Medicare Fraud, Waste, and Abuse

Medicare loses more to improper payments than most countries spend on their entire health systems. CMS’s FY 2025 improper payment estimates totaled approximately $57 billion across Medicare FFS ($28.83 billion at a 6.55% error rate), Medicare Part C ($23.67 billion at 6.09%), and Medicare Part D ($4.23 billion at 4.00%). These figures are not fraud estimates. They measure payments that did not meet program requirements, encompassing overpayments, underpayments, and payments where insufficient documentation prevented a determination of whether the payment was proper. The actual fraud figure is unmeasurable with precision because fraud, by definition, involves concealment. But the improper payment estimates establish a floor: at least $57 billion annually flows through Medicare in ways the program’s own rules do not authorize, and the enforcement apparatus recovers only a fraction of it.

The FWA landscape spans both FFS and MA, covers payment vectors from billing fraud to risk adjustment gaming to hospice eligibility manipulation, and intersects directly with the risk adjustment reform agenda that Series 2 of this publication documents. The chart review exclusion (MCR-02.02), the encounter-based RA future (MCR-02.04), and the DOJ enforcement surge documented in earlier articles are not just payment accuracy reforms. They are program integrity mechanisms designed to close the specific channels through which improper payments flow.

The Scope of Medicare FWA

#

The distinction between fraud, waste, and abuse matters analytically even when the three categories overlap in practice. Fraud is intentional: a provider billing for services not rendered, a plan submitting diagnosis codes it knows are unsupported, or a beneficiary using someone else’s Medicare card. Waste is systemic: payment mechanisms that produce expenditures without corresponding value because of design flaws, administrative inefficiency, or misaligned incentives. Abuse is improper but not necessarily intentional: a provider consistently upcoding evaluation and management visits beyond the complexity the clinical encounter warrants, not because of a deliberate scheme but because of inadequate training, poor documentation habits, or financial incentives that reward higher billing without oversight.

CMS’s FY 2025 FFS improper payment rate of 6.55% represents $28.83 billion, down from 7.66% and $31.70 billion in FY 2024. The reduction is notable, but the absolute number remains enormous. The majority of FFS improper payments fall into two categories: insufficient documentation to support the service billed and documentation that failed to demonstrate medical necessity. Durable medical equipment, prosthetics, orthotics, and supplies (DMEPOS) stood out with a 24.12% improper payment rate, approximately $2.27 billion. DME has been the most persistent FFS fraud vector for decades, with schemes involving billing for equipment never delivered, prescribing unnecessary equipment through kickback arrangements with ordering physicians, and using telemarketing and lead generation to identify Medicare beneficiaries whose identifiers can be used to submit fraudulent claims.

Other persistent FFS fraud vectors include unbundling (separating services that should be billed together as a single unit to inflate total reimbursement), E&M upcoding (billing office visits at higher complexity levels than the clinical encounter supports), and home health certification fraud (improper physician certification of homebound status to trigger home health services the beneficiary does not need or qualify for). DOJ’s FY 2025 National Health Care Fraud Takedown charged 324 defendants tied to over $14.6 billion in alleged fraud, including telemedicine and genetic testing schemes involving 49 defendants charged in connection with $1.17 billion in fraudulent claims, and wound care fraud exceeding $1.1 billion tied to fraudulent use of amniotic wound allografts.

The Part C improper payment rate increased from 5.61% in FY 2024 to 6.09% in FY 2025, or $23.67 billion. CMS noted that most Part C improper payments were attributable to situations where the MA organization’s supporting documentation failed to substantiate the beneficiary diagnosis data submitted for payment. This is the improper payment expression of the coding intensity and chart review problems that MCR-02.02 and MCR-02.07 analyze from the payment accuracy perspective. An unsupported diagnosis submitted for risk adjustment is, by CMS’s improper payment definition, an improper payment, regardless of whether the plan intended to defraud the program or simply submitted a diagnosis its documentation could not support.

MA-Specific FWA

#

The MA program’s FWA profile is structurally different from FFS because the payment mechanism is different. In FFS, fraud typically involves billing for services: fictitious claims, inflated claims, or claims for unnecessary services. In MA, the dominant FWA vector is risk adjustment manipulation: submitting diagnosis codes that inflate risk scores and capitation payments beyond what the beneficiary’s clinical reality supports.

Chart review upcoding is the mechanism this series has examined in detail (MCR-02.02). Retrospective chart reviews that identify diagnoses from historical records, problem lists, or incidental documentation, rather than from active clinical management during a provider encounter, inflate risk scores in ways that generate improper payments. The $7.2 billion chart review exclusion CMS proposed for CY 2027 is simultaneously a payment accuracy reform and an anti-FWA mechanism. CMS has framed it as program integrity: diagnoses not linked to encounters should not generate payments because there is no evidentiary basis to conclude the provider is actively managing those conditions.

Diagnostic code inflation extends beyond chart reviews. Plans that systematically train their coding staff, CDI programs, and contracted providers to capture diagnoses at the highest defensible specificity level are engaging in a practice that straddles the line between aggressive documentation and improper coding. The line is defined by whether the diagnosis is supported by the clinical record and whether the provider independently assessed the condition. When the plan’s coding infrastructure generates diagnoses that the provider did not independently evaluate, the coding has crossed from documentation improvement to improper payment generation, regardless of whether anyone intended to defraud the program.

Marketing and enrollment fraud operates as a separate MA-specific vector. The DOJ complaint against Aetna, Humana, Elevance, eHealth, GoHealth, and SelectQuote (MCR-04.03) alleges that disguised kickback payments to brokers produced enrollments tainted by Anti-Kickback Statute violations, which DOJ argues renders the associated capitation payments false claims under the FCA. Phantom networks, where plans list providers in directories who are not actually available to see patients, constitute a form of program abuse because beneficiaries enrolled based on network representations that did not reflect accessible care. OIG has documented MA prior authorization denial rates for services that would have been covered under Original Medicare, raising questions about whether prior authorization is being used as a cost management tool that crosses the line into improper denial of covered services.

The Hospice FWA Surge

#

Hospice represents a distinct and growing FWA concentration within Medicare that intersects with both FFS and MA payment streams.

The dominant hospice fraud pattern is length-of-stay manipulation. Medicare pays hospice on a per-diem basis, creating a financial incentive to enroll beneficiaries and keep them enrolled as long as possible. Longer enrollment generates more per-diem payments. Fraudulent hospice operations enroll beneficiaries who do not meet the terminal illness criteria (a life expectancy of six months or less if the disease runs its normal course), keep them enrolled for months or years beyond what their clinical trajectory would justify, and bill Medicare for the per-diem payments throughout. OIG found that Medicare improperly paid acute-care hospitals an estimated $190 million over a five-year period for outpatient services provided to hospice enrollees, reflecting how payment rules break down across care settings when hospice enrollment is improper.

General Inpatient (GIP) level-of-care gaming compounds the problem. GIP reimbursement is substantially higher than routine hospice care, and some hospice providers bill at the GIP level for patients who do not meet the clinical criteria for inpatient-level symptom management. The for-profit hospice sector has a documented quality and integrity differential relative to nonprofit hospice providers: for-profit hospices have higher rates of long stays, higher rates of live discharge (suggesting enrollment of patients who were not truly terminal), and lower staffing ratios (MCR-05.12).

The hospice FWA problem is growing because hospice enrollment is growing, particularly among MA enrollees whose plans contract with hospice providers. As hospice becomes a larger share of Medicare spending, the dollar magnitude of hospice-related improper payments increases proportionally, and the enforcement apparatus has not scaled at the same rate.

Enforcement Trends

#

DOJ and OIG are operating at historically high enforcement intensity. FCA settlements and judgments exceeded $6.8 billion in FY 2025, the highest annual total in the statute’s history, with over $5.7 billion from healthcare matters. The DOJ-HHS FCA Working Group, relaunched in July 2025, identified six priority enforcement lanes: MA risk score integrity, drug and device pricing, network adequacy and patient access, kickbacks, defective devices, and EHR manipulation.

Qui tam actions (whistleblower lawsuits filed under the FCA) continue to drive the enforcement pipeline. More than one-third of FY 2025 recoveries, nearly $2.3 billion, originated from cases the government declined to intervene in but that private relators pursued independently. The whistleblower bar is well-resourced and experienced in healthcare FCA litigation, and the pipeline of unsealed qui tam complaints suggests sustained enforcement volume for the foreseeable future.

Corporate integrity agreements (CIAs) imposed as conditions of FCA settlements require plans and providers to implement specific compliance infrastructure: independent review organizations, annual claims audits, coding accuracy reviews, compliance officer reporting obligations, and board-level oversight requirements. A CIA changes how an organization operates for five years, imposing external monitoring costs and operational constraints that go beyond the financial settlement amount. The Kaiser Permanente $556 million settlement did not include a publicly announced CIA, which was itself notable given the settlement’s size.

CMS is investing in AI-powered fraud detection, including a “war room” operation that analyzes claims in real time as they arrive at Medicare Administrative Contractors, identifying fraudulent patterns before payments are made rather than pursuing recovery after the fact. CMS reported preventing over $4 billion from being paid out in response to false or fraudulent claims and suspending or revoking billing privileges for 205 providers in the lead-up to the FY 2025 takedown. DOJ announced the creation of a Health Care Fraud Data Fusion Center leveraging cloud computing, AI, and advanced analytics to identify emerging schemes and break down data silos across enforcement agencies.

CMMI model integrity adds a new enforcement dimension as the innovation center’s model portfolio expands. WISeR’s AI-powered prior authorization system creates a risk that automated decision-making becomes an improper denial mechanism if the algorithms are not calibrated to clinical appropriateness. AHEAD’s geographic accountability framework creates new gaming vectors if hospitals under global budgets reduce necessary care to meet cost targets. Each model’s incentive structure creates a corresponding integrity risk that CMS must monitor alongside the existing FFS and MA fraud landscape.

FWA and Risk Adjustment Reform

#

The risk adjustment reform agenda documented in Series 2 is, at its core, an anti-FWA agenda. The chart review exclusion closes a specific channel through which unsupported diagnoses generate improper payments. Encounter-based RA narrows the documentation pathway to one that is auditable and clinically verifiable, reducing the surface area for coding fraud. The V28 model’s constraining methodology reduces the marginal return on aggressive coding within clinical hierarchies. A higher CPA, if CMS ever moves above the statutory 5.9% minimum, would address the aggregate coding intensity differential that MedPAC estimates produces $40 billion in annual overpayment.

Enhanced RADV audits complete the architecture. CMS committed in May 2025 to audit all eligible MA contracts each payment year and is beginning to apply extrapolation of audit findings starting with Payment Year 2018. Extrapolation means that the improper payment rate found in an audited sample is projected across the plan’s entire enrolled population, producing recovery obligations that can reach hundreds of millions of dollars for a large plan. The combination of extrapolation with the chart review exclusion and encounter-based RA creates a three-layer integrity framework: the exclusion prevents future improper payments from unlinked chart reviews, encounter-based RA prevents improper payments from non-encounter diagnoses when it arrives, and RADV extrapolation recovers past improper payments identified through audit.

The comprehensive anti-FWA agenda, if fully executed, would address the structural conditions that produce the $23.67 billion in Part C improper payments and the additional tens of billions in coding intensity-driven overpayment that MedPAC documents. Whether it is fully executed depends on CMS’s implementation capacity (MCR-03.05), DOJ’s sustained enforcement posture, and the political economy of reforming a program that now covers 55% of Medicare beneficiaries and generates hundreds of billions in annual revenue for the insurance industry.

Related Reading

#

MCR-02_02 Unlinked Chart Reviews: The $7.2 Billion Risk Adjustment Reckoning MCR-05_12 Hospice in Crisis: Benefit Design, Quality Failures, and the Medicare Spending Surge MCR-02_07 The MA Overpayment Ledger: What It Costs the Trust Fund and What Reform Would Save