Bianca K Frogner1, Gerard F Anderson, Robb A Cohen, Chad Abrams. 1. Department of Health Services Management and Leadership, School of Public Health and Health Services, The George Washington University, Washington, DC 20037, USA. bfrogner@gwu.edu
Abstract
BACKGROUND: The Medicare Advantage payment system underpays health plans that enroll beneficiaries with multiple and complex chronic conditions. OBJECTIVES: This article addresses 3 major problems in the current payment system: (1) underreporting of chronic disease prevalence in fee-for-service (FFS) Medicare claims data, (2) overpayment of healthier and underpayment of sicker beneficiaries in the current payment system, and (3) underpayment for new beneficiaries in Medicare Advantage plans that require the beneficiaries to have at least one chronic disease to enroll. RESEARCH DESIGN: We incorporate 2 years of data and a count of chronic diseases in the current Medicare payment model. We develop a separate payment adjustment for new enrollees. SUBJECTS: A nationally representative sample of FFS beneficiaries in the 2004-2006 Medicare 5% claims data. MEASURES: We use predictive ratios to evaluate whether our enhanced model improves the predictive accuracy over the current model overall and for subsets of beneficiaries. RESULTS: The underreporting of chronic disease prevalence in Medicare FFS by 20% leads to systematic bias in the disease coefficients and demographic adjusters. The enhanced model reduces the level of payment for healthy beneficiaries and increases the payment for beneficiaries with multiple and complex chronic conditions. It improves payment for plans that enroll new enrollees with specific chronic conditions. CONCLUSIONS: Our enhanced model reduces financial incentives for health plans to engage in risk selection against beneficiaries with multiple chronic conditions.
BACKGROUND: The Medicare Advantage payment system underpays health plans that enroll beneficiaries with multiple and complex chronic conditions. OBJECTIVES: This article addresses 3 major problems in the current payment system: (1) underreporting of chronic disease prevalence in fee-for-service (FFS) Medicare claims data, (2) overpayment of healthier and underpayment of sicker beneficiaries in the current payment system, and (3) underpayment for new beneficiaries in Medicare Advantage plans that require the beneficiaries to have at least one chronic disease to enroll. RESEARCH DESIGN: We incorporate 2 years of data and a count of chronic diseases in the current Medicare payment model. We develop a separate payment adjustment for new enrollees. SUBJECTS: A nationally representative sample of FFS beneficiaries in the 2004-2006 Medicare 5% claims data. MEASURES: We use predictive ratios to evaluate whether our enhanced model improves the predictive accuracy over the current model overall and for subsets of beneficiaries. RESULTS: The underreporting of chronic disease prevalence in Medicare FFS by 20% leads to systematic bias in the disease coefficients and demographic adjusters. The enhanced model reduces the level of payment for healthy beneficiaries and increases the payment for beneficiaries with multiple and complex chronic conditions. It improves payment for plans that enroll new enrollees with specific chronic conditions. CONCLUSIONS: Our enhanced model reduces financial incentives for health plans to engage in risk selection against beneficiaries with multiple chronic conditions.
Authors: Bruce Friedman; Peter J Veazie; Benjamin P Chapman; Willard G Manning; Paul R Duberstein Journal: Milbank Q Date: 2013-09 Impact factor: 4.911
Authors: Jennifer Hornung Garvin; Andrew Redd; Dan Bolton; Pauline Graham; Dominic Roche; Peter Groeneveld; Molly Leecaster; Shuying Shen; Mark G Weiner Journal: Perspect Health Inf Manag Date: 2013-10-01
Authors: Juan F Orueta; Roberto Nuño-Solinis; Maider Mateos; Itziar Vergara; Gonzalo Grandes; Santiago Esnaola Journal: BMC Health Serv Res Date: 2013-07-09 Impact factor: 2.655
Authors: Juan F Orueta; Roberto Nuño-Solinís; Arturo García-Alvarez; Edurne Alonso-Morán Journal: BMC Public Health Date: 2013-10-03 Impact factor: 3.295