Measuring coding intensity in the Medicare Advantage program.

Area of Science:

• Health Economics
• Health Policy
• Medical Informatics

Background:

• Medicare implemented a payment system in 2004 incentivizing Medicare Advantage (MA) plans to report diagnoses more than fee-for-service (FFS) providers.
• This created a disparity in diagnosis reporting between MA and FFS.
• Understanding this disparity is crucial for accurate healthcare resource allocation.

Purpose of the Study:

• To analyze the trends in risk scores for Medicare beneficiaries enrolled in MA and FFS plans.
• To determine if the increasing risk scores in MA reflect actual morbidity or coding intensity.
• To evaluate the effectiveness of Medicare's policy changes in mitigating coding intensity.

Main Methods:

• Utilized risk scores for all Medicare beneficiaries from 2004-2013 and Medicare Current Beneficiary Survey (MCBS) data from 2006-2011.
• Analyzed changes in average risk scores for all enrollees and for stayers (those in MA or FFS for two consecutive years).
• Examined prevalence rates by Hierarchical Condition Category (HCC) to identify coding patterns.

Main Results:

• Average MA risk scores increased faster than FFS scores annually, suggesting coding intensity.
• Under the 2004 model, the MA-to-FFS risk score ratio rose from 90% (2004) to 109% (2013).
• Survey data (2006-2011) showed a stable MA-FFS risk score ratio around 96%, indicating coding practices, not increased morbidity, drive MA score increases.

Conclusions:

• Medicare has implemented adjustments (e.g., 3.4% coding intensity adjustment in 2010, model revisions in 2013/2014) to address coding intensity.
• Despite these measures, the relative increase in MA risk scores persists.
• Further policy interventions are likely necessary to ensure accurate risk adjustment and equitable payments.