Medicare Advantage Risk Score Targets: When Finance Chooses the Diagnosis

Medicare Advantage risk score targets can put a revenue goal ahead of the medical record. The whistleblower complaint against Martin’s Point Health Care alleges that finance chose a target intended to increase payments while staying below a supposed audit trigger. The complaint says chart reviewers’ bonuses depended on hitting a revenue goal and management rejected deleting erroneous codes because doing so would make that goal harder to reach. The target, the incentive and the decision to leave the errors in belong in the same story.

Medicare Advantage risk adjustment changes a plan’s monthly payment through a risk score that, as the HHS inspector general explains, multiplies the plan’s base payment rate. A finance department’s desired result supplies none of the medical-record support a diagnosis needs, and taxpayers pay for unsupported diagnoses all along the payment trail from diagnosis to plan.

The check against the chart can come after payment. CMS’s risk adjustment data validation audits compare sampled diagnoses with supporting medical records. The plan also certifies the accuracy, completeness and truthfulness of its submitted data. That connects the chart-review operation to the people signing for its output.

Chart reviews that add or delete diagnoses without naming the encounter they belong to are what CMS’s description of the submission system calls unlinked chart review records.

Medicare Advantage Risk Score Targets Chosen by Finance

Martin’s Point’s alleged target was an average risk score of 0.92 for payment year 2017. The whistleblower’s complaint says the financial department chose it, rather than clinical personnel:

The supposed audit trigger is part of the whistleblower’s allegation about management’s intent. It gives an insider a concrete question about the number on a performance dashboard: who chose it, and what were they trying to achieve?

The complaint pairs the score with a harder-edged financial goal. Martin’s Point allegedly expected its chart-review program to recapture at least $26 million from Medicare that year. Senior managers repeatedly emphasized the goal and tied chart reviewers’ bonuses and job prospects to it. When staff raised the need to delete erroneous codes, management allegedly rejected the proposal because it would interfere with the revenue goal.

The same complaint alleges that the executives signing the data certifications also sat on the Revenue Steering Committee with the Medicare revenue operations director. On that account, the certification and the number were not made in different rooms.

When the Audit Finds the Number Is Wrong

An internal audit allegedly found errors while Martin’s Point was trying to explain a revenue shortfall. The review began because the chart-review program was behind its recapture goal. The whistleblower had pointed out a problem with the search for missing diagnoses: sometimes the earlier diagnosis was wrong.

The complaint explains why the search repeated: payment depends on conditions documented in the previous year. The program looked for conditions coded in an earlier year that had disappeared from later records. A historical cancer incorrectly coded as active could send a reviewer looking for another active-cancer code. The missing code could be a reason to correct the earlier submission, yet the revenue target rewarded finding something to add.

According to the complaint, management’s sample contained 48 already-submitted codes. Reviewers concluded that 28 were incorrect and should never have gone to CMS. The findings went to an executive meeting devoted to the chart-review program and revenue maximization. Meeting notes identified cancers coded as active when they were historical.

The whistleblower alleges that no executive at that meeting raised looking back further or correcting the errors. The conversation shifted to generating more money. To the whistleblower’s knowledge, even the specific codes identified in the sample were not deleted.

Martin’s Point later agreed to pay $22,485,000 to resolve the government’s allegations that its chart reviews produced unsupported diagnoses that it nevertheless submitted for higher payments. The settlement was smaller than the single-year recapture goal described in the complaint.

How the Incentive Reaches Physicians and Facilities

Physician and facility bonuses can carry the plan’s revenue goal into the medical record. The government alleged that the healthcare organization Kaiser Permanente set aggressive goals for adding risk-adjustment diagnoses, singled out physicians and facilities that fell behind, and emphasized that missing diagnoses cost the organization and the physicians money.

In the Justice Department’s account, Kaiser mined past medical histories and sent physicians queries urging them to add diagnoses through amendments to earlier visit records. The requests often arrived months after the visit, sometimes more than a year later. The government alleged that many added diagnoses had nothing to do with the visit and that physician and facility bonuses depended on meeting diagnosis goals.

According to the government, pressure to reach the number reached backward into a completed visit.

The government also alleged that Kaiser ignored physicians’ concerns about false claims and compliance audits identifying inappropriate amendments. Kaiser agreed to a $556 million settlement, far larger than Martin’s Point’s, over the alleged coding scheme.

A Medicare risk adjustment chart review examines existing medical records for diagnoses affecting payment, and it can identify a diagnosis missing from the submission or an unsupported diagnosis already in it.

A vendor’s contract can determine which direction the review rewards. The Group Health Cooperative complaint alleges that DxID received a percentage of the additional risk-adjustment payments it identified. Under that arrangement, an added diagnosis offered a fee. Finding a previously submitted error offered no corresponding incentive. That is the pressure built into contingency-fee chart review.

The complaint alleges that DxID did not recommend removing incorrect diagnoses originally submitted by providers. It says the vendor told the plan’s team that it neither looked for nor deleted previously submitted incorrect claims. This is how a review becomes a process that adds codes and leaves paid errors standing.

In the government’s later complaint, an Independent Health response to inferred diagnoses held that a condition easily inferred and clinically sensible should be coded, since it could be defended in an audit. That response puts the anticipated audit inside the coding decision itself.

The Medical Group Has Its Own Share

A medical group paid part of the plan’s capitation has its own financial stake in the risk score. The Seoul Medical Group complaint alleges that the group received about 40% of the payments attributable to its assigned Medicare Advantage patients. It also alleges that the group circulated lists of HCC diagnoses and instructed physicians to use them to generate higher payments.

The complaint says Health Net questioned rising risk scores. Advanced Medical Management, which reviewed the group’s charts, allegedly found unsupported diagnoses and recommended withholding part of scheduled primary-care physician bonuses to cover the expected revenue retraction. It also proposed withdrawing the unsupported codes and supplied a correction plan. The complaint says Seoul refused.

Sutter’s Correction Stopped at the Billing File

The Sutter Health complaint alleges a different way to contain an audit’s financial effect. At Palo Alto Medical Foundation, finance questioned why auditors were revisiting risk-adjustment data for which the organization had already been paid while its RAF scores were falling. The auditor was told to stop submitting corrections to the medical records.

A later instruction allowed corrections only on the billing side of EPIC. Physicians alone could amend the patient encounter. The whistleblower says the instruction would leave the payment problem in place: the diagnoses sent onward to the plan and CMS came from the patient’s medical record, not the billing file.

The audit was working. Its working was the complaint against it. In the whistleblower’s account, management knew that a billing-file correction would leave the inaccurate payment data untouched. That is why what happened after the internal warning matters as much as the error list.

The 2015 paper Upcoding or Selection? Evidence from Medicare on Squishy Risk Adjustment, hosted by Duke Economics, found coding intensity rose with closer insurer-provider integration and described physician-group meetings that displayed individual physicians’ scores, the pressure a target puts on the doctor. A 2019 study in Health Management, Policy and Innovation found Medicare Advantage patients looked substantially less risky under a diagnosis measure that plays no part in payment, the pattern behind coding intensity in Medicare Advantage.

That comparison makes the incentive visible. The case documents take the inquiry inside the business, to the target, the requested diagnosis and the response when somebody found an error.

Red Flags in Risk-Adjustment Coding

In risk-adjustment coding, the red flags for Medicare fraud come together as pressure to increase scores, unsupported diagnoses and a decision to keep them in submitted data. A target document can show the expected result. A bonus plan or vendor contract can show who benefits. The chart, audit findings and correction instructions can show how the organization responded when the medical evidence got in the way.

Those records connect different jobs inside the same operation. A physician may know why a requested diagnosis did not belong to the visit. A reviewer may know which errors were left in. Finance or data staff may know whether the correction reached the submission that determined payment. Together, those views can explain who knew, what they were told and what the company continued to bill.

A whistleblower case under the False Claims Act can pay 15% to 30% of what the government recovers.

The Target Is Only the Beginning

If you know who set the coding goal and what happened when the charts contradicted it, talk with our whistleblower experts. We connect the clinical evidence, the payment decisions and the people who made them to build your strongest Medicare Advantage whistleblower case, bringing in Medicare Advantage whistleblower lawyers who understand risk-adjustment coding. We stay on your side through the whole process.