SIO343: Excess Republican COVID Deaths Are Massive. But Not Surprising.

Transcript

This transcript was generated automatically and may contain errors — the audio is the record.

0:00:13 Thomas: Hello and welcome to Serious Inquiries Only. This is episode 343. I'm Thomas. That's Dr. Lindsey Osterman, professional co-host. How you doing?

0:00:21 Lindsey: I'm fantastic. A sedentary little today, honestly, but otherwise, okay. How are you?

0:00:24 Thomas: Mmm, yeah. I'm sedentary because it's a day that ends in day, or That's the old dad joke, daily that ends and why, I guess. Makes more sense.

0:00:30 Lindsey: Yep, yep. But, despite being sedentary, did get my Omicron booster a few weeks ago, so I feel pretty good about that.

0:00:41 Thomas: Oh nice, I got mine, yeah, longer ago than that because, you know, I am just more into science and medicine and all all that kind of thing.

0:00:42 Lindsey: Yeah. Wow. Okay. All right. Let's see how this recording's gonna go.

0:00:52 Thomas: No, I jumped on it right away. It was of those ones where it wasn't even displaying in my Kaiser app. Every time I mentioned my health insurance, anyone who doesn't know Kaiser is such a mystery. Eli's joked about it too on the other show. Anyway, yeah, it was not even available in the app, but then I was waiting and watching. It was refreshing. I was trying to buy a PlayStation 5, and then I saw it all of a sudden. I I was like, yeah, jump on it now.

0:01:16 Thomas: I got it right away. I think even Phoebe Smith is can get a third shot. So the first booster maybe is available for kids.

0:01:16 Lindsey: Wow, is impressive. Nice.

0:01:24 Thomas: I not sure. Well, I have the youngest group of kids, right? Two to five. So they've like two shots behind us, right. That's Yeah.

0:01:31 Lindsey: Yeah, yeah, that sounds right.

0:01:32 Thomas: Cool. Good news. Hey, is that at all related to what we're going to talk about today in any way?

0:01:33 Lindsey: No idea what you're talking Yeah.

0:01:37 Thomas: Or are we just making casual but natural conversation? How

0:01:42 Lindsey: Yeah. PSA, everybody. Get your boosters. Turns out to be important.

0:01:47 Thomas: Could you know that? Is there some way we could possibly measure any of those outcomes scientifically?

0:01:52 Lindsey: Yeah, there's like a whole method. And it's really scientific. And we are talking about a paper that came out very recently. You sent this to me could It's called excess death rates for Republicans and Democrats during the COVID -19 pandemic. So they're leaving a little mystery as to which way this relationship will go.

0:01:54 Thomas: Tell is. Why don't you tell us all about it?

0:02:12 Thomas: I wonder.

0:02:14 Lindsey: Came out, again, just very recently, 2022. It's actually not a peer -reviewed paper, which I don't know has been obvious from the reporting. It is an internal report from the National Bureau of Economic Research. So, context. Context?

0:02:19 Thomas: Oh, okay. Okay. So it's completely made up and designed for the lizard people to inject us, yeah, okay

0:02:28 Lindsey: Yeah. Pure propaganda. Yes, Exactly. Right. No, but it's a good report. It's a good paper. And there's a lot of good stuff to talk about here. Findings in this report are not exactly new. If anybody's been paying half attention, they've probably seen a number of headlines come out over the past year that suggests that COVID deaths are higher in red counties than blue counties.

0:02:48 Thomas: Not even making this up, I, when saw a headline around the same, I don't know if I was browsing the same thing or whatever, when I sent you this that said, like, ex dying more than Democrats of COVID, experts are baffled or something, you know, something like that. Hat. And I was like, oh, is this satire or like sarcasm? But then I saw it was the New York Times.

0:03:06 Lindsey: Huh?

0:03:06 Thomas: I I was Like, I don't think they do like sarcastic headlines. So could there be any mystery about this? That's the part I'm confused about.

0:03:14 Lindsey: Yeah, that's weird. Wait, I mean, was that connected to COVID specifically? Or was it?

0:03:17 Thomas: Yes. Yes, of COVID. Admit I didn't read the article. I was so baffled by the headline. I Was like, well, that's yeah, obviously Republicans would be dying more because they don't believe in science. But I I guess we'll have to find out what this report says here. I'll

0:03:18 Lindsey: Huh?

0:03:27 Lindsey: No, that that's really interesting. I would like to know what the story is behind the headline. If you find it again, let me know.

0:03:31 Thomas: Have to find it again.

0:03:32 Lindsey: Yeah, yeah, I agree with you. This is not surprising. This feels pretty much inevitable. Trump, Trump, and other Republican leaders lied their asses off about COVID and then later about vaccines in a way that seems like it was bound to have dangerous consequences for Republican voters, you know, assuming that those voters actually believed the bullshit that they were spewing. In fact, this report that we're talking about cites an experimental study that in March 2021, in which they found that Republican participants who saw a video clip indicating that Republican leaders endorsed the vaccine expressed significantly higher vaccination intentions compared to either a neutral control condition or a condition in which THEY saw endorsement from Democratic leadership.

0:04:14 Lindsey: And Trump and other Republican leaders would want them to get vaccinated.

0:04:17 Thomas: So, yes, and now I'm remembering back in the 2020 election, I'm pretty sure something came out shortly after that was like Trump's and the Republicans' lies about COVID may have killed enough people in Georgia to have made the election difference. So I feel like this is not a new thing, but this is a new report that has some more stuff.

0:04:33 Lindsey: It does.

0:04:37 Thomas: So but But it's interesting that it also, it sounds like they've isolated this a little bit further, like just directly, there's handed in terms of the, it's the fact that these leaders are spreading these lies that like seems to be pretty causally related to this phenomenon.

0:04:37 Lindsey: Does. Yeah. Yeah, Yeah, appears to be part of the picture, at least a causal part of it. May there be other things going on, but I thought it was worth sharing that finding because it does suggest that like what these leaders say matters and it has an effect on people's attitudes and thus their behavioral intentions and eventually their behaviors. So the stuff that they did prevented people from getting vaccinated. That's them from engaging in other forms of behaviors to prevent the spread of COVID.

So anyway, so based on the evidence collected thus far, it really does feel irresistible to conclude, like from those previous findings that red counties had more COVID deaths than blue counties, it feels irresistible to conclude that more Republicans than Democrats died of COVID because of the very stupid and dangerous lies that Republican leaders told about. Thing that's notable about this new report is that it documents this relationship using individual level data rather than analyzing this relationship at the county level.

And this is a really important statistical gap for them to fill because of something called the ecological fallacy. So when you find a relationship between variables at a group level, like a state or county level analysis, tempting to assume that the correlation holds among individuals, but this is actually a bad inference, as it turns out.

So it's entirely possible to find that things are correlated at the level of the state, but that that same relationship does not exist when you analyze individual data, or even the relationship reverses in some cases.

0:05:04 Thomas: This sounds like a cool stats thingy. I'm not a hundred percent sure. Is it because it be like a statistical anomaly maybe or no?

0:06:23 Lindsey: It not necessarily, it's just that you can get a correlation at a group level for reasons that have nothing to do with an individual level relationship. So I've got an example here. I did quite a bit of work in grad school with state-level associations because we were doing a lot of work on something called the culture of honor and its various outcomes in the United States. Southern and Western states have been designated as honor states because they have populations that are disproportionately descended from the Scotch-Irish immigrants who brought over this strong culture of honor, right?

Means like it's a culture that places a strong emphasis on reputation and endorses like really extreme and even violent recourse if somebody threatens your reputation. States are designated non-honor states because their populations have different historical roots that don't emphasize those honor values. So there's been a bunch of research that sort of analyzes data at the state level to get get a sense of relationships between honor, culture, and these different outcomes, right? Back to ecological fallacy.

In one paper, we showed that honor states have higher depression and suicide rates compared to non-honor states, and that's even after controlling for a bunch of potential confounds. So it's tempting to look at that correlation at

0:06:34 Thomas: Okay. Oh,

0:07:32 Lindsey: The level of the state, yeah, and assume that it means that individuals who endorse honor ideology are at higher risk of dying by suicide. But it could be that something completely different is driving that correlation. So it could Be that if you live in an area dominated by honor values that you don't share, it's alienating and isolating, and therefore the elevated suicides and honor areas are actually disproportionately among people who are low in honor values. That's totally possible. And that would mean that you get a positive correlation between honor and suicide at the state level, But a negative one at the individual level.

0:07:32 Thomas: I think I know where this is going. Yeah. Yeah. Oh, interesting.

0:08:06 Lindsey: Yeah. So

0:08:06 Thomas: Yeah. That's what I was thinking. So like it's the environment itself that is really causing the effect rather than like the individual beliefs of the person. Yeah, that's really interesting. And I could imagine that with COVID because of the like, you know, it's a community spread type thing.

0:08:15 Lindsey: That's right So Yeah, right. That's kind of where I went to. Yeah, so

0:08:24 Thomas: And even if that were happening here, it would still be like, you know, a bad thing about these beliefs, I guess. It would just be less bad, maybe. Like, it's not as directly causing people to die of COVID, but it is causing people in the environments of these people to die of COVID. Does that make sense?

0:08:41 Lindsey: I think so. Um there's several potential explanations if the explanation is not actually that republicans are dying more often, right? Which is possible, right that could be the reason that you get the state level of association. It's just not the only possibility um, I think another possibility is that like People are dying more in these counties because of something along the lines of there's more Republicans, so fewer people are getting vaccinated because of that. But that for whatever reason, that doesn't mean that Republicans are disproportionately dying more. Both Republicans and Democrats are both dying more because the lower sort of transmission. So I don't know that that's necessarily a better picture. As say, the Republicans are still sort of creating that environment.

0:09:11 Thomas: Yeah.

0:09:21 Thomas: I think I was trying to say that with, with way worse words, which is like, like, essentially, it's still an indictment of the fact it's just like a little bit different than saying, like, on an individual level, this leads to these outcomes. And honestly, you might expect that with stuff that is tragedy of the commons type stuff.

0:09:26 Lindsey: Right. Yeah.

0:09:38 Thomas: Like, you could almost turn it into that. Like, oh, it turns out, you individual tragedy doesn't correlate with what these beliefs on the tragedy of commons, but like it does as a group, right?

0:09:49 Lindsey: Yeah, yeah, The level association, the county level associations, there a piece of this that we are missing that maybe we would like to know about, which is who is dying, right? So the studies on this question still have that gap. What they're doing in the current paper is to fill that gap by looking at individual level data to see if the relationship holds between partisanship and death rates from, they do that by using voter registration data to link party affiliation to excess death rates at an individual level.

And so, holds up here, then we can be more confident that what is driving the association between excess death rates in red versus blue counties is actually this individual relationship. It's a good report. This, like, is, I think, a very meaningful question for them to answer. Yeah. So let's talk a little bit about the data they used. They obtained publicly available 2017 voter files in Florida and Ohio. So this is not a analysis. They just focused on people in these two states.

Files contained a lot of information. So they had full name, date of birth, county of voter registration, and party affiliation. And the Florida voter files also included gender. Ohio did not. These two states in particular? Voter registration data was really easily accessible for these two States. So this is a convenient sample.

Talked before, I think, about like when we did the episodes on how social media influences election outcomes and things, we talked about the fact that like you can gain voting information, including like voting records and registration information. But it varies from state to state in terms of how that information is. Some states charge fees. Some of them require more credentials or an application process. In Florida, it's real easy. You can just submit a request to that handles this, and they will send you discs. So you have to buy a CD player.

0:11:38 Thomas: How is this easy? What is a disc? How am I going to find this ancient technology? Got to go to like a historical salvage and

0:11:41 Lindsey: Yeah, it's pretty tough. You to got to go to your local, uh, shack.

0:11:49 Thomas: Then find a local Radio Shack. Does Radio Shacks still exist? I feel like there's a couple of them.

0:11:49 Lindsey: Yes. I don't think so. Did they didn't they close

0:11:55 Thomas: I I don't know. I

0:11:55 Lindsey: It. Yeah, you can send in the request, they send you the data files, and then you've got all the voter registration files. Um, in Ohio, you can just go to a website like right now if you want to, and download all the voter registration files. I

0:12:06 Thomas: Oh, yeah. We have talked about this on something. I can't remember what it was.

0:12:08 Lindsey: Yeah. Um, downloaded the complete voter registration data for Adams County in Ohio just because I could. It was really easy.

0:12:17 Thomas: It's just the power went to your head.

0:12:18 Lindsey: It's kind of creepy. It's like, it's got everybody's names and addresses and stuff. It's

0:12:20 Thomas: Yeah. Whoa. Wait. For real, I thought it would be like anonymized in some way.

0:12:23 Lindsey: Yeah. No.

0:12:26 Thomas: What, and You can tell, you can't tell how people voted, but you can call a party affiliation. Okay, um

0:12:27 Lindsey: No, yeah. Find Yeah. That creepy. I don't like it.

0:12:34 Thomas: Do As well.

0:12:34 Lindsey: It's a very searchable format.

0:12:35 Thomas: I Hope California protects that a little more. Yeah.

0:12:40 Lindsey: Does a little bit. I was looking at some of the state by state regulations, and it looked like, it looked like California had more restrictions on who can get access to what. Um, Yeah, I think, I think all states will have some voter registration information that's available publicly.

0:12:48 Thomas: Yeah. And For For You candidates to be able to run and to, you know, where to canvas and harass us and send us email, whatever. It's It's for

0:12:52 Lindsey: Yeah. Exactly, yeah. Right. So they got these voter registration records, and they only analyze data for registered Democrats and Republicans. So people that were unaffiliated or independent voters, or affiliated with some other party, were excluded from this analysis. Separately, they obtained detailed mortality data for through 2021 so that they could calculate excess COVID deaths. And they got this data from a service called DataVant. This actually sounds pretty cool. DataVant maintains a database that connects federal official death files with information from newspapers and funeral homes and other sources to construct an individual level database with information on most deaths in the U.S. It's like this database contains more than 80% of deaths. That That's a pretty impressive data thing. DataVant maintains the database internally and provides information to clients by request. So it was actually DataVant, and not the researchers themselves, that linked the voter file information information with the death data so that these researchers could then do the analysis they wanted to do. Yeah, so they linked the data from the voter registration files with the entries in DataVant.

And presto, they have a database with information on party affiliation of nearly 600,000 people age 25 and older who died between 2018 and 2021, as well as some basic demographic information, age and gender. They also used county -level data from the CDC on vaccination rates so that they could look at how excess death rates changed as a function of vaccine adoption in each county.

They got information on what proportion of each county in Florida and Ohio had been vaccinated as of June 6, 2021, because this was two months after the vaccines became available to all adults in Ohio and Florida, and therefore everybody who wanted to get vaccinated theoretically theoretically could have by that point, right? I can explain how they estimated excess COVID deaths, I need to explain how the researchers computed separate death counts for Republicans versus Democrats by month, county, and age group.

And the age groupings that they used were relevant to COVID risk. Mortality, so groupings were 25 to 64, 65 to 74, for 75 to 84, and 85 and older, right, which is linked with meaningful in terms of what your likely mortality is if you get a COVID infection.

Had these death counts broken out by party affiliation, month, county, and age group, meaning that one of their data points, for example, was how many registered Republicans between the ages of 25 and 64 died in Franklin County, Ohio in March 2018. They had that same count for March 2019, March 2020, March 2021, and so on through January of 2022. And then they had a comparable count for just Democrats in that age range and county and so forth. Does

0:13:37 Thomas: Yeah, straightforward, but... excited for this because it seems like really good data. Likewell, a lot you can tell from this.

0:15:45 Lindsey: There really is a lot they can do with it. So because they arranged their data this way, they were able to compute excess death rates that controlled for pre-COVID mortality differences that were there as a function of county of residence, age, time of year, and political party. To Yeah, be precise about it, their of excess COVID deaths is essentially the percent difference between total deaths in any given month for any given county, age range, etc. et cetera. COVID, compared with the same month in 2019, controlling

0:16:01 Thomas: Right, right, right.

0:16:16 Thomas: Mean, because like the first way you'd try to pick this apart probably is be like, well, maybe these people were going to die sooner anyway for some reason. But if you're kind of isolating all those variables and you're making the apples to apples comparison. Which idiot apples denied the science and didn't get the vaccine? Let's How do we like them apples?

0:16:26 Lindsey: Yeah, exactly. Exactly. I'm on the edge of my seat. Love them apples. Okay. So let's talk about the results. So first, they just computed excess deaths overall relative to 2019 over for one year pre -COVID, so they used 2018, and for post -COvid. So that's March 2020 through January 2022. this first analysis is just to establish kind of a baseline of what excess deaths looked like over time. It isn't broken out by party affiliation yet, so we're just establishing what the overall trend in excess deaths looks like.

0:17:02 Thomas: Yeah. Yeah.

0:17:02 Lindsey: In terms of why they included 2018 excess deaths, which doesn't really make any sense since that was before COVID, right? But they included 2019 because it's useful to look at a pre-COVID year in addition to the years that we're actually looking at to get a sense of annual fluctuation, right. And so when you look at their graph, what you see is that, quote-unquote, excess deaths in 2018, just hover around zero, like you would expect them to, they're essentially identical to 2019, um, just little random fluctuations around zero percent change, right?

0:17:17 Thomas: Oh, Wait, there's no excess. So, oh, it's excess deaths due to COVID. Well,

0:17:36 Lindsey: Mean all they have is excess deaths, right? They don't have like cause of death in this data set, right?

0:17:40 Thomas: Then how, sorry, how is an excess death defined then? Oh,

0:17:44 Lindsey: Death, like the way that they get this, it's just like the the percentage change relative to 2019, so

0:17:49 Thomas: Okay, so that's why it would be zero, is it's relative to a, yeah, Okay, gotcha. Because I was like, there's not zero premature deaths, I was trying to figure that out, like Year over year, same

0:17:52 Lindsey: Yeah. No, Yeah, Um, it just means that, yeah, the same in 2018 as 2019, essentially.

0:18:03 Thomas: Number of people driving too fast, roughly doing Dangerous stuff, Whatever,

0:18:06 Lindsey: Right. Yeah. Yeah. Yeah. And just to underline, right, like it's not just that they're comparing the overall rate for 2019. Like they have this broken out on a month and county and age range basis. So it's like the number, like average, like the numbers of 25 to 64-year-olds who who live in this particular county, died at approximately similar rates in January of 2018 versus January of 2019. That's the level of granularity of their comparisons, which is important, I think. Right?

So 2018 essentially looks the same as 2019 in terms of quote-unquote excess deaths, meaning that those rates look about the same. Excess deaths during the pandemic show exactly the peaks that you would expect if you think back to the first two significant COVID waves that we got. So the first peak that you see here is in July 2020. The second one is in December 2020 slash January 2021. So no surprises there. We were all there. We saw those come and go, right?

Thing worth noting here is that it looks like their estimate of excess deaths is tracking the information that we have about excess deaths from other sources that use different methods of estimating excess COVID the deaths, right? So that's useful to know. Next, they examined excess deaths over time separated by party affiliation. What you see in their data here is pretty striking. So excess deaths among Republicans and Democrats look no different from each other in 2018.

0:18:08 Thomas: Yeah,

0:19:26 Lindsey: So they're on top of each other. Mm -hmm. I thought so too. I though it was possible that there was a Difference because of various factors, but nope, nope, no different. They also don't look different for the the first part of the pandemic. So for the first few months after COVID became a thing in the U.S., which was March 2020, Democrats and Republicans have excess death rates that are completely on top of each other. There's no difference there. Excess deaths among Republicans start pulling slightly ahead of excess deaths among Democrats around October 2020, and thereafter remain at least nominally higher than Democrats, at least a little higher. But the difference is pretty small through late 2020 and into early 2021. And it's not statistically significant. There are like maybe two months during this earlier period in the pandemic in which the difference is just like barely statistically different from zero. But generally speaking, Republicans and Democrats did not have different excess death

0:19:28 Thomas: That's Interesting Already That's A little interesting thing, yeah, I

0:20:26 Lindsey: Rates early in the epidemic, according to their estimates. Well,

0:20:27 Thomas: And Um, Remind Me, uh, if I'm, correct me if I am wrong, but I seem to recall that at first COVID wasn't terribly politicized, and it sort of took time for, then for it to become a real partisan issue, uh, at the hands of Trump and others.

0:20:42 Lindsey: I think that I think That's fair. But I mean, we're talking about into early early 2021. So this is like, yeah.

0:20:47 Thomas: Oh, it's that long? Oh, okay.

0:20:49 Lindsey: Yeah. So thought that was a little bit surprising. I would have, well, we're getting to the vaccine.

0:20:50 Thomas: Yeah, that's interesting. Well, well, no, I guess because, I mean, it took so long for the vaccine to come out that, okay,

0:21:01 Lindsey: But like early on, you

0:21:01 Thomas: I guess what I'm saying is there, I don't, there wasn't much we could do early on, regardless mean, I guess social distancing and stuff, but okay, well I guess then that's the question is, that's why I'm kind of interested in some other data on this. I haven't seen like how much those measures helped at that time, you know?

0:21:08 Lindsey: Distancing.

0:21:19 Thomas: And like either Democrats and Republicans were doing those roughly equally or they didn't help that much, right?

0:21:18 Lindsey: Well, yeah, so they weren't doing them roughly equally. There's, we have very good data that like, Republicans and Democrats differed pretty pretty markedly in terms of their attitudes toward spread prevention measures like social distancing and their behavior, their actual behaviors measured with like cell phone data and stuff looked very different. So Republicans and Democrats were behaving differently. But yeah, at least from like this estimate that we're getting here, it doesn't look like that was translating into differential death rates, which I did find somewhat surprising.

0:21:27 Thomas: Okay. Yeah. Yeah?

0:21:56 Lindsey: I was like, okay, but surely there was a difference in like transmission rates early in the pandemic. And for whatever reason, that wasn't translating into excess deaths early on for some reason. But from what to find, the story is a little more complicated. The And data on this aren't awesome. So it's kind of, I know, I feel like we have an incomplete picture still of what was going on here. Yeah, so there's lots of evidence that Republicans and Democrats differed in what they were doing in terms of social distancing. But the available evidence on transmission rates is way more complicated." So one study suggested that transmission was actually way higher in blue counties than red counties.

0:22:30 Thomas: All I was just going to say, yeah, do just for population because

0:22:32 Lindsey: But exactly right. Population density is going to be hugely different there. Urban centers lean blue, rural, more sparsely populated areas lean red. And so that's going to obviously impact rates in a way that is difficult to control at an individual level, potentially. Yeah, so I think that that's part of what's going on. Yeah.

0:22:34 Thomas: Yeah. Yeah, yeah. No, No, that totally makes sense. I wasn't sure if that was controlled for in these in data, these because thinking back, my moron Republican family that denies all the science on this really felt pretty confident and cocky about it for the longest time. Part And of that was because, you know, Democrats live in places where people are and Republicans don't. So obviously it would be easier to avoid transmission in the country where I'm from, where the next person is five miles away.

0:23:17 Lindsey: Yeah.

0:23:18 Thomas: But it caught up to them because you can't do that forever.

0:23:22 Lindsey: That's right. That'S right. I mean, especially, like, so early on, especially you were probably okay for a few months after COVID got to the US in rural areas, just because of how long it takes for the cases to sort of work out. Right. So, yeah, I think that analysis is spot on. Was all this panic around it. And I think that Republicans who were listening to the messaging from leaders were looking at their experiences and thinking that confirmed that this wasn't actually a big deal. And that sucks because obviously was going to come a point in which that wasn't going to be true anymore. Gap in excess death rates between Democrats and Republicans does not become statistically significant until vaccines became widely available to all U.S. adults.

Month after the vaccines were up opened to all adults, so that's April 2021, the gap between the rate of excess deaths among Republicans and Democrats continued to widen. They had data in their set through January of 2022. So as of that month, earlier this year, Republican death rates were continuing to increase relative to Democrats. Republicans and democrats followed similar patterns in terms of when excess deaths peaked and troughed in a more absolute sense. And those patterns were in line with the spread of Delta and then later the Omicron variant. But in terms terms of the gap between Democrats and Republicans, that just steadily became larger and larger and larger from summer 2021 through at least January of 2022. Republican excess death rate was about 10% higher than that of Democrats in late summer of 2021. So it was like a little more than 20% excess death among Democrats versus about 35% excess deaths among Republicans.

0:23:34 Thomas: Yeah. Yeah. Wow. Wow.

0:24:57 Lindsey: And that gap widened to about 25% in the winter of 2021, when you saw about 10% excess deaths relative to 2019 among Democrats versus 35% among Republicans, right? So that's a, yes, they really are. This is not a, you % difference that happens to be statistically significant. These are big gaps.

0:25:10 Thomas: So these are significant numbers. Yeah. Right. I often ask you like, oh, what the hell does this effect size mean that you're talking about? But this seems pretty clear. Yeah.

0:25:23 Lindsey: This think that all sudden, and I've got some data for a point later, but like, Like, all said and done, right, talking about probably hundreds of thousands of people dying because they didn't get vaccinated, even though they had vaccines available to them. Those are concentrated in red states, probably, and among Republicans specifically. It's not just red areas.

0:25:44 Thomas: There any sort of like health care access control something, you could be a problem here?

0:25:50 Lindsey: Think that the fact that they are controlling for county would account for that.

0:25:55 Thomas: Yeah. Good point.

0:25:56 Lindsey: Yeah. Yeah, no, I think the way they broke this out was really smart in terms of addressing confounds like that. So the fact that this difference only emerged after the introduction of vaccines is pretty suggestive. That vaccine uptake, which we know from other studies, is associated with party affiliation, is probably a central part of why Republicans are dying at higher rates than Democrats. But we can't say that for sure, based on this analysis alone. So they did one final interesting analysis that I think can help bolster that inference.

0:26:08 Thomas: Yeah.

0:26:27 Thomas: Sorry. Why can't we say that for sure exactly from It

0:26:29 Lindsey: We don't have information about each individual's vaccination status when they died. Right. So it could be that there's some confound here that's producing this relationship or, but you can't completely rule it out. Right?

0:26:38 Thomas: Seems far -fetched, but I'll accept that. It must be a logical possibility. Yeah. Yeah. Okay.

0:26:45 Lindsey: That's still there. Or, you know, it could regardless of whether we got a vaccine, it was just going to take some time for that gap to emerge because of other behaviors and that that happened to be when the gap emerged, right? There's lots of other ways you could explain it.

0:26:53 Thomas: Oh, I see. Yeah. I I guess that makes sense. Okay. So you're just saying specifically the vaccine we can't tie it to. Oh, gotcha.

0:27:01 Lindsey: Yes, that's We can't say that they're dying because they're unvaccinated. But did do one final interesting analysis that I think can help bolster that inference. Again, you it's not going to be 100%, but this does get us, I significant step closer to making that inference pretty strongly. So they compared the excess deaths among Republicans and Democrats as a function of the percentage of vaccinated residents in each county.

0:27:02 Thomas: Yeah. Hmm.

0:27:23 Lindsey: And they examined this both before and after vaccines were widely available in order to rule out the possibility that counties with higher vaccination rates were different prior to the actual introduction of vaccines in other ways that were meaningfully impacting COVID deaths.

0:27:38 Thomas: Interesting. Interesting.

0:27:39 Lindsey: Yeah. So in other words, they're they're looking at the relationship between percentage of vaccinated residents and percentage of excess Republican and Democrat deaths in each county. And they're looking at this relationship both early in the pandemic period, prior to the vaccines, and later in the in epidemic, after vaccines, because even prior to the availability of vaccines, a county that had a 20 percent vaccination rate probably important ways from a county that eventually had a 60% % vaccination rate, and in ways that could have impacted excess deaths, right? So the point of this analysis is to help them strengthen that inference from the main analysis that vaccines are probably preventing excess deaths among Democrats and not among Republicans.

0:28:18 Thomas: This be because vaccines are a magic, wonder, science, impossible miracle that it's amazing that we got so fast?

0:28:25 Lindsey: -huh, I wouldn't rule that out. Yeah. Yeah, So it's much easier to infer that vaccines vaccines are magical, and that that's responsible for this gap that we're seeing between Republicans and Democrats, if we see a difference here after of vaccines that isn't there before the introductionof vaccines.

0:28:26 Thomas: Okay. But that would mean that Brett Weinstein is an idiot and I just don't know if I can accept that.

0:28:46 Lindsey: I don't know if I'm going to be able to go on either, right?

0:28:47 Thomas: Yeah. Yeah, could I even sleep if that's true? Wow.

0:28:50 Lindsey: Yeah. I mean, can we sleep anyway? So are you going to do?

0:28:53 Thomas: Not with Arlo in my house. Can't do it.

0:28:57 Lindsey: All first thing to say is that there is a very strong negative correlation, even before vaccines became available, between the percentage of a county that was eventually vaccinated by June 2021 and excess COVID deaths. So even prior to vaccines being there, counties that eventually had really low vaccine uptake had higher excess deaths than counties with high vaccine uptake. And this suggests that vaccine percentage is associated with other pre -existing county differences that were relevant to COVID mortality.

0:29:26 Thomas: And those could be lots of things, right? That could be the the measures we were talking about maybe, right?

0:29:29 Lindsey: Yes, exactly. Right. Yeah. You mentioned health care availability and things like that. Yes. So it could be things likethat.

0:29:34 Thomas: Be that. Yeah. Is there any drilling down into what that is or is that beyond the scope here?

0:29:36 Lindsey: Yeah, it's not something they really discuss in this report. No, but I think that that's the very clear sort of implication of of this relationship. Relationship?

0:29:42 Thomas: Right. Yeah, because could say the possibilities being pretty different there. Like I could see it and correct me if I'm misunderstanding the data, but I would see a real difference if it's like, well, yeah, those people have way less access to health care. And so they're not as well informed about the science or they weren't able to get the vaccine as much. Or I could see it being like, well, having gotten vaccinated later correlates really highly with having taken it seriously and done measures and masked up and social distance. And those are kind of two different things. It'd be interesting to know which might be more responsible.

0:30:19 Lindsey: Yeah, well, let me tell you about like the full findings here. And because I do think that it speaks to what you're saying a little bit.

0:30:23 Thomas: Sure. Sorry, I thought you did. My

0:30:26 Lindsey: No, Yeah, you two sort of panels here, right? Right. One is looking at the relationship between vaccinated percentage and excess deaths for Republicans and Democrats before vaccines are introduced. The second panel is that same relationship, but after vaccines are produced and we're going to see if there's a difference there. So What we're seeing before the vaccines are introduce is that there's are really strong negative relationship between percentage of each county that's vaccinated and what their rate of excess deaths are. And that, I think, is speaking to these kinds of differences in like health care availability and things like that. Right. The striking difference in their findings pre and post vaccines is that before vaccines are introduced, the relationship between percentage of vaccinated residents and excess COVID deaths looks absolutely no different among Republicans and Democrats.

They're right on top of each other. So affiliation made no difference here for both Democrats and Republicans alike. They're both dying, you know, around the same rate in to how vaccinated the county eventually became. So it's showing that that environment, whatever's going on in that county that's less versus more vaccinated is affecting Democrats and Republicans equally. However, after the introduction of vaccines, you see the following differences according to party affiliation. So first, excess deaths among Democrats are overall lower than Republicans.

We already knew that from the main analyses. And the excess death rates among Democrats aren't as strongly correlated with the percentage of the county that is vaccinated, as you see in Republicans.

0:30:44 Thomas: Right.

0:31:59 Thomas: Interesting.

0:31:59 Lindsey: So in other words, excess deaths among Democrats are pretty low, regardless of whether they live in a highly vaccinated county or not.

0:32:07 Thomas: Interesting.

0:32:07 Lindsey: They're So, like, they're slightly higher in counties with low vaccination rates, but really not by much. It's almost a flat line there. By contrast, among Republicans, excess death rates in each county are still strongly correlated with the percentage of the county that is vaccinated. In other words, the relationship between excess death rate and percent of vaccinated residents looks about the same among Republicans after the vaccine as it did before the vaccine. So there are way higher excess death rates among Republicans in counties with low vaccination rates. In counties with high vaccination rates, Republican excess death rate rates are essentially no different from those of Democrats. So I think the reasonable inference there is probably clear. Democrats are not suffering as much in counties with low vaccination rates because they're vaccinated.

0:32:08 Thomas: Interesting.

0:32:53 Thomas: They got vaccinated.

0:32:55 Lindsey: Republicans, on The other hand, are dying more if they live in those red unvaccinated areas because transmission is higher because no one's vaccinated, but they are strongly benefiting from herd immunity in the highly vaccinated areas. And they're probably saying to themselves what you were saying your family was saying, which is like, look, you not vaccinated and I'm still not dead two years into the pandemic, right? It's like, yeah, because everybody around you was vaccinated. Good for you. So bottom line of this final analysis is that the pattern of excess deaths changed favorably for Democrats following the introduction of the vaccines in a way that suggests the Democrats were being protected by the vaccines. Shocker. Yeah.

0:32:56 Thomas: It just sucks that like all this amazing science to be like. And so, as you can see, one plus one seems to equal two.

0:33:43 Lindsey: Yep. Yep. Pretty cool. Yeah, don't think I need to tell SIO listeners to get vaccinated. They They've already done that. But, like, if anybody is, like... getting a little comfortable with... Like, get the boosters. Like, seriously, get

0:33:56 Thomas: Yeah, yeah, yeah. Good call. Good imagine there's... there's plenty of people that haven't gotten the boosters. I think I saw something about that, actually, that we need to be on that as as science -minded people. Get your boosters immediately. Mean, just as an example, like I'm way more ethically, you Way more moral than Lindsay because I got it like two weeks sooner or so. So be like me, you Know, it's been great.

0:34:00 Lindsey: Yeah. Mm -hmm.

0:34:15 Lindsey: We already knew that. I

0:34:17 Thomas: Yeah. I'm on a pedestal over here. Lindsay's practically going to hell. It doesn't exist, but if it did, it would be for people who got their booster two weeks later than me.

0:34:22 Lindsey: For you. Am going to hell. That's... Even though... Yeah, Yeah, but I'm still going. True. But, yeah, no, seriously. Get The Boosers. And hey, you know, the Omicron, I had really bad symptoms with both my initial vaccination and with my first with booster. The Omicran booster was fine.

0:34:41 Thomas: Yeah, it wasn't much for me either. There you go.

0:34:43 Lindsey: Yeah. Highly recommend that you not become an excess death. Do some boosters.

0:34:47 Thomas: Yeah. Be a normal death. The correct amount, everybody.

0:34:49 Lindsey: Yeah. Oh, yeah. So, yes. So just to rehash the take home points, let's drive these home. So their analysis allowed them to establish that at an individual level, partisanship is associated with excess death rates after vaccines became widely available for COVID -19. We had group level data suggesting this association. It wasn't necessarily clear what was driving that or, know, that Republicans were necessarily dying at higher rates than Democrats. This allows us to make that inference because Is. It is, I think. And they also did the supplemental analysis, which I really liked, with the vaccination percentage in each county that strongly suggests that this in vaccine uptake is probably the mechanism responsible for the ever-increasing gap in excess deaths between Republicans and Democrats, right? Of results is highly suggestive of that conclusion. I think this was a really solid science thingy, for

0:35:19 Thomas: Crystal clear.

0:35:43 Thomas: Absolutely. This is like one of the science thingies in terms of how clear it is that we've had on this show. I mean this is just And those numbers, again, I mean, those are major numbers. It would be one thing if it's like, yeah, you know, 100 more people in the country died because of, you know, but this this is a lot of people. Preventable deaths basically died of Trumpism.

0:36:06 Thomas: I mean it's it's and you know it's also interesting. I think I think Republicans absolutely know this. Republican leaders know this and you saw them try. You saw them trying to get to start flipping that message a little bit. And you remember like Trump in a couple speeches was like, ah, but you know, the vaccine's a hoax, but definitely get it. I got it. It's definitely, remember that? And then the crowd is like, boo. And he's like, yeah, nevermind. You know, he tried a little bit because I'm sure, you the evil, the competent evil people behind the scenes were like, we're killing a lot of voters that need to.

0:35:47 Lindsey: Yeah. Yes. Yeah, mm- -hmm. Mm- -hm. Yeah.

0:36:39 Lindsey: Your Voter Base That's

0:36:40 Thomas: We need. Yeah, it's not great. We've we just run the numbers and it's actually not good if you kill your own voters in large numbers. Weird. So can we try to get the vaccine on stage or whatever the hell it was on that point, you've lost control, though, you

0:36:40 Lindsey: Right. Yeah. Oh, Yeah. It was way too late at that point, yeah.

0:36:55 Thomas: Really interesting. And to me, it raises the question of who was in control to begin with. You know, it's a complicated thing. I don't know. Like, would there be without Trump? Would there have still been how much of this? That's not important to this study, by the way. But I'm just curious, like, as a As a curious, like, political question, um, don't know, like, I'm not sure what who's driving whom more, but the point is, by by now, I mean, you've lost it, like, you can't just decide, oh wait, this is killing people, well, I'll just undo all the demagoguing I did, like, instantly. Wow.

0:37:11 Lindsey: Yeah.

0:37:27 Lindsey: No, that's exactly right, um, and and speak to to, like, just the volume here, I looked up estimates from the, this is from the Brown School of Public Health, and estimates suggest that over 300,000 deaths have occurred because people who had available to them have refused to get them. 300,000 deaths. And from data, like, we have from this study, we know that those deaths are heavily concentrated in red areas. For Ohio and Florida specifically, they estimate that we're talking unnecessary deaths due to unvaccinated status on the order of like 30,000 to 50,000 people in each state, a lot of people.

0:38:03 Thomas: Do We need to put that in terms of 9-11s for anybody to understand, that's Like a lot, lot of nine, it's like 20 9-11s, yeah.

0:38:07 Lindsey: Yeah. But many dead people that didn't have to be dead because these cynical

0:38:14 Thomas: And The way, that's not the extent of the dead people.

0:38:19 Lindsey: No.

0:38:20 Thomas: It'd be like if on 9-11, no, I shouldn't Get that, a Hundred thousand people died, but, like, then a bunch more people didn't believe that the towers were on fire And stayed in and you're like, yeah, Also That many more people never should have died at all, in any way. It

0:38:23 Lindsey: Oh, yeah, no. Therefore died. Yeah, -hmm. I nope, actually, like, that that analogy quite A bit, I think, that that Does Make The point, yeah.

0:38:40 Thomas: Doesn't Mean That The only Yeah, you know, like, it's just so crazy, the numbers we're looking at, like, we've just, I've said this a hundred times in the last two, three years, whatever it's been, we have no, we're unable to fathom the scale of death and of tragedy of COVID, I really think, like, it's So hard psychologically, to the point where even these excess death numbers, if you, if you could imagine pre-COVID some news report that's like, hey, 50,000 people just fell over dead today, day, you'd be like, oh my fucking god, Like, what, just for no reason, like, maybe a biological weapon attack, like, whatever, it would, would have had World War Three over that if it was a foreign enemy, and that's just excess deaths, That's not even all the deaths.

0:38:56 Lindsey: Yeah. Yeah.

0:39:21 Lindsey: That's That's Right, that's just the deaths from people who didn't get vaccinated after they could, right? That's it, that doesn't count the just hundreds of thousands of deaths prior to the vaccines being available, um, crazy, and yes, the thing to center in those numbers is, is the human tragedy that is at the center of this. So many people have lost family members. So many People have suffered tremendously because of the lies that Republican leaders told. That is awful. Really bad about the following thought that I'm about to share. I'm ashamed of myself. I'm not proud of it. Again, these numbers. These are reflections of actual human suffering.

And that's that is terrible. Like, a ton of compassion for the Republican voters that just trusted their leaders. And when they told them this wasn't a big deal and that vaccines were not necessary, they believed them. Like, That fucking sucks, do feel compassion for that. With that said, I cannot help but feel just slightly satisfied because I know that the motivation for telling those lies in the first place had to do with maintaining political power. And at at the the rates that we are talking about with these excess deaths, like, I don't think it's inconceivable that in places where you have tight races, often Republican and Democrats, that this makes a difference.

0:39:23 Thomas: Oh, Yeah. Yeah. Yeah. Yeah. Yeah.

0:40:41 Lindsey: Red That hour areas is probably not going to matter. And super blue areas probably not gonna matter. But anywhere that's mixed, I think that they might have fucked themselves out of like, holding on to seats.

0:40:51 Thomas: Let's hope so.

0:40:52 Lindsey: Again, I feel gross for going there on the basis of data like this. My ideal self. Yeah.

0:40:58 Thomas: It's, yeah, no, it, it I mean, obviously on some level, no one's in control of everything, anything. And I get that, like, it's hard to blame in a typical, like no one has free will way. It'S, ought to be compassionate to everybody and empathetic to everybody. But least, I at who trusted science and, you people who are who vote in leaders that don't try to lie and kill people, at least they didn't die more. That's like the one bright side. That's nice that like the People Who Did Those Things didn't suffer unnecessary deaths.

0:41:28 Lindsey: Yeah. I that framing.

0:41:34 Thomas: Shouldn't say unnecessary because there's a lot of unnecessary deaths, but didn't didn't offer more excess deaths, I guess.

0:41:37 Lindsey: Yeah. Right. There was some positive consequence for being lucky enough to be in that position of

0:41:39 Thomas: Yeah. Trust science, everybody. So frustrating. I, you can come up with some examples from history when science went pretty bad, but like it doesn't strike me as that hard to separate some of those like key failures from what's going on now and and from advice that's clearly good on an individual level about vaccines. Like it's just, it's, I don't want to hear another anti-vaxxer be like, "But Tuskegee."

0:42:12 Lindsey: Yeah?

0:42:12 Thomas: You're like, okay, that's not, I'm sorry. I get, I get that as a reason people mistrust. I But like, if I'm talking to some fucking white liberal from California and they're like Tuskeegee, I'M like, all right, shut up. That's not, yeah.

0:42:16 Lindsey: Sure. Yeah 100%. Yeah, I think the point of that, of Tuskegee, was that, yeah, middle class white Americans are definitely targeted for, yeah. That was the point, right? That's what happened in Tuskeegee?

0:42:34 Thomas: Yeah. And I shouldn't have said liberal. I I mean, I, but, I'm in California, I do run into some, uh, anti-vaxxers that are not conservative, you know, and, and they do exist. So that was what jumped to my mind, but yeah. This is, uh this is horrible. Um, it is something that I want to literally, I might try to actually send this to my mom. I'm curious. Like, this to my mom and said, hey, remember how, you know, me and our members of our conservative family disagreed on the science? Might be evidence that's relevant to that question of maybe who had the right take there.

0:42:42 Lindsey: Yeah? Hmm. That, I would be super interested in a follow-up about how that goes. Would like to think that this would be persuasive to people.

0:43:16 Thomas: Would like you to imagine whatever the least satisfying, you now, result would be and imagine that's it. And the least satisfying result isn't just like no no no I can't see it, I can see the least satisfying result is like, oh yeah, maybe that's true, anyway, um, Thanksgiving, do you guys want, you know, like that it'll be something like that.

0:43:22 Lindsey: Yeah. Yeah.

0:43:35 Thomas: You know, just just completely disregard or what, avoiding, you know, contending with it, that what I get often with with these these things, yeah, change of subject, um, would be more, and satisfying is not the right word, but it would feel better to at least be like disagreeing in a way that means they're committed to like actually trying to figure out the truth.

0:43:43 Lindsey: I'm very familiar with this dynamic, yeah.

0:43:57 Thomas: That would be better, yeah, well, fantastic science thingy, we give it uh several palps up.

0:43:57 Lindsey: Yeah, yeah, agree. I tend to veer toward engagement myself, but yeah. Yeah.

0:44:06 Thomas: You news about the spider palps, um, they believe in science. So there weren't a bunch of excess deaths in the spider palp community, the spider community. So the palps are still palping full so the palp system rates this as two spider palps way up, I think, right?

0:44:12 Lindsey: Yes. Palps are notorious for their rigorous science education, I've noticed, yeah?

0:44:23 Lindsey: Yeah. We need to figure out the scale. I don't like only having zero palps to two palps. It's not granular enough.

0:44:29 Thomas: Oh, that's true. That's, yeah, no, you're right. We need like, we need like hundreds of palps, like hundreds. Like if there's a hundred palps which means what, 50 spiders? Is that what we determined?

0:44:36 Lindsey: Gross. So many spiders, yeah!

0:44:41 Thomas: A couple for one for backup emergency purposes. So if there's 50 spiders in a room and they're raiding the science thingy, how do you think how do they think their palps would find it?

0:44:42 Lindsey: Mm -hmm. Yeah, and Also it's complicated, it's a complicated metaphor because it's like, the good thing is if both the palps break off, like if you don't lose both your palps, it's like a unsatisfying spider sexual experience, I think, so It's Very hard to use the scale, I don't know, we're gonna Have to work on It

0:45:06 Thomas: Yeah, a catch 22. All right. We'll just work on a new scale. That scale gets two palps way down. So we'll work on another one. All right, thanks so much for listening. And again, reminder, please, please support the show. patreon.com/seriouspod so that Lindsay can afford to still research stuff like this full time for us. Give us this.

0:45:11 Lindsey: But Got Free Ramen Noodles in the laundry room the other day, so it was pretty cool, apropos

0:45:35 Thomas: What does that mean? So like, OK, one of you doesn't have to pledge now, but everybody else does. I don't know what the what does that means? So if you were going to play, if one of us was going to pledge, pledge your pledge minus one cup of noodles dollar amount. Is that the point here?

0:45:36 Lindsey: Of Nothing, yeah. It Yep, yeah.

0:45:54 Thomas: No, pledge the full amount. She needs more ramen noodles, let's face it. Right. Thanks so much for listening, everybody. Vaccinated. Get your boosters, please. And continue to trust the overwhelming scientific consensus.

0:45:56 Lindsey: Indeed.

0:46:15 Clip: Thank you so much for listening to Serious Inquiries Only. If you like the show, the absolute best thing you can do is support us at patreon.com/seriouspod. The second best thing you can do, is please share it on social media or by word of mouth. Speaking of social media, follow me on Twitter at Serious Pod. If would like to be Facebook friends, send me a request at facebook.com Thomas podcast. If not up for that level of intimacy, that's fine. Maybe you want to join the discussion at facebook.com slash groups slash SIO members. Thank you again for listening. We'll see you next time.