Local Determinants of Morbidity: Using Multi-Scale Spatial Modeling to examine associations in US counties between socio-demographic indicators, built environment characteristics and Covid-19 deaths
Topics: Spatial Analysis & Modeling
, Social Geography
,
Keywords: Covid-19, health geography, epidemiological modeling, pandemic, excess deaths, regression, geographic weighted regression, multi-scale geographic weighted regression
Session Type: Virtual Paper Abstract
Day: Saturday
Session Start / End Time: 2/26/2022 11:20 AM (Eastern Time (US & Canada)) - 2/26/2022 12:40 PM (Eastern Time (US & Canada))
Room: Virtual 33
Authors:
Dave J Cook, University of Maryland, Center for Geographic Information Science
Taylor M Oshan, Associate Professor
,
,
,
,
,
,
,
,
Abstract
This paper examines the geography of Covid-19 morbidity in the U.S. from 2/2020 to 6/2021. Rates of Covid-19 morbidity are often seen on a map through a common geographic lens. A shared public lens, though, does not accurately account for the contextual indicators and characteristics, the local determinants, that influence disparities and differences in morbidity. Covid-19 morbidity varies greatly across counties, regions and populations. The pandemic’s impact has largely been measured through death certificates. Recent assessments, however, of county-level excess deaths- the difference between observed and expected numbers of deaths in the same time periods- determined that death records did not fully capture pandemic morbidity. With the percentage of excess deaths not properly attributed, the scope and scale of the pandemic is not being analyzed or properly captured geographically. This research will build a county-level excess death model and test excess deaths against socio-demographic indicators and built environment characteristics using machine learning. Ordinary Least Squares will be used to define a baseline, aspatial model. Geographically Weighted Regression (GWR) will then be applied to better understand the local dynamics of the explanatory variables. GWR assumes, however, that all processes are modeled on the same bandwidth which can obscure nuanced relationships of place and space. Therefore, a multi-scale geographically weighted regression (MGWR) will also be applied to improve understanding of local and regional spatial processes and the scale at which they operate, along with clarifying how socio-demographic and built environment characteristics influence Covid-19 morbidity overall.
Local Determinants of Morbidity: Using Multi-Scale Spatial Modeling to examine associations in US counties between socio-demographic indicators, built environment characteristics and Covid-19 deaths
Category
Virtual Paper Abstract
Description
This abstract is part of a session. Click here to view the session.
| Slides