DOJ settled $777M in Medicare Advantage fraud cases against Cigna, Humana, and Anthem. Here is why.

In September 2023 [6], The Cigna Group agreed to pay $172 million to settle a False Claims Act case brought by the U.S. Department of Justice [6]. The allegation: Cigna’s Medicare Advantage business had submitted diagnostic codes for its enrollees that were not supported by the patients’ medical records, but that triggered higher payments from the federal government [6]. The settlement did not require Cigna to admit liability. The pattern of conduct the case described, the same pattern the DOJ and HHS-OIG had documented in cases against UnitedHealth, Humana, and Anthem in prior years, has continued [3][4][5].

Medicare Advantage is the privately administered alternative to traditional Medicare. About half of Medicare beneficiaries are enrolled in it [2]. The federal government pays the private insurer a per-enrollee monthly payment, adjusted by the enrollee’s documented health conditions. Sicker patients trigger higher payments. The structure is designed to ensure that plans that take on sicker enrollees are not financially penalized. The structure also creates a direct financial incentive for the plan to document as many qualifying conditions as possible, accurate or not [1][3].

The Official Story

The Medicare Advantage industry’s stated position is that the risk-adjustment system is appropriate, that diagnostic coding represents an accurate medical record of patient conditions, and that the False Claims Act cases that have been settled were either disputed disagreements over coding interpretation or isolated instances of contractor error. Plans uniformly maintain that they comply with the Centers for Medicare and Medicaid Services regulations on diagnostic submission, including the requirement that submitted codes reflect documented physician findings during face-to-face encounters.

The HHS-OIG 2020 report on Medicare Advantage diagnostic coding documented a pattern that contradicts the industry’s framing [3]. The report estimated that diagnoses reported only on health risk assessments, the home-visit reviews that plans routinely conduct on their enrollees, generated approximately $9.2 billion in estimated Medicare Advantage payments in 2017 alone [3]. Many of those diagnoses were not corroborated by any other clinical record from the same year. The OIG’s framing, careful and bounded, was that the pattern raised concerns. The numbers were the same numbers DOJ filings have used in subsequent cases [4][5][6].

Follow the Money

The financial mechanics of Medicare Advantage upcoding work in three layers.

A Medicare Advantage plan enrolls a beneficiary. The plan receives a monthly capitated payment from CMS that depends on the enrollee’s risk score, calculated from the diagnostic codes attached to that enrollee in the previous year [2]. A 65-year-old with documented diabetes, heart failure, and chronic kidney disease generates a substantially higher monthly payment than a 65-year-old documented as having only hypertension. The risk score is built from the hierarchical condition category (HCC) framework that CMS uses to allocate payments [2].

The plan has multiple ways to add diagnostic codes to an enrollee’s record. The most clinically appropriate is the codes a treating physician attaches during a face-to-face visit when actually diagnosing or managing the condition. The least clinically appropriate, but most efficient at generating revenue, is the codes a plan-contracted reviewer attaches during a home visit or chart review conducted specifically to identify additional diagnoses that can be coded for risk-adjustment purposes [3]. The DOJ cases against UnitedHealth, Humana, Cigna, and others have alleged systematic use of the latter pathway in ways that violate the federal False Claims Act [4][5][6].

The resulting payment difference, multiplied across an insurer’s millions of enrollees, runs into the billions of dollars per year per plan [2][3]. The settlements, when they occur, recover a small fraction of the cumulative upcoded payments. The cases are slow, the discovery is contested, the legal costs are substantial, and the plans treat the eventual settlement as a cost of doing business at the scale Medicare Advantage operates [4][5][6].

The Network

The largest Medicare Advantage insurers in the United States, viewed by enrollment, are UnitedHealth Group, Humana, CVS Health (which operates Aetna), Centene, and Cigna [2]. UnitedHealth’s Medicare Advantage business operates inside the same corporate parent that owns Optum and OptumRx, the pharmacy benefit manager analyzed in our earlier piece on the three companies that set American drug prices. Aetna sits inside CVS Health, also part of the same corporate cluster. The Medicare Advantage market is concentrated in roughly the same hands that concentrate the U.S. pharmacy-benefit market.

The DOJ has filed or intervened in False Claims Act cases against most of the major plans over the past decade [4][5][6]. The UnitedHealth case filed in 2017 remains pending [4]. The Humana case settled for $11.4 million in December 2024 [5]. The Cigna case settled for $172 million in September 2023 [6]. Anthem, now Elevance Health, settled an earlier case for $594 million in 2020 [1]. Aetna has had multiple cases at smaller scales [1]. The collective settlements, while substantial in absolute terms, are small relative to the annual risk-adjustment revenue at issue [2].

What Was Buried

The 2024 GAO report on Medicare Advantage oversight documented that CMS does not consistently track which plans have produced patterns of coding behavior consistent with the patterns the DOJ has alleged in False Claims Act cases [1]. The agency’s risk-adjustment audit program, which would in theory detect systematic upcoding, has historically reviewed a small fraction of the contracts and a small fraction of the diagnostic codes within those contracts [1]. The audit findings, when they occur, produce payment recoveries that are again small relative to the totals at stake [1]. CMS has announced rule changes that would tighten the audits, but implementation has been delayed multiple times by industry litigation [1][2].

MedPAC’s 2024 annual report estimated that Medicare Advantage payments per enrollee exceed traditional Medicare payments per enrollee by approximately 22 percent, after adjusting for the underlying health of enrollees, and that diagnostic upcoding accounts for a substantial share of that gap [2]. The Commission’s framing, again careful, was that the payment policy needs structural reform. The structural reform has not been enacted at scale [2].

The Stakes Now

Medicare Advantage enrollment continues to grow. Roughly 33 million people, more than half of Medicare beneficiaries, were enrolled in private plans in 2024 [2]. The federal payment to the plans for Medicare Advantage exceeded $450 billion in 2024 [2]. The structural condition that produces the upcoding incentive, federal payment that scales with documented diagnoses, has not changed. The False Claims Act cases continue to be filed. The settlements continue to be paid. The next case is being investigated.

Reform proposals fall into two categories. The first would tighten the risk-adjustment audit process and require plans to repay overpayments at scale matching the audit findings. The second would change the underlying capitation structure to reduce the financial incentive for diagnostic upcoding. Industry opposition has been substantial in both cases. The plans operate in nearly every congressional district, employ large workforces, and contribute heavily to congressional campaigns of both parties. The structural reforms remain proposed rather than enacted [2].

The One Thing That Matters

The federal government pays Medicare Advantage plans for documented diagnoses. The plans make money by documenting more diagnoses. The plans hire reviewers whose job is to find diagnoses to document. The DOJ files cases when the documentation crosses the line into clear fraud. The cases settle. The reviewers keep documenting. The federal payment to the plans continues to grow.

The dynamic resolves only if the structural payment relationship changes. As long as the federal government pays per documented diagnosis, and as long as the federal audit capacity remains a small fraction of the documentation flow, the upcoding pattern is the rational response to the payment design. The cases the DOJ files are not deterrents at the scale of the underlying revenue. They are operating costs, paid at the time of settlement, deducted from the much larger ongoing federal payment stream that justifies the model.