Association Study of Health Outcomes Based on Combined Data from Primary and Ancillary Surveys

• Activity Number: 257
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Tianle Hu*+ and Trivellore Raghunathan and Ana Diez-Roux
• Companies: University of Michigan and University of Michigan and University of Michigan
• Keywords: maximum likelihood ; ancillary survey ; social epidemiology

Abstract

Traditional epidemiological studies focus on the association between health status of individuals and their risk factors. More recently, however, there has been increasing interest in the effect of neighborhood level characteristics on the individual health outcome. In some epidemiological studies, researchers have begun collecting data to measure specific features of neighborhoods using ancillary surveys, conducted on a separate set of individuals. The challenge is then to combine data from the main survey and ancillary surveys, so that maximal information can be extracted to assess the role of neighborhood-level factors. We propose a hierarchical model for combining data from the different sources using a likelihood-based approach. The simulation study shows that the proposed approach yields estimates with good sampling properties. We also conduct the analysis for a real dataset.