Using trajectory modeling of spatio-temporal trends to illustrate disparities in COVID-19 death in flint and Genesee County, Michigan

Richard Casey Sadler; Thomas W Wojciechowski; Zachary Buchalski; Alan Harris; Danielle Lederer; Matt Peters; Pamela Hackert; C Debra Furr-Holden

doi:10.1016/j.sste.2022.100536

Using trajectory modeling of spatio-temporal trends to illustrate disparities in COVID-19 death in flint and Genesee County, Michigan

Spat Spatiotemporal Epidemiol. 2022 Nov:43:100536. doi: 10.1016/j.sste.2022.100536. Epub 2022 Aug 27.

Authors

Richard Casey Sadler¹, Thomas W Wojciechowski², Zachary Buchalski³, Alan Harris⁴, Danielle Lederer⁵, Matt Peters⁶, Pamela Hackert⁷, C Debra Furr-Holden⁸

Affiliations

¹ Associate Professor, Michigan State University, Flint, MI, USA. Electronic address: sadlerr@msu.edu.
² Assistant Professor, Michigan State University, East Lansing, MI, USA.
³ Data Analyst, Michigan State University, Flint, MI, USA.
⁴ GIS Analyst, Michigan State University, Flint, MI, USA.
⁵ Chief Epidemiologist, Genesee County Health Department, Flint, MI, USA.
⁶ Epidemiologist, Genesee County Health Department, Flint, MI, USA.
⁷ Medical Health Officer, Genesee County Health Department, Flint, MI, USA.
⁸ C.S. Mott Endowed Professor of Public Health, Michigan State University, Flint, MI, USA.

Abstract

COVID-19's rapid onset left many public health entities scrambling. But establishing community-academic partnerships to digest data and create advocacy steps offers an opportunity to link research to action. Here we document disparities in COVID-19 death uncovered during a collaboration between a health department and university research center. We geocoded COVID-19 deaths in Genesee County, Michigan, to model clusters during two waves in spring and fall 2020. We then aggregated these deaths to census block groups, where group-based trajectory modeling identified latent patterns of change and continuity. Linking with socioeconomic data, we identified the most affected communities. We discovered a geographic and racial gap in COVID-19 deaths during the first wave, largely eliminated during the second. Our partnership generated added and immediate value for community partners, including around prevention, testing, treatment, and vaccination. Our identification of the aforementioned racial disparity helped our community nearly eliminate disparities during the second wave.

Keywords: COVID-19; Epidemiological methods; Group-based trajectory modeling; Health inequalities; Racial disparities; Spatial analysis.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

COVID-19*
Humans
Michigan / epidemiology
Seasons