Upcoding: Evidence from Medicare on Squishy Risk Adjustment.

Area of Science:

• Health Economics
• Health Insurance Policy
• Medical Billing and Coding

Background:

• Health insurance plans in the US often face incentives to inflate patient diagnoses reported to regulators.
• These reported diagnoses directly influence the risk-adjusted payments plans receive, creating a potential for financial gain through upcoding.

Purpose of the Study:

• To quantify the extent of diagnosis-based risk score inflation in private Medicare plans compared to traditional fee-for-service Medicare.
• To estimate the financial and behavioral consequences of this upcoding practice.

Main Methods:

• Comparative analysis of diagnosis-based risk scores between private Medicare enrollees and fee-for-service Medicare beneficiaries.
• Econometric modeling to estimate excess public spending and behavioral distortions.

Main Results:

• Enrollees in private Medicare plans exhibit 6% to 16% higher diagnosis-based risk scores than under fee-for-service Medicare.
• Upcoding is estimated to generate billions in excess public spending.
• Coding intensity correlates positively with vertical integration, suggesting insurer-provider principal-agent problems.

Conclusions:

• Diagnosis upcoding in private Medicare plans inflates risk scores, leading to substantial financial waste and market distortions.
• Vertical integration may exacerbate upcoding, indicating a need for regulatory oversight and aligned incentives within health insurance markets.