A Reference-Invariant Health Disparity Index Based on Rényi Divergence

• Activity Number: 666
• Type: Contributed
• Date/Time: Thursday, August 2, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Makram Talih*+
• Companies: National Center for Health Statistics
• Keywords: Epidemiological Methods ; Health Inequalities ; Generalized entropy class ; Alpha-gamma divergence ; Survey Data ; Taylor Series Linearization

Abstract

One of four overarching goals of Healthy People 2020 is to achieve health equity, eliminate disparities, and improve the health of all groups. In health disparity indices (HDIs) such as the mean log deviation (MLD) and Theil index (TI), disparities are relative to the population average, whereas in the index of disparity (IDisp) the reference is the group with the least adverse health outcome. While the latter may be preferable, reference group selection is affected by statistical reliability. To address this issue, we propose a new HDI, the Rényi index (RI), which is reference-invariant. The RI is related to the Atkinson index, which is parameterized by a disparity aversion parameter. Also, the MLD and TI are limiting cases of the RI. When groups are population-weighted, the MLD and TI are decomposable; similarly for the RI, even when groups are equally-weighted (like in the IDisp). We derive design-based standard errors for the RI and symmetrized RI (SRI), and for their within- and between-group components. Using data from the National Health and Nutrition Examination Survey, we examine how varying the disparity aversion parameter or the group weighting scheme affects the SRI.