The Republican-Democrat COVID Divide

(Originally posted as a blog entry on 6/13/21. Graphs will continue to be updated)

Misinformation and outright lies has been a major part of the problem of the COVID response in the US. This seems to be a particular problem related to the right and right-leaning media.

Methods

2020 election outcome data for the presidential race was used to group voters by county into six strata:

  • 80-100% Republican
  • 60-80% Republican
  • 50-60% Republican
  • 50-60% Democrat
  • 60-80% Democrat
  • 80-100% Democrat

Each of these strata were pooled for both 2010 population per the US Census Bureau and the incidence of COVID Cases. Due to the geographical election structure not aligning with county structure in Alaska, Alaska was excluded from this analysis.

The pooled COVID cases for each strata were then calculated as an incidence rate per 100,000 per calendar day. A 7-day moving average was then applied to each strata to remove some of the normal weekday variations and each line on the graph was colored to represent the degree of dominance of a political party (dark red represents >80% Republican, dark blue represents >80% Democrat) by 2020 presidential vote.

The hypothesis was that the more strongly a county voted Republican, the more likely it was to listen to and be influenced by misinformation and disinformation channels.

Results

There is a clear gradient in case rates depending on the political leanings of counties. This has remained consistent over time with two exceptions.

When the pandemic first started, a large proportion of the spread was in New York City. This was before there was a good understanding of the mode of transmission and in a very dense population area. As the virus spread into less populous area of the country and more knowledge was gained about transmission, other factors became more important in spread, such as messaging and beliefs.

The other anomaly in this pattern is in the spring of this year. This could be an effect of college students traveling during spring break, who are at an age where they are eager to return to normal and more likely to be a source of asymptomatic community transmission.

For better clarity in the just how big the differences are in the extremes, the next graph only shows the >80% wins in the election. Except during the rapid case declines that were due to the vaccination efforts that reduced spread and the previously noted exceptions, the rates of new cases in strongly republican counties were almost four times as high as those in strongly democratic.

Discussion

This becomes much more important as it related to vaccination efforts. As of this 7/31/22, the vaccination rates by state correlate well with voting. A current view of this map can be found here.

The majority of lies and disinformation seems to spread mostly in right wing media and is instigated by twelve different people. This is causing irreparable harm to health and to the economy. Sadly, as new variants come to dominate cases, the spread will be most obvious among those who have been fooled by these sources. What remains to be seen is whether the media will be held responsible for the damage they have caused.

Do your part. Get vaccinated. Encourage others to do so. As B.1.617.2 (the delta variant) gains a foothold, your health and life may depend on it.

An Epilogue

There is a similar pattern for deaths when stratified the same way, which is not surprising at all. The two big jumps for about a month in May 2020 are due to states catching up on death reports and submitting them all at once.

8/21/21 Addendum. Fox “News” carries some responsibility for what is happening by allowing lies like this.

8/26/21 Others are picking up on this source of the problem as well.

3/14/22 For the record, one year ago today was the earliest I believe I said anything about this data.

6/7/22 My work contributed to some academic research on this topic. “…The authors thank the Centers for Disease Control and Prevention and the New York Times for facilitating access to county-level COVID-19 mortality data. They also thank Charles Gaba of ACASignups.net, David Leonhardt of the New York Times, Gregory Travis, and Michael Olesen for their work documenting the differences in vaccine uptake, COVID-19 mortality, and county voting records. Their work to elucidate these differences inspired the work presented here.

The authors provided another view of the data that paints the same kind of picture as what I was trying to communicate. See The Association Between COVID-19 Mortality And The County-Level Partisan Divide In The United States if you want further analysis, or a blog post about this topic as well.

6/16/22 I had a notification from Facebook today that showed me the earliest date that I had seen this pattern was two years ago today. As the data became clearer on the differences, I obviously moved away for not discussing the politics of disease and death and figured out better ways to visualize the data. It has played a big role both in harming individuals, but also in harming society, yet those that need to hear this continue not to listen. The only thing that I changed on it was the permission to view it, which had been set to friends only, but now is public.

10/9/22 A study in Nature shows just how dangerous right wing media sources are in causing excess morbidity and mortality by reducing vaccine uptake among their viewers. With a little luck, some enterprising attorneys will see this as very ripe grounds for a class action lawsuit by those who have lost someone or are dealing with chronic disease issues as a result of their infection and relying on media sources like Fox for COVID information.