For too long, Medicare Advantage has rewarded plans for becoming better at documenting how sick their members are. The next era should reward them for demonstrating how much healthier they make their members. Medicare Advantage is now the dominant alternative to traditional Medicare, enrolling about 55% of beneficiaries eligible for the program.
It is popular for understandable reasons: many plans charge little or no additional premium beyond the Part B premium and offer lower cost sharing or supplemental benefits such as dental, vision and hearing coverage. The paradox is that Medicare Advantage can be less expensive for the beneficiary while costing the Medicare program more per enrollee.
Medicare Advantage was designed as a private-plan alternative to traditional fee-for-service Medicare, not as a supplement to it. Its payment system, however, created a powerful incentive: document more diagnoses, increase the risk score, receive more money. The incentives worked — perhaps too well. Over time, Medicare Advantage has become extraordinarily sophisticated at coding.
Medicare Advantage needs to move from rewarding coding to rewarding caring.
This is not just a theoretical issue. America’s largest health insurers just reported another remarkably profitable quarter. In the second quarter of 2026, six of the nation’s largest insurers collectively reported roughly $13.5 billion in net income.
Public disclosures and earnings calls from UnitedHealthcare, Humana and Aetna indicate that MA plans contributed significantly to their overall profitability.
Meanwhile, CMS raised expected Medicare Advantage payments by 5.06% for 2026, adding more than $25 billion in a single year. Medicare Payment Advisory Commission (MedPAC) estimates that Medicare will pay Medicare Advantage plans approximately $615 billion — about $76 billion, or 14%, more than traditional Medicare would spend for those beneficiaries.
There is nothing inherently wrong with insurers making money; profit can reward innovation, efficiency and better care. But when taxpayers provide hundreds of billions of dollars to these companies, taxpayers deserve to know what they are buying.
Because of a lack of transparency, we do not know how much Medicare Advantage contributes to health and well-being, and we do not fully understand its effect on costs. Medicare Advantage should not ask America to trust that it creates value. It should be required to prove it.
Here is the paradox. MedPAC estimates that MA plans’ projected spending on medical services, before administrative costs, is 15% to 18% below what comparable care would have cost in traditional Medicare. Yet Medicare is projected to pay MA plans about 14% more overall than it would spend for comparable beneficiaries in traditional Medicare.
These are different flows of money: the first is what MA plans project they will spend providing medical care; the second is what the federal Medicare program pays the plans. The contrast is striking: taxpayers can pay more even while plans spend less on medical services. At University Hospitals, we contract with MA plans at 100% of Medicare rates.
Yet in 2025, MA ultimately paid us 9.5% less than traditional Medicare, resulting in $95 million less revenue, largely due to higher denial rates.
What explains this gap — and the profits that presumably accompany it? We do not know, because insurers’ public financial statements do not provide enough detail to disentangle the causes. Some factors may be benign. Plans may negotiate lower prices, eliminate unnecessary care, substitute home care for hospitalization and coordinate chronic disease better.
But there are also reasons for concern. MedPAC identifies favorable selection as a major contributor to higher MA payments, estimating that it raises 2026 payments by roughly 11%. This does not simply mean that MA beneficiaries are “healthier.” It means that, after accounting for the characteristics captured by Medicare’s risk-adjustment model, MA enrollees tend to have lower expected spending than their risk scores would predict. Medicare therefore pays plans as though these beneficiaries will cost more than their underlying spending patterns suggest.
Coding intensity is a separate but reinforcing issue. MedPAC projects that, even after CMS’s coding adjustment, MA risk scores in 2026 will remain about 4% higher than they would have been if the same beneficiaries were in traditional Medicare, contributing about $22 billion to higher payments.
More diagnoses and higher risk scores do not necessarily mean the patients are actually sicker; they can reflect more intensive documentation of conditions that already exist. Favorable selection and coding can therefore coexist: a population can cost less than its risk score predicts while also being coded more intensively.
These changes reflect innovation in coding and scoring. Substantial technology investment has focused on helping health plans identify diagnoses, improve documentation and optimize risk scores. But those innovations do not necessarily improve patient outcomes.
We have built sophisticated technology to find and document illness; we should be equally sophisticated at preventing illness and creating health. Finding another diagnosis can increase revenue, but it does not control blood pressure, prevent hospitalization or help an older adult remain independent at home. Technology can make healthcare better, but it can also make saying no faster and cheaper.
Another murky issue is the logic behind prior authorization — the common requirement that a clinician obtain a health plan’s approval before delivering a service, drug or device. Prior authorization creates value for patients and hospitals when it prevents unnecessary interventions.
But prior authorization can also destroy value by delaying or denying necessary care. An Annals of Internal Medicine study examining three large national insurers found that only approximately 14% of procedure codes requiring prior authorization at one insurer required it at all three. If science were the primary driver, we would expect much greater agreement. Instead, the rules vary remarkably across plans.
In a 2026 review of skilled nursing facility admissions, HHS Office of Inspector General found that Medicare Advantage organizations overturned 95% of appealed prior-authorization denials, raising concern that some beneficiaries were initially denied medically necessary care.
Automation and predictive algorithms add another concern, but the evidence should be stated carefully. A Senate investigation documented increasing automation in prior-authorization workflows as denial rates for post-acute care rose at some large insurers; the insurers have disputed that AI makes final denial decisions.
The stronger point is that automated tools can accelerate a flawed process at scale. When denials are later overturned, patients face delays, health systems absorb the costs of continued care and capacity constraints, and clinicians and staff shoulder the administrative burden of appeals.
Technology is not creating value if it merely reduces transaction costs for the payer by shifting costs and friction to patients and providers.
When hospitalizations decline while mortality, functional status, healthy days at home and total cost improve, that is genuine value. When utilization falls because necessary care is delayed or denied and outcomes worsen, that is rationing by friction.
So how do we distinguish better care from rationed care? Make payment decisions transparent and measure the outcomes. CMS should establish a standardized, audited Medicare Advantage Value Statement that allows the public to answer four questions for every plan: What did you pay for? What did you deny? What clinical rules did you use to make those decisions? What outcomes did you achieve?
We may find that Medicare Advantage plans lead to genuine advances in quality at lower cost. If so, we should celebrate those gains — but we should also be able to see them. In a taxpayer-financed program of this scale, not knowing is no longer acceptable.
Transparency must extend to every aspect of Medicare Advantage operations. In the case of prior authorization and claim approvals, the choice of what to pay for should be grounded in the best available science, with the rationale and data underlying it available to all.
Plans should disclose coverage and payment policies, prior-authorization requirements, denial and appeal rates, and the clinical evidence underlying those decisions. CMS should connect those decisions to meaningful, risk-adjusted outcomes including mortality, hospitalization, complications, functional status and days healthy at home.
The goal is to learn what works and what does not in Medicare Advantage. If plans coordinate care better, eliminate waste and keep people healthier, transparency will prove it, and the best plans should be rewarded. But we have built a roughly $600-billion taxpayer-funded program without giving the public enough information to know whether we are purchasing better health or better margins.
Payment rules, denial rates, profits and outcomes should all be transparent. Patients, clinicians, policymakers and taxpayers deserve to know what their money is buying.
Peter Pronovost, MD, PhD, FCCM, Chief Quality and Clinical Transformation Officer, and President of the Healthcare Transformation Institute, University Hospitals Cleveland
Bradley C. Bond, CFA, CPA, MBA, Chief Financial Officer, University Hospitals Cleveland
