Online Program Home
My Program

Abstract Details

Activity Number: 349 - Longitudinal, Spatial, and Bayesian Methods
Type: Contributed
Date/Time: Tuesday, July 31, 2018 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #328839
Title: Estimating Small Area Life Expectancy Using Hellinger Distance Spatial Correlation
Author(s): Robert E. Johnson* and Sarah Lotspeich
Companies: Vanderbilt University and Vanderbilt University
Keywords: similarity measure; age-specific mortality; multidimensional scaling
Abstract:

With recognition that health outcomes vary across smaller geographic areas and are shaped by the social and physical environment, interest has mounted in measuring disparities with greater granularity. Life expectancy (LE), as a health metric, offers the advantage of being a facile statistic that resonates with the public and policymakers and that is strongly associated with other measures of overall health. LE is traditionally calculated at the national or state/provincial level, and concerns about precision and accuracy arise when attempting to estimate or interpret LE on a small geographic scale. Hierarchical and geospatial modelling are methods, often used in tandem, that borrow information from other areas to stabilize estimates. Rather than assume LE is continuous over small geographic areas, we focus on continuity over areas with similar age-specific mortality distributions. We investigate replacing the geographic distance metric used to define spatial correlation with Hellinger distance, used to quantify similarity between probability distributions. The model is developed using a set of counties with stable LE estimates and tested using simulated small and larger areas.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program