did not use the extra hearing and vision benefits during the period 2017 to 2021, and during that same period MA plans spent just 10 percent of their overpayments on all supplemental services (not counting the cost of reducing out-of-pocket payments and Part B premiums). Although individual MA plans varied widely, the analysis demonstrated median reductions in out-of-pocket spending by MA beneficiaries (as compared to those in TM) of $23 per dental visit, $20 per pair of eyeglasses, and $22 per Durable Medical Equipment purchase (largely driven by hearing aids).10

To sum up, the MA overpayments finance some services not covered by TM and reduce some out-ofpocket costs, but total expenditures on these benefits is difficult to determine, many enrollees do not use these benefits, and funneling tax dollars through MA plans is an inequitable method of financing supplemental benefits.11

The second category of quality measures—quality of care—is even more confounded than measures of coverage. In 2019, MedPAC declared CMS’s five-starrating report card for MA plans to be so inaccurate it should be replaced.12 In 2025 MedPAC characterized the literature on MA quality as “inconclusive” due to “wide heterogeneity in terms of study populations, metrics evaluated, and data sources used.”13

There are at least two major obstacles to accurate measurement of the quality of health insurance companies: health insurance is a bundled product (it covers thousands of services), and risk-adjustment is needed for most services—but risk adjustment is crude. Because health insurance covers so many services, creating a single accurate measure of quality, or even a few that purport to be all encompassing, is impossible; on the other hand, a measure limited to just a few services (for example, the percent of diabetic enrollees who receive an annual eye exam, kidney function test, and HbA1c test) represents a microscopic and unrepresentative slice of the universe of services typically covered by health insurance. Crude risk adjustment is another obstacle to accurate measurement of quality, just as it is for measurement of cost.

Access to health care constitutes a third category of plan quality measurements, along with coverage and quality of care. Access is affected by multiple variables, including narrow networks and “ghost networks” (networks that do not include the providers listed in network directories). In 2022, the average MA enrollee had access to half as many doctors as TM enrollees did, and 12 million MA enrollees were in networks that excluded 70 percent of the physicians in their county.14,15

Access problems caused by limited networks are often aggravated by inaccurate directories. According to MedPAC’s June 2024 report, “inaccuracies are rampant.”16 The Office of Inspector General for the Department of Health and Human Services reported in 2025 that 55 percent of mental health professionals listed as in-network by MA plans were not available to the plan’s enrollees.17 CMS does very little to enforce network adequacy regulations.18 According to MedPAC, “CMS has the authority to impose sanctions for noncompliance with network adequacy standards, but has never done so.”19

Disputes between MA plans and providers also threaten access. According to a 2025 report from KFF, “Just this year, at least 38 hospital systems serving all or parts of 23 states have cut ties with at least 11 Advantage plans after failing to agree on payment and other issues, according to a review of news releases and press reports. Over the past three years, separations between Advantage plans and health systems have increased 66%… .”20

Aggressive use of prior authorization constitutes another barrier to access for MA enrollees. In a recent review of prior authorization in the Medicare program, Biniek et al. offered this contrast between MA and TM: “Virtually all enrollees in Medicare Advantage (99%) are required to obtain prior authorization for some services—most commonly, higher cost services, such as inpatient hospital stays, skilled nursing facility stays, and chemotherapy. This contrasts with Traditional Medicare, where only a limited set of services, including certain outpatient hospital services, non-emergency ambulance transport, and durable medical equipment, require prior authorization.” According to the authors, the rate at which prior authorization is used is far higher in MA than TM – 1.7 per MA enrollee in 2024 versus 0.02 per TM enrollee. Utilization of prior authorization by MA plans has accelerated dramatically, from 30 million authorizations in 2020 to 53 million in 2024.21

Prior authorization has several destructive effects. The worst, of course, is the impact on patients. Prior authorization often denies necessary services, and even when necessary services are eventually authorized, the delay can damage patient health. Prior authorization of cancerrelated services, for example, extends the delay between diagnosis and treatment and causes higher post-surgical mortality rates.22 Another side effect is long-lasting damage to physician morale.

Conclusions and Recommendations

For several decades the evidence has demonstrated that the MA program is driving Medicare’s costs up, not down, is at best failing to improve quality on balance, and is probably having a net destructive effect on quality. We propose that Congress eliminate the MA overpayments and use those funds to create a low out-of-pocket maximum in TM, as well as adding pharmacy, vision, hearing, and dental coverage to TM. Such a level playing field would more fairly reveal the preferences of Medicare beneficiaries.

Reducing the overpayments will not be simple. Repealing two of the minor sources of overpayment— the counter-productive “quality bonus” and ineffective subsidies for MA plans that operate in rural areas—is not complicated. But reducing the two most important sources of overpayment—favorable selection coupled with crude risk adjustment, and upcoding—is complex and will be difficult. Legislation authorizing reducing overpayments should recognize these problems and include a provision limiting how long CMS will have to determine to what extent the overpayments can be reduced and at what cost.

Beyond helping their patients navigate the important personal decision between Traditional Medicare and Medicare Advantage, physicians should take a leadership role in elevating these concerns within organized medicine and the broader community.

https://digitaleditions.walsworth.com…

References

  1. Falkson JL. HMOs and the politics of health system reform. American Hospital Association; 1980.
  2. Cooper B., Vladeck BC. Bringing competitive pricing to Medicare: Theory meets reality, reality wins. Health Aff (Millwood). 2000;19(5):49-54,49-50.
  3. Medicare and You 2026. https://www.medicare.gov/publications/10050medicare-and-you.pdf
  4. Weisbart E, et al. “No real choices: How Medicare Advantage fails seniors of color.” Physicians for a National Health Program. 2025. Accessed May 21, 2026. https://pnhp.org/system/assets/uploads/2025/10/MAEquityReport2025_Final.pdf
  5. To cite just a portion of that research: The US General Accounting Office, now called the Government Accountability Office (US GAO), published Changes to HMO Rate Setting Method Are Needed to Reduce Program Costs in 1994; Growing Enrollment Adds Urgency to Fixing HMO Payment Problem in 1995; HCFA Could Promptly Reduce Excess Payments by Improving Accuracy of County Payment Rates in 1997; Fewer and Lower Cost Beneficiaries with Chronic Conditions Enroll in HMOs in 1997; and Payments Exceed Cost of Fee-for Service Benefits, Adding Billions to Spending in 2000.
  6. Newhouse, J. “Patients at risk: Health reform and risk adjustment.” Health Aff. Spring (I) 1994:132-146, 132. https://www.healthaffairs.org/doi/epdf/10.1377/hlthaff.13.1.132
  7. “The Medicare Advantage program: Status Report.” Medicare Payment Advisory Commission. March 2025 Report to Congress, Chapter 11, p 323. https://www.medpac.gov/wp-content/uploads/2025/03/Mar25_MedPAC_Report_To_Congress_SEC-1.pdf
  8. “Our Payments, Their Profits.” Physicians for a National Health Program. 2023. https://pnhp.org/system/assets/uploads/2024/11/MAOverpaymentsReportFinal.pdf
  9. “Supplemental Benefits in Medicare Advantage.” Medicare Payment Advisory Commission, Report to Congress, June 2025, p 61. https://www.medpac.gov/wpcontent/uploads/2025/06/Jun25_MedPAC_Report_To_Congress_SEC.pdf
  10. Cai CL, Iyengar S, Woolhandler, S et al. “Use and costs of supplemental benefits in Medicare Advantage, 2017- 2021.” JAMA Netw Open. 2025 8(1); e2454699. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2829183
  11. Weisbart E, et al. “No real choices: How Medicare Advantage fails seniors of color.” Physicians for a National Health Program. 2025. Accessed May 21, 2026. https://pnhp.org/system/assets/uploads/2025/10/MAEquityReport2025_Final.pdf
  12. “Redesigning the Medicare Advantage quality bonus program.” Medicare Payment Advisory Commission. Report to the Congress. June 2019. Chapter 8. https://www.medpac.gov/wp-content/uploads/import_data/scrape_files/docs/default-source/reports/jun19_medpac_reporttocongress_sec.pdf
  13. “The Medicare Advantage program: Status report.” Medicare Payment Advisory Commission. Report to the Congress. March 2025, chapter 11, pp 372-373. https://www.medpac.gov/wp-content/uploads/2025/03/Mar25_MedPAC_Report_To_Congress_SEC-1.pdf
  14. Herman, B. “Medicare Advantage enrollment races past 33 million.” STAT News. Feb 16, 2024. https://www.statnews.com/2024/02/16/medicare-advantage-enrollment/#:~:text=Roughly%2033.4%20million%20adults,for%20the%20taxpayer-funded%20program
  15. Rae M., et al. “Medicare advantage enrollees have access to about half of the physicians available to traditional Medicare beneficiaries.” KFF. October 27, 2025. Accessed May 21, 2026. https://www.kff.org/medicare/medicare-advantageenrollees-have-access-to-about-half-of-the-physicians-available-to-traditionalmedicare-beneficiaries/
  16. “Provider networks and prior authorization in Medicare Advantage.” Report to Congress. MedPAC June 2024, page 77. https://www.medpac.gov/wp-content/uploads/2024/06/Jun24_MedPAC_Report_To_Congress_SEC.pdf
  17. “Many Medicare Advantage and Medicaid Managed Care Plans Have Limited Behavioral Health Provider Networks and Inactive Providers.” Department of Health and Human Services, Office of Inspector General. October 2025. Data Brief: Many Medicare Advantage and Medicaid Managed Care Plans Have Limited Behavioral Health Provider Networks and Inactive Providers, OEI-02-23-00540.
  18. “Provider networks and prior authorization in Medicare Advantage.” Report to Congress. MedPAC June 2024, page 68. https://www.medpac.gov/wp-content/uploads/2024/06/Jun24_MedPAC_Report_To_Congress_SEC.pdf
  19. Rae M., et al. “Medicare advantage enrollees have access to about half of the physicians available to traditional Medicare beneficiaries.” KFF. October 27, 2025. Accessed May 21, 2026. https://www.kff.org/medicare/medicare-advantage-enrollees-have-access-to-about-half-of-the-physicians-available-to-traditional-medicare-beneficiaries/
  20. Biniek JF, Sroczynski N, Freed M, et al. Medicare Advantage insurers made nearly 53 million prior authorization determinations in 2024. KFF, 2026. https://www.kff.org/medicare/medicare-advantage-insurers-made-nearly-53-million-prior-authorization-determinations-in-2024/#6e420acb-2fc14707-8689-ac19594e493a
  21. “Medicare Advantage: A Disadvantage for Complex Cancer Surgery Patients.” Raoof, M, Ituarte PHG, Haye S, et al. J Clin Oncol. 2023;41(6):1239-1249 (2023). https://ascopubs.org/doi/pdf/10.1200/JCO.21.01359?role=tab
  22. If I Betray These Words: Moral Injury in Medicine and Why It’s So Hard for Clinicians to Put Patients First. Dean, W and Talbot S. 2023.