Spatial disease risk mapping with local odds ratio, risk ratio and attributable risk indicators

Authors: Chao Song*, Department of Geography, Dartmouth College; Institute of Spatial Information Technology and Big Data Mining, School of Geoscience and Technology, Southwest Petroleum University, Yan Chen Bo, State Key Laboratory of Remote Sensing Science, Research Center for Remote Sensing and GIS, and School of Geography, Beijing Normal University, Jin Feng Wang, LREIS, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Xun Shi, Department of Geography, Dartmouth College
Topics: Medical and Health Geography, Spatial Analysis & Modeling, Geographic Information Science and Systems
Keywords: Local disease risk mapping model; OR, RR and AR; Hand, foot and mouth disease (HFMD); Spatial autocorrelation; Environmental risk factors
Session Type: Paper
Day: 4/14/2018
Start / End Time: 2:00 PM / 3:40 PM
Room: Grand Couteau, Sheraton, 5th Floor
Presentation File: No File Uploaded


Purpose: Mapping disease relative risk is important in the field of spatial epidemiology. In general, original epidemiology relative risk indices like odds ratio (OR), risk ratio (RR) and attributable risk (AR) are global concepts for population groups in a study area. The aim of this article is to devise a general method to downscale the global indicates to local places for disease risk mapping.
Methods: A local disease risk mapping model is proposed that both consider the spatial autocorrelation and the environmental risk factors using spatial local OR, RR and AR. Both climate and socio-economic exposures are applied as environmental risk factors for modeling. We use hand, foot and mouth disease(HFMD)case at the county level in China as the example.
Results: The results indicated that the proposed method was easy to apply and predictions are made with high accuracy which was achieved at a 95% confidence level. Local OR and RR maps show similar macro-scale spatial patterns, but with different local spatial heterogeneity. Local AR map represents the sensitivity of selected variables in different geographical units. It is useful to predict the risk of no-data regions in practical disease prevention and fill the missing-data in risk mapping.
Conclusions: The local disease risk mapping model provides a general method to estimate relative risks indices OR, RR and AR in local places and can be easy applied for other diseases risk mapping.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login