The Importance of Diagnosis Coding for Medicare Advantage

Clinical documentation integrity (CDI) professionals work in a variety of settings, and although I mostly focus on topics related to hospital inpatient billing, this week I want to focus on diagnosis coding related to risk adjustment within Medicare Advantage (MA) plan payments.

Enrollment in MA has surpassed that of traditional Medicare, and it is now the dominant healthcare delivery model for those eligible for Medicare benefits. The payment methodology for Medicare Part A and Medicare Part B is guided by the Inpatient Prospective Payment System (IPPS). However, Medicare Part C (MA Organizations) is governed by a different payment methodology.

Instead of paying healthcare providers directly, under Medicare Part C, monthly payments are made to private healthcare companies that provide insurance coverage for Medicare beneficiaries who enroll in their plan.

MA plans make bids to Medicare that represent the expected average cost per standard beneficiary, administrative costs, and a profit margin. The standard bid is compared to a benchmark: expected costs adjusted for local markets under statutory formulas, based on average spending on traditional Medicare beneficiaries. Adjustments are also made for a MA plan’s rating in the star system, which measures the plan’s quality of care.

This benchmark establishes the monthly base payment per beneficiary to the MA plan. Because it is an average cost, adjustments are made to the base payment rate based on each enrolled beneficiary’s demographic and health risk characteristics. These risk characteristics include age, disability, dual eligibility, institutional status, prior health conditions, and risk adjustments using Centers for Medicare & Medicaid Services (CMS) Hierarchical Condition Categories (CMS-HCCs).

Most diagnoses used in the risk adjustment calculations are reported on physician and outpatient claims. CMS-HCCs led to the development of the risk-adjustment coding subspecialty. Prior to the emphasis on risk-adjustment coding, there was minimum scrutiny regarding the diagnoses included on a professional or hospital outpatient claim, since reimbursement was based on the assignment of Current Procedural Terminology ® (CPT) codes.

Diagnoses were used to demonstrate a covered benefit, medical necessity, and elements required for the proper assignment of evaluation and management (E&M) codes. It has also contributed to outpatient CDI efforts, especially for health systems that offer MA plans.

Not surprisingly, a study by the Medicare Payment Advisory Commission (MedPac) found that MA beneficiaries have higher risk scores, on average, compared to similar traditional Medicare beneficiaries, due to more exhaustive reporting of diagnosis codes.

More diagnosis codes can lead to higher risk-adjustment payments. The MedPac 2025 Report to Congress estimates that Medicare will spend 20 percent ($84 billion) more for MA plan beneficiaries than if those enrollees were receiving care under Medicare Part A and Medicare Part B (traditional Medicare). The report concludes that favorable selection and coding intensity are primarily responsible for these higher payments:

• “Favorable selection” occurs when beneficiaries with lower actual spending relative to their risk score enroll in MA, which results in spending lower that the traditional Medicare average.
• “Coding intensity” refers to the tendency for more diagnosis codes to be recorded for MA enrollees, which causes risk scores – and payments – for the same beneficiaries to be higher when they are enrolled in MA than they would be in traditional Medicare.

MedPac concluded that “MA plans have a financial incentive to ensure that their providers record all possible diagnoses because adding new risk-adjustment-eligible diagnoses raises an enrollee’s risk score and results in higher payments to the plan.” This conclusion is based on estimates wherein MA risk scores were about 17-percent percent higher than similar traditional Medicare beneficiaries due to higher coding intensity in 2023.

Investigations by entities that perform oversight of Medicare have revealed questionable diagnosis collection mechanisms that deviate from ICD-10-CM Official Guidelines for Coding and Reporting. Section IV, Diagnostic Coding and Reporting Guidelines for Outpatient Services, states that “Guidelines in Section I, Conventions, general coding guidelines and chapter-specific guidelines, should also be applied for outpatient services and office visits.” Outpatient coders are also instructed to “list additional codes that describe any coexisting conditions.” Additional outpatient coding guidance instructs:

• Chronic diseases treated on an ongoing basis may be coded and reported as many times as the patient receives treatment and care for the condition(s);
• Code all documented conditions that coexist at the time of the counter/visit and that require or affect patient care, treatment, or management;
• Do not code conditions that were previously treated and no longer exist; and
• History codes may be used as secondary codes if the historical condition or family history has an impact on current care or influences treatment.

Questionable collection mechanisms used by some MA plans include:

• Chart reviews (which document diagnoses not captured through the usual means of reporting diagnoses);
• Health risk assessments (which sometimes rely on unverified enrollee-reported data); and
• Potentially fraudulent diagnosis data.

Coding intensity has subsequently become the subject of the occasional U.S. Department of Health and Human Services (HHS) Office of Inspector General (OIG) investigation as well. The OIG investigations resulted in a toolkit to help decrease improper payments to MA plans. OIG audits through 2023, when the toolkit was published, found that approximately 70 percent of diagnosis codes reported by MA plans were not supported by the associated medical records.

Additionally, some diagnosis codes were unsupported over 90 percent of the time! Some of the lowlights included the following:

• Acute stroke: 96-percent error rate;
• Acute heart attack: 95-percent error rate;
• Breast cancer: 96-percent error rate; and
• Colon cancer: 94-percent error rate.

As a seasoned CDI professional, I am not surprised by these codes having such high error rates, since they are diagnoses for which documentation must differentiate between a current, acute condition and a historic condition. It is also likely, in my opinion, that many of these coding errors can be attributed to data collected from claims for which providers code their own records, rather than relying upon a professional coder who understands the nuances associated with proper code assignment.

Next week I will dig deeper into this topic by exploring Risk Adjustment Data Validation (RADV) audits and their findings.

Programming note:

Listen to Cheryl Ericson report this story live during Talk Ten Tuesday on November 4,10 Eastern with Chuck Buck.