Published: Lung-Chang Chien & Ge Lin Kan

Lung-Chang Chien and Ge Lin Kan (both Environmental and Occupational Health) co-authored "Disparity of Imputed Data from Small Area Estimate Approaches – A Case Study on Diabetes Prevalence at the County Level in the U.S." to assess concordance and inconsistency among three small area estimation methods — multi-level logistic regression, spatial logistic regression, and spatial Poison regression — that currently are providing county-level health indicators in the U.S. They used diabetes prevalence at the county level from the 2012 sample of Behavioral Risk Factor Surveillance System as an example. The mapping results show that all three methods displayed elevated diabetes prevalence in the South, while point estimates are apparently different among different methods. This study provides the evidence about the need of building up a unified small area estimate method with necessary clusters and confounding variables to prevent possible inconsistencies in prevalence estimates. Both Chien and Kan are with the epidemiology and biostatistics section of their department.

People in the News

Ryan Brunty leans against a large, furry version of his yeti
People | December 11, 2019
Ryan Brunty's melancholy, vulnerable yeti, Yerman, has become a sensation locally and online. Now it's making its way to Hot Topic.
woman holding colorful fabric
People | December 11, 2019
Alumna Jaimee Newberry turns a mommy-daughter holiday project into a booming custom clothing company.
couple in front of product display
People | December 11, 2019
With a focus on "clean-beauty" products, 2009 alums Vontoba and Psyche Terry find their shelf space in the crowded skincare market.