The tutor is required to watch 3 videos, each 50 minutes long, and write whatever the professor says in the video. Here are the supporting articles, but I'll use dropbox to upload the video links. Y
Working Paper Series
Political lean: A crucial variable for monitoring COVID-19
in the United States
N. Krieger, PhD1, J. T. Chen, ScD1, C. Testa, BS1, P. D. Waterman, MPH1,
and W. P. Hanage2
October 8, 2021
HCPDS Working Paper Volume 21, Number 5
Abstract
Descriptions and monitoring of the social and spatial population distribution of COVID-19 cases and
deaths in the US have largely relied on individual- and county-level sociodemographic data and
vulnerability indices that draw primarily or exclusively from data available in US health records and US
census data. In this brief report, using US data from September 1, 2020 to September 15, 2021, we
provide empirical evidence demonstrating that county-level data on political lean (Republican vs.
Democrat for the 2020 US presidential election) adds critical information to understanding population
distributions of COVID cases and deaths – and also document the importance of socioeconomic variables
in addition to data on racialized groups. In particular, during the period from July 1, 2021-September 15
(corresponding to the delta surge, occurring when COVID-19 vaccines were authorized for all US adults
for at least 3 months or more), the two county level variables that most sharply differentiated risk
comparing the highest to lowest quintiles for COVID-19 rates (per 100,000 person-years) were: (a)
political lean: highest Republican lean vs. highest Democratic lean, for cases: rate ratio (RR) = 2.39 (95%
confidence interval [CI] 2.25, 2.55) and for deaths: RR = 3.34 (95% CI 2.99, 3.73), and (b) percent below
poverty line, for cases: RR 1.93 (95% CI 1.15, 2.4) and for deaths: RR = 5.08 (95% CI 3.14, 8.97). By
contrast, the least differentiation was provided by % people of color (highest vs. lowest quintile): for
cases, RR = 0.95 (95% CI 0.89, 1.02), and for deaths: 0.83 (95% CI 0.74, 0.93). However, combining
these single variables with political lean magnified the risk contrast between county quintiles. Thus,
people residing in the counties jointly with the highest poverty and highest political lean toward
Republicans were nearly 6 times more likely to die (rate ratio: 5.90; 95% CI 4.95, 7.07) from COVID-19
compared to those residing in the counties jointly with the lowest poverty and highest political lean
toward Democrats. Additionally, people residing in counties jointly with the highest % people of color
and highest political lean toward Republicans were almost 5 times more likely to die (rate ratio: 4.77,
95% CI 3.70, 6.20) from COVID-19 compared to people residing in counties jointly with the lowest %
people of color and highest political lean toward Democrats. We accordingly posit that county-level
political lean is a crucial variable that should be used routinely to monitor county-level trends in COVID-
19 cases and mortality, alongside and in conjunction with sociodemographic and socioeconomic data.