Medicare Advantage plans are paid through risk adjustment. The government sends each plan a fixed monthly amount per enrollee, and that amount rises when the enrollee’s record contains certain diagnosis codes. The system assumes sicker patients cost more to cover, so plans with documented sicker populations receive higher payments.
That design creates a standing incentive to add diagnosis codes. When a code is added without clinical support, or amended after the fact to raise a payment, the government treats it as upcoding. Medicare Advantage upcoding fraud is the practice of submitting inaccurate risk-adjustment data to inflate those monthly payments.
The federal government has spent years examining the practice. A 2020 study of Medicare risk adjustment found that enrollees in private Medicare plans generate diagnosis-based risk scores 6% to 16% higher than the same enrollees would generate under fee-for-service Medicare. In October 2024, the HHS Office of Inspector General estimated $7.5 billion in Medicare Advantage overpayments tied to questionable health risk assessments. Two settlements in 2026 show the enforcement reality behind those numbers.
What Counts as Medicare Advantage Upcoding
Upcoding occurs when a provider submits codes to Medicare, Medicaid, or a private insurer for more serious, and more expensive, diagnoses than the record supports. In Medicare Advantage, the fraud takes a specific form. The insurer knowingly submits, or causes a contractor to submit, inaccurate risk-adjustment data so that the plan collects a larger capitated payment than the enrollee’s true condition warrants.
The Centers for Medicare & Medicaid Services defines Medicare fraud in similar terms: a provider knowingly submits false information to the federal government to receive Medicare payment. The KFF program integrity analysis published in March 2025 walks through that definition and the tools the agency uses to identify improper payments.
Phantom Diagnoses and Home-Visit Upcoding
In January 2025, I published a post on this site titled Phantom Diseases and Medicare Advantage Fraud. The post examined a Wall Street Journal investigation from August 2024 that documented how Medicare Advantage plans used home visits to add diagnosis codes for conditions with little or no foundation in the patient’s record.
The phrase “phantom diseases” describes the result. A patient is assessed at home, often briefly, and the encounter generates codes for conditions the patient was never treated for and may not have. The codes raise the patient’s risk score, and the plan’s payment rises accordingly. The pattern is not a billing error. It is a deliberate use of encounters to manufacture risk-adjustment data.
Kaiser Permanente Pays $556 Million Over Retrospective Addenda
The enforcement pace picked up in January 2026. Becker’s Hospital Review reported that Kaiser Permanente reached a $556 million settlement with the Department of Justice over Medicare Advantage upcoding allegations. Office of Public Affairs | Kaiser Permanente Affiliates Pay $556M to Resolve False Claims Act Allegations | United States Department of Justice
The conduct at issue involved retrospective addenda. According to the reporting, Kaiser used data-mining queries after patient visits to add diagnosis codes that increased risk scores and payments. The addenda were generated after the encounter, rather than from conditions identified and treated during the visit itself. Those retroactive additions formed the core of the government’s case.
Villages Health System Self-Discloses and Pays $541.5 Million
The second major settlement came from a physician practice. On August 27, 2026, the Justice Department announced that Villages Health System LLC agreed to pay $541.5 million to resolve Medicare Advantage upcoding allegations. Office of Public Affairs | The Villages Health System LLC Agrees to $541.5M Settlement to Resolve False Claims Act Allegations | United States Department of Justice
Villages self-disclosed the conduct to the government. The Justice Department described a process of “retrospective amendments” and “sprints” in which Villages added diagnosis codes to patient records months after the relevant visits, and in some cases more than a year later. Some of the codes were added by individuals who were no longer with the practice when the amendments were made.
The distinction matters. This was not a health plan with a data-mining system generating addenda at scale. It was a practice that went back into old records, added diagnoses, and submitted the amended data to Medicare Advantage plans to raise risk-adjusted payments.
The Broader Enforcement Picture
These settlements did not occur in isolation. Most major Medicare Advantage insurers have been accused of upcoding, and the Senate has been examining the program’s payment practices. In January 2026, a Senate investigation found that UnitedHealth, the $400 billion healthcare giant, aggressively gamed Medicare Advantage, according to Healthcare Dive.
The HHS Office of Inspector General has continued to flag the problem. In June 2026, the Center for Economic and Policy Research reported on a new HHS report finding millions in Medicare Advantage overpayments tied to upcoding. The Justice Department’s scrutiny dates back further. Medicare Advocacy noted in February 2024 that Medicare Advantage plans were already under examination by the department.
The federal definition remains constant. Medicare fraud occurs when a provider knowingly submits false information to the federal government to receive Medicare payment. The 2026 settlements show the government applying that definition to risk-adjustment data, not just to fee-for-service claims.
What the Settlements Mean for Providers
For providers, the lesson is direct. Diagnosis codes affect payment in Medicare Advantage, and codes added after a visit attract scrutiny. The Kaiser settlement shows that retrospective addenda generated by data-mining queries are treated as fraud when the underlying records do not support the codes. The Villages settlement shows the same rule applied to a physician practice that amended records months or more than a year after care.
