Extension of the Peters-Belson Method to Estimate Health Disparities Among Multiple Groups Using Logistic Regression with Survey Data

• Activity Number: 492
• Type: Contributed
• Date/Time: Wednesday, August 3, 2016 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Yan Li* and Barry Graubard and Pengyu Huang and Joseph L. Gastwirth
• Companies: Joint Program in Survey Methodology and National Cancer Institute and Fors Marsh Group and The George Washington University
• Keywords: Taylor linearization ; complex survey data ; multinomial logistic regression ; proportional odds logistic ; unexplained disparity

Abstract

Determining the extent of a disparity, if any, between groups of people, is of interest in many fields, including public health for medical treatment and prevention of disease. An observed difference in the mean outcome between an advantaged group (AG) and disadvantaged group (DG) can be due to differences in the distribution of relevant covariates. The Peters-Belson (PB) method fits a regression model with covariates to the AG t0 predict, for each DG member, their outcome measure as if they had been from the AG. The difference between the mean predicted and observed outcomes of DG members is the (unexplained) disparity of interest. We focus on applying the PB method to estimate the disparity based on logistic regression models using data collected from complex surveys with multiple DGs. Estimators of the unexplained disparity, an analytic variance estimator that is based on the Taylor linearization variance method, as well as a Wald test for testing a joint null hypothesis of zero for unexplained disparities between multiple minority groups and a majority group, are provided. Simulation studies and analyses of disparity in BMI in NHANES 1999-2004 are conducted.