October 15 Vaccine Analysis

FINDINGS AND ANALYSIS



Thanks to Brown University, we’ve got the monthly Vaccine Hesitancy Analysis mostly automated. This will be my last “Vaccine Hesitancy Monthly Analysis. You can view and download the data set from this google sheet.

  1. Vaccinations now account for 759,254 lives saved (deaths averted) this year, based on my calculations. The benefits of vaccinations are not evenly distributed across states and counties, often due to policy choices and politics. The top 100 counties have a vaccination rate, on average, of 74% versus the bottom 100 with only 22% on average. Let that gap sink in for a moment.
    • In my state, Nevada, comparing the lowest vaccinated county to the highest county, Storey has only 18.4% fully vaccinated. Next door, Carson City, the location of many state workers has the highest rates at 57.2%.
    • I’ve sorted “County Demo” tab by state and vaccination rate by county so you can scan and see the top and bottom counties easily. (Note, in this sheet, a few counties, mostly in Alaska, are not included in the sort for purposes of the regression).
  2. There are 8 States banning vaccine mandates and 19 states with mandates for state employees (including Nevada), according to NASHP.
    • The states banning vaccine mandates lag overall national vaccination rates and had slower growth over the past 30 days when controlling for the differences in prior month rate.
    • The average across the 8 states is 49.6% fully vaccinated as of Oct 15, versus 61.5% for the 19 states mandating vaccines for state employees. The 19 states with vaccine mandates for state workers grew vaccination rates 18.4% faster than the 8 states banning vaccines when controlling for the different base rate (a base adjusted growth calculation uses a denominator of 100% - base % to normalize differences in percentage base levels).
    • There are 70.5 million Americans living in the 8 states with vaccine mandate bans. Had these states achieved the same vaccination rate as the average for the 10 most vaccinated states of 67%, 12.3 million more people would have been fully vaccinated in these states and 40,895 more people in these 8 states would not have lost their life to COVID-19 this year.
  3. For the first time, the majority of increases in vaccines are not explained by severity of COVID in the county, politics, race and religion.
    • First, religion has never been all that significant of a factor explaining differences between counties with higher or lower vaccination rates. I’ve left it in the model just to check, given some discussion of religious exemptions.
    • Second, Race has decreased in significance in the model over the past 90 days, and now Black and Hispanic are growing at a faster rate (see below).
    • by severity of COVID in the county, measured by deaths per capita, continues to be the most significant factor, but it isn’t explaining the shifts in county level vaccination rates over the past month as it has in the past.
    • he big move in the model is new factors playing a larger role in explaining the incremental vaccinations of the past month. My model predicting vaccinations and the growth in vaccinations was very good, based on using the deaths per capita from COVID (the most important factor by far) along with Politics, Race and Religion (and advertising). The model had a pretty good r2, which had remained fairly stable over the past six months. However, looking at the incremental vaccinations over the past month, the r2 drops by 2/3rds. The increase in vaccinations over the past 30 days is most likely influenced by employer and government mandates which seem to have influenced more to get vaccinated than otherwise would have been predicted based on the prior six months patterns. (Some notes of how to take this analysis further are offered in the “Areas for others to investigate.”)
  4. Race: Both Black Americans and Hispanic Americans show higher than average increase in vaccination rates over the past month.
    • Counties with 40% of more of Black Americans, with a combined population of 17.7 million (192 counties), and 53.1% of population on average recorded as Black, based on 2019 Census, saw an increase of 5.2% growth in vaccinations over past 30 days on a base adjusted basis, 28% above the national average for all counties.
    • Counties with 40% of more of Hispanic Americans, with a combined population of 44.0 million (162 counties), and 58.3% of population on average recorded as Hispanic, based on 2019 Census), saw an increase of 4.6% growth in vaccinations over past 30 days on a base adjusted basis, 15% above the national average for all counties.
  5. Politics: Counties voting heavily for Trump in 2020 continue to fall behind on vaccination rates
    • Counties where Trump won by a 50% or more vote margin (2:1), with a combined population of 23.6 million, saw an increase of 3.4% growth in vaccinations over past 30 days on a base adjusted basis, -17% BELOW the national average for all counties. (92% of population on average recorded as White, based on 2019 Census.)
    • Counties where Biden won by a 50% or more vote margin (2:1) with a combined population of 30.7 million, saw an increase of 6.2% growth in vaccinations over past 30 days on a base adjusted basis, 55% above the national average for all counties. (55% of population on average recorded as White, based on 2019 Census.)
    • As a further note on the rural vs non-rural divide, consider that the counties where Biden won by 50% or better margin have an average population of about 580 thousand, whereas the counties where Trump won by 50% or better have an average county population of just 21 thousand people. (While population density, education and religion have all been evaluated in my model, the 2020 election margin are highly correlated with these variables and the election results better explains/predicts vaccination rates).


Using the same vaccine hesitancy framework inherited from HHS, where the bottom 350 counties are classified as “Very High,” we can see:

  1. The lowest 350 counties barely increased their vaccination rate from 27% to 29%. The bottom 350 are increasingly rural, and now only represent less than 2% of the overall US population.
  2. The next group, 470 counties listed as “High” have also shifted more rural (4.4%). Fully vaccinated went from 33.1% to 35.7%.
  3. The next 1180 counties, classified as “moderate,” went from 40.5 to 43.0% Fully vaccinated. This population represents 18% of total US.
  4. Those not in the bottom 2000 counties are classified as “not hesitant” and represent 77% of total US population. Vaccination went from 56.2% to 58.5%. On a base adjusted basis, this group had the largest growth rate. That said, the US lags behind most EU countries, despite having earlier access as scale to vaccines.

For more information on the model, click on the dashboard and select Model Methodology at the bottom left.

Full Screen Dashboard

Related Analysis

The views expressed on this site are my own. The company I founded, Marketing Evolution, is not responsible for this content. Copyright © 2020 Rex Briggs. All Rights Reserved.