Development and Comparative Predictive Validity of an Outpatient Medication Exposure Measure for Risk Adjustment Using Retrospective Claims Data

The purpose of this study was to develop and test a measure of outpatient prescription utilization (medication exposure measure, or MEM) that may be coupled with the CMS-HCC and CMS-RxHCC methodologies to improve risk-adjusted payments to Medicare Part C and Part D plans. Studies have identified prescription measures that predict future expenditures; however, many are easily manipulable by health plans or practitioners, thus limiting their utility as risk-adjusters. The addition of a non-manipulable prescription utilization measure to existing risk-adjustment models may improve prediction, reducing adverse risk selection incentives by health plans. A secondary objective of this study was to evaluate the utility of adding prescription measures to the Charlson's Comorbidity Index, Elixhauser's Index, and RxRisk to predict year-2 expenditures, hospitalization counts, emergency department visits, and mortality. Study Design: The study design utilized a retrospective cohort from the 5% Medicare national sample, which used year-1 (2007) inputs to predict the year-2 (2008) economic and clinical outcomes. The sample included beneficiaries with continuous enrollment in fee-for-service Medicare Parts A, B, and D for a minimum of 12 months in the base year and a minimum of 1 month in year-2. An interaction between the end-of-year Medicare Part D benefit phase and the prescription measures was included to account for the influence of the coverage gap (i.e., "donut hole") on the prescription measures. Results: Overall, the addition of the prescription-based measures to risk-adjustment models resulted in enhanced predictive validity for the economic and clinical outcomes tested compared to the risk-adjustment model alone. The addition of any prescription measure to the risk-adjustment models did not meaningfully improve model performance in predicting year-2 medical expenditures; however, the prescription measures, particularly the MEM, markedly improved prediction of year-2 pharmacy expenditures. Conclusions: Although adding MEM to the HCC models used to predict medical expenditures does not appear to be a useful method of enhancing risk-adjusted payments, the MEM performed particularly well with the RxHCC predicting year-2 pharmacy expenditures. Incorporating the MEM into Medicare Part D risk-adjustment models (i.e., with RxHCC) would improve risk-adjusted capitated payments from both the perspectives of CMS and the health plans.

Name thesis:degree_name

M.S. in Pharmaceutical Science

Level thesis:degree_level

Thesis

Year dc:date.available

2012

Author dc:creator

• Null, Kyle Dennis

Contributors dc:contributor

• John P. Bentley
• Jeffrey Hallam
• Yi Yang

• Capitation
• Hcc
• Health Care Financing
• Medicare
• Risk-Adjustment
• Rxhcc
• Pharmacy and Pharmaceutical Sciences

Repository record dc:identifier

https://egrove.olemiss.edu/etd/211

OAI identifier oai:identifier

oai:egrove.olemiss.edu:etd-1210