A new study led by Dr. Richard Kronick at the University of California San Diego reveals that differences in diagnostic coding practices between Medicare Advantage (MA) plans and Traditional Medicare (TM) resulted in an estimated $33 billion in excess payments to private insurers in 2021 alone. The findings, published in the Annals of Internal Medicine, suggest that aggressive coding practices by MA plans inflate patients’ risk profiles, triggering higher reimbursements from the federal government—even when clinical severity remains unchanged.

The Study

The study used a retrospective analysis of data from the Centers for Medicare & Medicaid Services (CMS), including over 697 active MA contracts spanning 2015 to 2021. Researchers examined two key indicators: “persistence” (the rate at which diagnoses continue from one year to the next) and “new incidence” (the rate of new diagnoses year-over-year). By comparing these patterns between MA and TM enrollees, the team isolated the extent to which coding intensity alone—rather than actual changes in health status—contributed to revenue differences.

Troubling Findings

Findings showed that in 2021, MA plans had risk scores 18.5% higher on average than those in TM. These scores, which are used to calculate plan reimbursements, reflect how sick a patient appears on paper. The persistence of diagnoses was notably higher in MA (78.1%) versus TM (72%), and new diagnosis rates were also elevated (46% in MA vs. 33% in TM). This inflation of risk scores resulted in billions in extra federal payments—with UnitedHealth Group alone receiving $13.9 billion, or $1,863 in additional revenue per member, far exceeding the MA plan average of $1,220.

While the study does not allege fraud, it underscores a critical vulnerability in the MA reimbursement structure. As the authors note, differential coding may distort payment fairness and exacerbate spending without improving patient outcomes. Moreover, the variation in coding practices across insurers suggests that any reform to MA payment policy could have uneven impacts depending on how intensively each insurer codes.

Where to Go From Here?

The authors emphasize the need for targeted policy reforms to correct these imbalances, especially as MA enrollment continues to grow. Without adjustments, differential coding could continue to divert federal dollars toward inflated reimbursements instead of actual patient care. The study did not receive any commercial funding, and no financial conflicts of interest were disclosed by the research team.

Source: Annals of Internal Medicine, April 2025 Edition
Corresponding Author: Dr. Richard Kronick, UC San Diego
Media Contact: Yadira Galindo | y2galindo2@ucsd.edu

For full study access: https://www.acpjournals.org/doi/10.7326/ANNALS-24-01345