Alternatives for using multivariate regression to adjust prospective payment rates

Area of Science:

• Health Economics
• Econometrics
• Healthcare Policy

Background:

• Medicare's prospective payment system relies on adjustments for accurate reimbursement.
• Key adjustments include indirect-teaching, disproportionate-share, and large-city status.
• These adjustments significantly impact annual Medicare payment distribution ($3 billion).

Purpose of the Study:

• To analyze the application of multivariate regression in Medicare payment adjustments.
• To demonstrate the impact of different regression model specifications on these adjustments.

Main Methods:

• Multivariate regression analysis was employed.
• Specific focus on indirect-teaching, disproportionate-share, and large-city adjustments.
• Comparison of outcomes based on varying regression model specifications.

Main Results:

• Regression model specification critically influences Medicare payment adjustments.
• Different specifications lead to varying distributions of approximately $3 billion in annual payments.
• The study demonstrates the tangible financial implications of regression modeling choices.

Conclusions:

• The choice of regression models is crucial for equitable Medicare payment distribution.
• Understanding the impact of model specification is essential for policymakers.
• Accurate regression modeling ensures fairness in healthcare reimbursements.