Unlinked Chart Review Records and Audio-Only Diagnoses: Which Diagnoses Count for Medicare Advantage in 2027?

The Centers for Medicare & Medicaid Services will exclude diagnoses from unlinked chart review records from Medicare Advantage risk scores for 2027, with an exception for beneficiaries switching Medicare Advantage parent organizations. The agency is also excluding specified audio-only diagnoses. For a coder, reviewer or plan employee, the practical question is which service stands behind the diagnosis and whether its record qualifies for payment. The chart-review result, the encounter and the payment data have to tell the same story.

What Are Unlinked Chart Review Records?

Unlinked chart review records report diagnoses without identifying the previously submitted encounter data record to which they belong. The agency therefore cannot tell from the review record which healthcare items or services generated those diagnoses. A linked chart review supplies that connection to the encounter.

A retrospective chart review looks through existing medical records for diagnoses. Its findings can support additions or deletions to the information used for risk adjustment. In its 2019 report prepared for CMS, RAND makes a distinction that matters to anyone counting visits or reconciling a vendor’s output:

The review reports what someone found in the chart. The encounter records the service provided to the patient. A second look at a chart can change the payment data without another visit taking place.

That matters because a diagnosis can change the plan’s payment. Eligible diagnoses feed into hierarchical condition categories, and those categories and demographic factors determine the enrollee’s risk score. Medicare multiplies the plan’s base payment by that score. The financial effect reaches beyond the bill for the visit that produced the diagnosis.

What CMS Finalized for 2027

The Centers for Medicare & Medicaid Services finalized the diagnosis-source exclusions in its April 6, 2026 Rate Announcement. For Medicare Advantage organizations other than PACE, diagnoses submitted on unlinked chart reviews leave the risk-score calculation unless the switching exception applies. The audio-only exclusion depends on the services recorded alongside the diagnosis.

The agency kept its 2024 risk adjustment model for these plans instead of recalibrating it. It still changed which diagnosis sources enter the score. The unlinked chart review exclusion also applies to Part D risk scores, with the switching exception.

Plans can continue submitting encounter data records and linked chart reviews. They can also submit unlinked chart reviews, even when the diagnoses on them will be excluded from the score. PACE organizations retain the use of unlinked chart-review diagnoses for 2027.

Health care services that generate no traditional claim still belong in the encounter data. A later chart review can attach additional diagnoses to that service record.

Final risk-adjusted payments for 2027 use diagnoses from 2026 dates of service. The rule reaches the records being created before the payment year begins.

Why the Switching-Plan Exception Matters

The exception applies when the beneficiary was enrolled with a different Medicare Advantage parent organization in the prior year. The agency expressly excludes a beneficiary coming from fee-for-service Medicare from this exception. The diagnoses must still meet the other risk-adjustment criteria.

The Centers for Medicare & Medicaid Services gives an operational reason. A receiving plan can face constraints linking diagnoses to the prior year’s encounter data after the beneficiary changes organizations. The exception preserves a route for those diagnoses to enter the score despite that linkage problem.

Audio-Only Diagnoses Depend on the Other Service Lines

The Centers for Medicare & Medicaid Services excludes diagnoses from audio-only services identified by modifiers 93 or FQ when no other line on the encounter data record, chart review record or fee-for-service claim is eligible for risk adjustment. A record containing another qualifying service receives different treatment from a record containing only audio-only service lines.

The agency describes this as an operational update to its longstanding face-to-face requirement. Real-time, interactive audio and video can meet that requirement when the visit satisfies the other criteria. A diagnosis from a qualifying face-to-face visit during the applicable data collection period can be considered for risk adjustment.

Why CMS Wants the Service Behind the Diagnosis

The final announcement reports that approximately 85 percent of the 88.8 million unlinked chart review records submitted in 2023 for 2024 payment could not be associated with a submitted encounter. The diagnoses arrived, but the matching service records largely did not.

The older payment studies show why that missing connection matters. In its 2019 chart-review report, the Department of Health and Human Services’ Office of Inspector General found that over 99 percent of the chart reviews it examined added diagnoses. Diagnoses appearing only on chart reviews and on no service records generated an estimated $6.7 billion in risk-adjusted payments for 2017.

A separate October 2024 OIG report estimated $7.5 billion in 2023 payments from diagnoses appearing only on health risk assessments or assessment-linked chart reviews, with no other service record in the preceding year. Almost two-thirds came from in-home assessments and their linked reviews. That study measured the assessment channel, including the work behind home-visit diagnosis allegations.

The inspector general identified two concerns: inaccurate diagnoses producing improper payments, or beneficiaries missing needed care for serious conditions. Taxpayers can pay more for a diagnosis while the patient goes without the care it calls for.

The coding intensity adjustment addresses differences in coding patterns across Medicare Advantage and fee-for-service Medicare. The new source rule gives CMS a more specific connection between a diagnosis and a service.

When a Review Finds an Unsupported Diagnosis

The Medicare Advantage fraud question is what the organization learned about a diagnosis and what it did with that information. One-way chart review allegations concern programs that use findings to add revenue while leaving identified errors in the payment data.

The government’s complaint against Anthem alleged that its review program sought additional codes while disregarding findings that failed to substantiate codes already submitted. Its instruction to its vendor, as the complaint describes it, was explicit: Medi-Connect was to direct its review and coding toward new codes that would earn Anthem more money.

The same complaint alleged that Anthem had no process to compare previously submitted diagnoses against the vendor’s review results for the same patients and visits, even though its programmers could build that comparison. The alleged failure sat between finding the codes and acting on what the review revealed.

The HealthCare Partners settlement announcement described a parallel allegation: the company searched for missed diagnoses while ignoring inaccurate codes that should have been deleted. A whistleblower had brought those one-way review allegations to the government. A whistleblower case under the False Claims Act can pay 15% to 30% of what the government recovers.

These examples focus a risk-adjustment fraud inquiry on the clinical support, the review findings and the company’s response. The new payment rule addresses which sources CMS uses. The historical cases turn on what the organizations allegedly submitted or left uncorrected after examining the charts.

What the Coder and the Payment Team Each See

A coder may see a diagnosis rejected by review. A vendor may see instructions to send additions. The payment team may see which diagnosis codes actually reached CMS. Connecting those views is how we examine the payment trail with an insider.

For this policy, the useful connections include the review record and its encounter, the services shown on an audio-only record, and the enrollment history behind a switching exception. For an unsupported diagnosis, the connection extends to the correction and its effect on payment. A finding in a review file and a correction in the payment data can be different things.

The CMS deadline memo makes the direction of correction consequential. Organizations must update and correct data before and after the final deadline, but CMS processes only deletions submitted after that deadline in later risk-score runs. The submission window can close to additional payments while corrections remain due.

When You Can See the Missing Connection

If you know why a diagnosis reached the payment file despite a contrary review finding, speak with our whistleblower experts about the records and the decisions behind it. We stay beside you through the case and bring in Medicare Advantage whistleblower lawyers who understand chart-review and risk-adjustment cases to build your strongest case from the clinical evidence, the submissions and the money.