兰德-Specialty-Payment-Model-Opportunities-and-Assessment_-Oncology-Simulation-Report_43页_1mb

Summary of "Specialty Payment Model Opportunities and Assessment: Oncology Simulation Report"

Core Content

This report presents the results of a simulation analysis of a proposed payment model for specialty oncology services, specifically chemotherapy care, developed by the Centers for Medicare & Medicaid Services (CMS) through the CMS Alliance to Modernize Healthcare (CAMH) Federally Funded Research and Development Center (FFRDC), operated by MITRE, and conducted by RAND Health. The goal of the simulation is to assess the potential impacts of the model and inform its design for possible testing by the Center for Medicare and Medicaid Innovation.

Main Objectives

• To evaluate the financial and behavioral impacts of a specialty payment model for oncology services.
• To simulate the effects of care management and performance-based payments on Medicare spending and oncology practice revenues.
• To explore the implications of different model design features and participation assumptions.

Key Points

1. Payment Model Overview

• The Oncology Care Model (OCM) is a proposed episode-based payment model designed to incentivize high-quality, low-cost care.
• Practices are eligible for:
  • Per-beneficiary per-month (PBPM) care management payments of $160 per month for each chemotherapy episode.
  • Performance-based payments based on per-episode spending relative to a target, with two options:
    • One-sided: Practices receive payments if actual spending is below the target.
    • Two-sided: Practices face upside and downside financial risk.
• The model includes a ceiling on performance-based payments (up to 20% of the spending target) and limits on amounts owed to CMS in the two-sided option.

2. Simulation Methodology

• The simulation uses a baseline dataset of chemotherapy episodes for 2016, derived from 2010 Medicare FFS claims.
• The dataset includes 8 prevalent cancer types and was inflated to 2016 using category-specific inflation factors.
• Only practices with 50 or more chemotherapy episodes per year were assumed eligible for participation.
• 10% of eligible practices were assumed to participate in the model.

3. Behavioral Responses and Model Impacts

• The care management payments of $160 per month per episode are relatively large compared to FFS revenues, which are around $2,000 per episode.
• These payments could increase Medicare revenues by about 50% on average for participating practices.
• Performance-based payments depend on accurate spending targets. However, due to unexplained variation in spending (sampling error), the model may result in "noise bonuses" if the targets are not well calibrated.
• To break even, practices would need to reduce utilization and intensity by about 4%.
• This behavioral response is possible but uncertain, based on research on similar models.
• The break-even point could be reduced by:
  • Lowering the PBPM care management payments.
  • Lowering the performance-based payment incentives.

4. Data and Methods

• The baseline dataset includes:
  • Demographic and location data for beneficiaries.
  • Cancer type based on ICD-9 codes.
  • Actual spending per episode.
  • Attributed practice information.
• The dataset was inflated to 2016 using Producer Price Indices.
• Spending was categorized into 11 types, with different shares treated as revenue to the attributed practice (see Table 2.1).

5. Episode-Level Spending

• The mean spending per chemotherapy episode in 2016 is estimated at $26,641.
• Prostate cancer has the lowest mean spending ($12,968) and lowest physician practice revenue ($353).
• Breast cancer and colorectal cancer have the highest mean spending and revenue.
• Over 70% of practices with attributed episodes had fewer than five episodes in 2010, indicating a significant portion of practices may not be true oncology practices.

Key Findings

• The care management payments represent a substantial new revenue source for oncology practices and a substantial new cost for Medicare.
• The performance-based payment model may result in noise bonuses due to inaccurate benchmarks.
• A 4% reduction in utilization and intensity is required for Medicare to break even.
• Selective participation (e.g., only high-volume practices) may reduce the risk of noise bonuses.
• The simulation model does not include quality measures, and the impact of quality on spending is not evaluated.

Conclusion

The simulation analysis highlights the potential for the OCM to change the financial landscape for oncology practices and Medicare. While the model could lead to significant savings if practices respond as expected, the risk of noise bonuses and the uncertainty of behavioral responses suggest that careful design and benchmarking are essential. The findings also emphasize the importance of practice volume in determining eligibility and the impact of sampling error on the accuracy of performance metrics.

Key Tables and Figures

• Table 2.1: Spending categories and their inflation factors, revenue shares, and mean spending per episode in 2016.
• Table 2.2: Cancer types, number of episodes, mean spending, and mean physician practice revenue in 2016.
• Table 2.3: Distribution of oncology practice volume.
• Figure 3.1: Simulated errors in spending targets and their implications for bonuses in the one-sided option.
• Figure 5.1: Medicare spending on chemotherapy episodes by participating practices.
• Figure 5.2: Simulated Medicare revenues at participating practices.
• Figure 5.3: Distribution of simulated changes in practice-level revenues.
• Figure A.1: Simulated participation probability.

References and Acknowledgments

• The research was conducted by MITRE and RAND Health.
• The Model Design Report (Huckfeldt et al., 2014) provided foundational insights.
• The authors acknowledge the contributions of Paul B. Ginsburg, Sarah Nowak, and other experts from CMS and the Brookings Institution.

Abbreviations

• ACE: Acute Care Episode
• ACO: Accountable Care Organization
• BPCI: Bundled Payment for Care Improvement
• CCW: Chronic Conditions Warehouse
• CMS: Centers for Medicare & Medicaid Services
• DME: Durable Medical Equipment
• FFS: Fee-for-Service
• MedPAR: Medicare Provider Analysis and Review
• OCM: Oncology Care Model
• PBPM: Per-Beneficiary Per-Month
• PCMH: Patient-Centered Medical Home
• TIN: Tax Identification Number