Complete Health: Psychiatric Coding Guidance and Physician Prompts

In the Complete Health Medicare Advantage fraud settlement, Complete Health Partners Holdings agreed to pay $14.1 million over allegations that incorrect psychiatric and substance-use coding guidance led doctors to add unsupported diagnoses. The Department of Justice alleged that the company helped generate the unsupported codes and then took a share of the higher payments they produced. Taxpayers paid for diagnoses the patients’ records did not justify, according to the government’s account.

The Justice Department announced the settlement on August 3, 2026. Complete Health, headquartered in Jacksonville, Florida, was a management services organization that managed, owned or otherwise operated affiliated provider groups in Florida, Alabama and Colorado. The allegations concerned conduct from 2020 through 2023.

How Incorrect Guidance Allegedly Became Physician Prompts

The government alleged that Complete Health distributed incorrect coding guidance to both coders and physicians. Coders allegedly reviewed patients’ medical records for additional chronic-condition diagnoses. Complete Health then allegedly prompted doctors to add the identified codes even when they lacked support or clinical justification.

According to the Justice Department, doctors added inaccurate diagnoses in response. Those codes increased payments to the Medicare Advantage plans, which passed a portion of the increase to Complete Health.

The allegation reaches back before the code went to the insurer. It concerns the instructions that shaped the review, the diagnosis the coder selected and the request that reached the doctor. A physician looking at a single query sees the proposed addition. Someone familiar with the coding guidance may see why the same kind of addition keeps being requested. We look at those parts together when assessing risk-adjustment fraud.

How the Additional Diagnoses Reached the Payment

Complete Health’s contracts gave it a percentage of the Medicare payments the plans received. That arrangement connected the provider organization’s revenue to the patient’s recorded risk score.

Medicare pays private plans a monthly amount for each enrolled beneficiary, adjusted for expected health costs. The model groups eligible diagnoses into clinically related categories that contribute to a risk score. Generally, more severe diagnoses or conditions associated with costlier treatment produce higher payments. That is how a diagnosis becomes revenue even when the billing concern is a code added to an existing record.

The American Institutes for Research explains the incentive in its June 2026 report, Medicare Advantage Fraud, Waste, and Abuse Policy Options. It describes sub-capitation: plans pay a provider group upfront, and that group pays its physicians on a capitated basis. The contract can reward coding intensity: set as a percentage of premium, it raises the provider’s revenue whenever more intensive coding raises Medicare’s payment to the plan.

That analysis explains why provider compensation belongs beside the coding instructions in a whistleblower inquiry. The contract shows who benefits when another diagnosis reaches the payment calculation.

HCC 55 and HCC 59 Were the Categories at Issue

The Complete Health allegations concerned HCC 55, Drug and Alcohol Dependence, and HCC 59, Major Depressive, Bipolar, and Paranoid Disorders.

The Justice Department described several problems with the submitted diagnoses: they were not clinically valid, lacked proper support in the medical record, or had not been considered in the patient’s care, management or treatment. Those distinctions matter to someone reviewing a query about major depression or dependence.

A request to add a condition can look routine on a physician’s screen while carrying a payment consequence elsewhere in the business.

The Case Began on the Insurer’s Side

A former Associate Director of Risk Adjustment at Viva Health brought the qui tam case. Her employment was at the insurer, rather than Complete Health. That matters to healthcare insiders who work on the plan’s side of diagnosis submissions and may recognize problems originating in a provider group’s coding practices.

The government joined the case for settlement.

The Justice Department said the whistleblower would receive approximately $2.47 million as her share of the federal recovery. A whistleblower case under the False Claims Act can pay 15% to 30% of what the government recovers.

Where the Prompt Meets the Patient Record

We examine how the coding guidance, the request to the doctor and the encounter record connect to the diagnosis submitted for payment. A coder may know why the condition was selected, a physician may know why it did not fit the visit, and a plan or finance employee may see how it reached the payment calculation.

A Depression Diagnosis That Conflicted With the Visit

A separate whistleblower case involving Independent Health, Group Health Cooperative and the coding vendor DxID shows why the encounter record matters. The government joined that case and, in its September 2021 complaint, alleged that DxID coded major depressive disorder from a patient’s problem list and Group Health Cooperative submitted it despite conflicting progress notes.

The complaint quoted the physician’s assessment, which recorded the depression as resolved, a passing response to his sister’s death, and noted that he had asked to stop his antidepressant earlier in the year when he felt better and was still doing well.

The problem list supplied a diagnosis. The physician’s assessment described depression that had resolved. The government’s allegation was that the coding process used the former despite the latter. An insider who understands how the review handled conflicting information can explain a decision that the final diagnosis code conceals.

That complaint also alleged that DxID encouraged providers to add diagnoses after the visit, sometimes up to a year later. A Medicare Advantage physician asked to add a diagnosis after a visit may recognize that workflow. In the Complete Health case, the government’s focus was the incorrect guidance and prompts for unsupported additions.

Federal Scrutiny of Coding Prompts and Risk Sharing

The Department of Health and Human Services Office of Inspector General’s February 2026 Medicare Advantage compliance guidance identifies physician prompts and provider risk-sharing arrangements among the practices examined in federal investigations. It describes electronic-record queries, including prompts generated by artificial intelligence, that encourage diagnoses patients did not have or that did not affect their care.

The same guidance addresses chart reviews that inappropriately increase risk scores, failures to remove unsupported codes, and unsupported diagnoses from home assessments. It separately identifies providers submitting unsupported diagnoses to inflate payments they receive from plans under risk-sharing arrangements.

The overlap with Complete Health’s alleged mechanism is direct: a request to add the diagnosis, followed by a financial benefit to the organization behind it. For the insider, the useful perspective spans both. The clinical record explains the diagnosis, and the contract explains why someone wanted it added.

You Saw What the Query Asked For

If you know how a coding instruction became an unsupported diagnosis, speak with our whistleblower experts about the coding requests you saw. We work with you to build your strongest Medicare Advantage whistleblower case from the clinical contradiction through the payment trail. We bring in Medicare Advantage whistleblower lawyers who understand provider coding and risk sharing, and stay beside you through the case.