Coding-Driven Changes In Measured Risk In Accountable Care Organizations.

Area of Science:

• Health economics
• Healthcare policy
• Health services research

Background:

• Risk adjustment in healthcare payments relies heavily on claims data.
• Potential for claims data to reflect coding practices rather than true health status changes.
• Medicare's Accountable Care Organization (ACO) program utilizes risk adjustment for payment.

Purpose of the Study:

• To quantify the divergence between claims-based (HCC) and survey-based (CAHPS) risk scores in Medicare ACOs.
• To assess whether observed risk score growth is attributable to coding practices or changes in beneficiary health.
• To evaluate the impact of different risk adjustment methodologies on healthcare payments.

Main Methods:

• Comparison of risk scores derived from Centers for Medicare and Medicaid Services Hierarchical Condition Categories (HCC) and Consumer Assessment of Healthcare Providers and Systems (CAHPS) data.
• Analysis of risk score growth trends between 2013-2016 within Medicare ACOs.
• Examination of the variability in the gap between HCC- and CAHPS-based risk score growth across ACOs.

Main Results:

• HCC-based risk scores demonstrated significantly faster annual growth (2.1%) compared to CAHPS-based risk scores (0.3%) during 2013-2016.
• A wide variation in the gap of risk score growth was observed across different ACOs.
• The primary driver for the observed risk score growth gap appears to be HCC coding practices, not necessarily ACO performance or CAHPS model limitations.

Conclusions:

• Claims coding practices, rather than true health status changes, likely account for the majority of risk score growth in Medicare ACO beneficiaries.
• The findings suggest potential distortions in risk adjustment payments due to coding intensity.
• Further investigation into coding practices and their impact on healthcare payment models is warranted.