Does Density Aggravate the COVID-19 Pandemic?

Figures & data

Table 1 Variables used to explain COVID-19 per capita death rate (N = 913).

	Description	Mean	SD	Source
Outcome variables
Death rate	No. deaths per 10,000 population as of May 25	2.12	3.15	Dong et al. (2020)
ln of death rate	Natural log of no. deaths per 10,000 as of May 25	−0.02	1.26
Virus rate	No. COVID-19 cases per 10,000 population as of May 25	42.91	77.48
ln of virus rate	Natural log of no. COVID-19 cases per 10,000 as of May 25	3.21	0.98
Explanatory variables
Days since first case	No. days since the first case was confirmed	68.94	8.78	Dong et al. (2020)
Activity density	Population + employment per square mile	907	2,387	ACS 5-year estimates (U.S. Census Bureau, 2020a) and LEHD (U.S. Census Bureau, 2020b) for 2017
ln of activity density	Natural log of activity density (population + employment per square mile	5.78	1.37	ACS 5-year estimates (U.S. Census Bureau, 2020a) and LEHD (U.S. Census Bureau, 2020b) for 2017
Metropolitan population	MSA population	2,027,153	3,481,763	ACS 5-year estimates (U.S. Census Bureau, 2020a)
ln of metropolitan population	Natural log of MSA population	13.54	1.40
% of Black population	% of Black or African American race	12.49	14.15
% of college-educated population	% of adult population with some education beyond high school	57.44	10.44
% of population aged 60 +	% of population aged 60 years and over	21.73	4.69
Primary care physician rate	Primary care physicians per 10,000 population	6.54	3.57	County Health Rankings (University of Wisconsin Population Health Institute, 2020)
% of adults currently smoking	% of adults that reported currently smoking	16.61	3.13
Air pollution	Average daily amount of fine particulate matter in micrograms per cubic meter (PM2.5)	9.93	1.66
ICU beds rate	ICU beds per 10,000 population	2.04	2.09	Kaiser Health News (2019)
COVID-19 testing rate	Statewide number of COVID-19 testing per 10,000	437.45	157.26	The COVID Tracking Project (2020)
Enplanements rate	Annual enplanements in metropolitan area per 10,000 population	21,102	25,557	Federal Aviation Administration (2018)
% Staying at home	Average % of population staying at home (i.e., no trips with a non-home trip end more than 1 mile away from home) since the first case was confirmed until May 25	27.23	5.49	Maryland Transportation Institute (2020)

Table 2 Pearson correlation between explanatory variables and confirmed virus and death rates.

	Virus rate	ln of virus rate	Death rate	ln of death rate
Days since first case	0.054	0.185**	0.162**		0.135**
Activity density	0.204**	0.270**	0.364**		0.235**
ln of activity density	0.125**	0.280**	0.233**		0.196**
MSA population	0.285**	0.352**	0.454**		0.322**
ln of MSA population	0.152**	0.255**	0.257**		0.250**
% of Black population	0.151**	0.318**	0.221**		0.235**
% of college educated	−0.098*	−0.015	−0.029		−0.038
% of smokers	0.027	−0.032	−0.048		−0.018
Air pollution	−0.018	0.046	−0.021		0.027
Primary care physicians rate	−0.008	0.065	0.088**		0.069*
ICU bed rate	−0.040	<0.001	−0.049		−0.055
% of population aged 60+	−0.090	−0.174**	−0.003		0.028
Enplanement rate	0.043	0.086**	0.055		0.085*
% Staying at home	0.124**	0.275**	0.341**		0.284**
COVID-19 test rate	0.247**	0.341**	0.408**		0.327**

Notes: *Correlation is significant at the .05 level (two-tailed). **Correlation is significant at the .01 level (two-tailed).

Figure 1. Causal path diagram for COVID-19 death rate in terms of county density and other variables (correlational arrows among explanatory variables are not shown for the sake of simplicity and clarity).

Table 3 Direct effects of density, socioeconomic, and health care characteristics on COVID-19 infection and mortality rates.

Variables	Coefficient	SE	Critical ratio	p Value
ln of activity density → ln of virus rate	0.005	0.035	0.159	.874
% of Black population → ln of virus rate	0.018	0.002	8.716	<.001
% of population aged 60+ → ln of virus rate	−0.021	0.006	−3.474	<.001
% of college-educated population → ln of virus rate	−0.021	0.004	−5.902	<.001
ln of MSA population → ln of virus rate	0.135	0.028	4.81	<.001
COVID-19 testing rate → ln of virus rate	0.002	2^−4	8.838	<.001
Enplanement rate → ln of virus rate	−3^−6	1^−6	−2.204	.028
% Staying at home → ln of virus rate	0.056	0.009	6.276	<.001
% of smokers → ln of death rate	0.021	0.011	1.847	.065
% of Black population → ln of death rate	0.003	0.002	1.581	.114
ln of virus rate → ln of death rate	0.974	0.028	35.391	<.001
% of population aged 60+ → ln of death rate	0.048	0.005	8.958	<.001
ln of MSA population → ln of death rate	0.087	0.022	4.033	<.001
ln of activity density → ln of death rate	−0.118	0.058	−2.05	.040
ICU bed rate → ln of death rate	−0.029	0.014	−2.043	.041
% Staying at home → ln of death rate	0.044	0.018	2.371	.018

Note: Analysis was restricted to counties without missing values. Chi-square = 3.528; degrees of freedom = 2; p value = .171; root mean square error of approximation = 0.029 (p value = .698); comparative fit index = 1.0.

Table 4 Direct, indirect, and total effects of the county-level density and other variables on COVID-19 mortality rate.

	Direct effect	Indirect effect	Total effect
ln of activity density	−0.118	0.005	−0.113
ln of MSA population	0.087	0.132	0.219
Enplanement rate	0	−3^6	−3^6
COVID-19 testing rate	0	0.002	0.002
% of population aged 60+	0.048	−0.021	0.028
% of college-educated population	0	−0.020	−0.020
% of Black or African American population	0.003	0.017	0.021
% of smokers	0.021	0	0.021
ICU bed rate	−0.029	0	−0.029
ln of virus rate	0.974	0	0.974
% Staying at home	0.044	0.055	0.099

Dong, E., Du, H., & Gardner, L. (2020). An interactive web-based dashboard to track COVID-19 in real time. The Lancet Infectious Diseases, 20(5), 533–534. https://doi.org/10.1016/S1473-3099(20)30120-1

 Google Scholar

U.S. Census Bureau. (2020a). American community survey 5-year estimates, 2013–2017 tables. https://data.census.gov/cedsci

 Google Scholar

U.S. Census Bureau. (2020b). LEHD origin-destination employment statistics data (2002–2017). https://lehd.ces.census.gov/data/#lodes. LODES 7.4.

 Google Scholar

University of Wisconsin Population Health Institute. (2020). County health rankings & roadmaps. https://www.countyhealthrankings.org/app/2020

 Google Scholar

Kaiser Health News. (2019). Millions of older Americans live in counties with no ICU beds as pandemic intensifies: Analysis of hospital cost reports filed to the Centers for Medicare & Medicaid Services. https://khn.org/news/as-coronavirus-spreads-widely-millions-of-older-americans-live-in-counties-with-no-icu-beds/

 Google Scholar

The COVID Tracking Project (CTP). (2020, May 25). COVID-19 data API. https://covidtracking.com/

 Google Scholar

Federal Aviation Administration (2018). Enplanements at all commercial service airports (by rank). https://www.faa.gov/airports/planning_capacity/passenger_allcargo_stats/passenger/

 Google Scholar

Maryland Transportation Institute. (2020). University of Maryland COVID-19 impact analysis platform. https://data.covid.umd.edu

 Google Scholar