Recognizing Differential Incidence Intensity of Geographical Disease Clusters Allowing for Confounding Variables

• Activity Number: 462 - Novel Spatial and Spatio-Temporal Models in Public Health
• Type: Contributed
• Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Chih-Chieh Wu* and Sanjay Shete
• Companies: National Cheng Kung University and The University of Texas MD Anderson Cancer Center
• Keywords: geographical cluster; hierarchical cluster; spatial clustering

Abstract

We propose to develop statistical methods for differentiating incidence intensity of geographical disease clusters, adjusted for confounding variables. The proposed methods prioritize incidence intensity and recognize geographical incidence intensity patterns of the areas with high incidence and areas with low incidence, respectively, incorporating known or suspected risk factors as confounding variables. These methods can construct hierarchical (in intensity) clusters of mutually neighboring high-risk areas and mutually neighboring low-risk areas on a map, respectively. With the models allowing for covariates, adjusted for risk factors, investigators can determine if some risk factors can explain the occurrence of geographical clustering of mutually neighboring high-risk areas or neighboring low-risk areas. The investigation into geographical areas of large or “peak” incidence is as useful and meaningful as geographical areas of incidence paucity in epidemiology. We use the data on the spatial occurrence of sudden infant death syndrome in North Carolina counties over 1974-1978, for orientation.