Maintenance Phase - November 16, 2021


Is Being Fat Bad For You?

Topics
Diet Book Deep Dive: Angela Lansbury's "Positive Moves" Vibrators

Episode Stats


Length

1 hour and 25 minutes

Words per minute

181.97

Word count

15,488

Sentence count

685

Harmful content

Misogyny

15

sentences flagged

Toxicity

179

sentences flagged

Hate speech

64

sentences flagged


Transcript

Transcript generated with Whisper (turbo).
Misogyny classifications generated with MilaNLProc/bert-base-uncased-ear-misogyny .
Toxicity classifications generated with s-nlp/roberta_toxicity_classifier .
Hate speech classifications generated with facebook/roberta-hate-speech-dynabench-r4-target .
Topics generated with Qwen2.5-3B-Instruct.
00:00:00.000 hi everybody and welcome to maintenance phase the podcast that's just concerned about your health
00:00:15.460 oh my god we're in your mentions on instagram you're glorifying it
00:00:22.140 that one really lived or died by the line reading and i feel pleased with how it turned out if i'm
00:00:27.680 honest. I'm Aubrey Gordon. I'm Michael Hobbs. This month, by astonishingly popular request,
00:00:35.740 our bonus episode for Patreon patrons at every level is the HBO Max documentary miniseries,
00:00:43.100 The Way Down. Yes, I have been told that I have very disappointing opinions on this matter.
00:00:47.260 So enjoy those. And today we are talking about is being fat bad for you?
00:00:56.800 I'm both very excited for this to get the Michael Hobbs, like terrier digging through a million facts kind of treatment. I am also nervous about it because this is the kind of thing that people with profound anti-fat bias use against fat people all the time.
00:01:17.860 Yeah. 0.99
00:01:18.120 Anytime you open up this conversation, it's like opening up Pandora's fucking box for fat people. 0.98
00:01:24.940 This like this little intro is so telling, I think, because the way that fat people and thin people experience this conversation is worlds apart. 0.99
00:01:34.320 Yes.
00:01:34.680 Thin people experience this conversation as like, well, you know, I read in Newsweek the other day that because everybody's fat, we're all going to live, you know, 13 years less than we thought we would.
00:01:43.080 And fat people experience this as like, you are going to die. 0.94
00:01:46.600 Yes. And I would say fat people experience it as you are going to die because often thin people who read those Newsweek articles then turn around and tell fat people you are going to die. 0.86
00:01:56.660 When a thin person decides to tell a fat person we're going to live 13 years less or whatever because of you, because you're fat, whether or not that thin person intends to do so, the idea that that introduces is I'm superior to you.
00:02:11.380 Yes. Because I have figured out how not to do this thing and you are failing because you have
00:02:16.080 not figured out how to do this thing. Right. Anti-fat bias has, I would argue, a more decisive
00:02:21.280 impact on fat people's health than actual just physical body weight.
00:02:26.860 Than adipose tissue. Correct.
00:02:28.520 A term that I have read far too many times this week. Yes.
00:02:32.620 All right. This is so many caveats.
00:02:35.020 I know. Let's keep caveating. Just never, never start.
00:02:38.100 Let's never start and only be like, one more thing before we get into it.
00:02:42.940 My friend has this joke that every left-wing podcast is just people saying,
00:02:46.840 I'm just going to zoom out for a second over and over again.
00:02:49.600 And that's what we are doing.
00:02:51.580 Oh, no, that's me.
00:02:53.300 To me, I think a really important – god, I'm doing it, aren't I?
00:02:58.300 Contextualization thing about this issue is people – most of the academic literature treats this as some sort of footnote.
00:03:06.200 But to me, it's very important to acknowledge that stigma against fat people and the belief
00:03:12.060 that fat people are unhealthy long predates any science showing that.
00:03:17.700 And there's this really good book called Fat, A Cultural History of Obesity, where they
00:03:21.380 identify the first reference of an obesity epidemic was in 1620.
00:03:26.740 You know, the Catholic Church invented the seven deadly sins and one of them was gluttony.
00:03:31.180 And sloth. 0.82
00:03:31.900 Well, exactly.
00:03:32.500 Yes.
00:03:32.980 Yeah.
00:03:33.160 So like there's always been this moral component and there's always been this outgroup component.
00:03:38.860 So something I didn't know until I read this book was that diabetes used to be known as the Jewish disease.
00:03:45.320 What?
00:03:45.560 The idea was that there was something about Jewishness that gave people diabetes. 0.60
00:03:49.760 It was like a genetic predisposition, basically. 0.99
00:03:52.280 Oh, boy.
00:03:52.760 So this has always been associated with like this is a group that I do not like and I'm going to assign to them like some sort of health status.
00:04:00.060 Right.
00:04:00.360 I mean, if we think back to the BMI episode, right, the earliest codification of the BMI was the fattest 15% of people would be considered overweight, right?
00:04:11.280 It didn't have anything to do with health risks.
00:04:14.000 It didn't have anything to do with anything but the fattest among us need to be defined in this way, right?
00:04:21.080 Then you create a bunch of funding streams to find out why it's so unhealthy and so terrible to be what we already think of as terrible.
00:04:27.680 Right. 0.80
00:04:27.780 You know, it can sound a little bit conspiratorial to be like, everybody hates fat people.
00:04:31.800 And like, that's why all the science says this.
00:04:33.620 And so I do think that that's overly simplistic, right?
00:04:36.240 But I also think that it's like, it's also overly simplistic to think that this has no
00:04:40.420 effect.
00:04:41.100 Yeah.
00:04:41.480 So the earliest studies of like, what does it do to your health to be a fat person?
00:04:45.640 I'm going to skip a lot of this because we talked about it in our BMI and obesity epidemic
00:04:49.580 episodes.
00:04:50.380 But the only data that was available was from life insurance companies.
00:04:55.060 Right.
00:04:55.460 Right. This is the 40s, 50s kind of era. Yeah. Yes. So, like, think about who had excess income in the 40s and 50s. Yeah. To spend on life insurance. It's going to be disproportionately white, overwhelmingly male. Yeah. Middle class, upper middle class and straight up wealthy. Right. Like there's like a whole wide, wide range of people that you're missing.
00:05:18.900 Exactly. So basically, the big innovation at this time and the closest thing we get to reliable information on fat people and health is the rise of the cohort study. So are you familiar with this?
00:05:31.960 Uh, no. I mean, like, I know cohort studies like exist, but I don't know about the rise of them.
00:05:38.560 I'm not fucking with you. 0.95
00:05:39.960 I don't know. I feel like I'm being trapped somehow. 0.99
00:05:42.440 so basically everyone knows these life insurance tables are bullshit for all the reasons you just 0.51
00:05:51.240 said and so what they start doing in the 1940s and the 1950s is they start getting these large 0.90
00:05:57.140 groups of people and they get like a representative sample of the country so they start getting
00:06:02.440 together like an entire small town you know one of the most famous ones is like a bunch of nurses
00:06:07.240 It's like 70,000 nurses, huge numbers.
00:06:11.400 And then you track the same people over time.
00:06:14.200 Yeah.
00:06:14.300 I mean, it's essentially like the seven up series.
00:06:17.440 Yes, exactly.
00:06:18.560 Science, right?
00:06:19.260 Yes.
00:06:19.720 That's like, we're checking in periodically.
00:06:21.360 We want to sort of see how you're doing on these specific sort of like sets of measures.
00:06:26.080 Yes.
00:06:26.500 After a couple of years, after longer periods, I mean, some of these cohort studies have
00:06:30.460 been going for like decades at this point.
00:06:32.280 So once you wait long enough, people start dying.
00:06:35.480 And then you can get into all of this risk relative, you know, mortality rates rising kind of thing of like, okay, so why if we would expect 25 deaths, why are there 50 deaths among fat people?
00:06:47.840 Like that is how you can start doing these gradations and figuring out like what is killing Americans.
00:06:53.020 And so the sort of the central issue with calculating these mortality rates is you can't just look at the raw numbers, right?
00:07:00.680 So if you say, like, okay, if we look at normal people, 500 of them died.
00:07:05.200 But then if we look at smokers, 1,000 of them died, right?
00:07:08.580 So it's like, oh, okay, so smokers are twice as likely to die.
00:07:11.560 But that's too simple, right?
00:07:13.360 Because smokers are not the same as non-smokers.
00:07:16.920 Smokers are much more likely to be poor.
00:07:19.120 They're much more likely to be uneducated.
00:07:21.120 They're more likely to be unemployed.
00:07:22.800 They're slightly older than the rest of the population.
00:07:24.500 There's all these other factors.
00:07:25.860 So you have to sort of adjust so that you're comparing like poor, uneducated smokers to poor, uneducated nonsmokers.
00:07:35.200 And then you can say, ah, OK, so this is what accounts for the smoking.
00:07:39.700 Right. And if you don't do some level of statistical control, then you do what your brain is sort of wired to do, which is fill in the blanks with what you think you know, which are often not rooted in data or science or anything.
00:07:55.040 Exactly. So what you just said is basically the Rosetta Stone for this entire episode. Because if you look at the raw numbers, it's true that very fat people have shorter lifespans than thin people. It's complicated for the lower weights and there's a lot of scientific debate about sort of when that association kicks in and we'll get into it. But that correlation, that statement that very fat people are more likely to get heart disease and diabetes and everything you've read about in a million Newsweek articles, that's never really been debated.
00:08:23.500 The debate is about what the association means because obesity is not the only thing
00:08:30.340 that's associated with having a shorter lifespan. Catholics live two years longer than evangelicals
00:08:37.320 and Jewish people live two years longer than Catholics. People who live in West Virginia 0.74
00:08:42.560 die six years younger than people who live in California. One of the weirdest statistics I
00:08:47.960 came across is that people with master's degrees have 10% higher mortality rates than people with
00:08:54.740 PhDs. What? So next time you see somebody posting about like, hey, I got my master's degree on
00:08:59.620 Instagram, go into their mentions like, you're glorifying having a master's degree. I'm just
00:09:04.840 concerned about your health. I know. Quit your master's program. Another really consistent one
00:09:10.040 is that unmarried people have between two and two and a half times the mortality rate of married
00:09:16.640 people. What I love about the unmarried statistic is that when you tell people, okay, unmarried 0.84
00:09:22.700 people are more likely to die, you can like see the gears in their brain start working. People 0.99
00:09:27.800 are like, oh, it's probably something to do with like if you fall down the stairs, if you're a
00:09:31.980 married person, like someone is there to call 911 or maybe somebody's going to hassle you about
00:09:36.960 getting a doctor's appointment for like the lump in your chest if you're married, whereas if you're
00:09:41.140 not, you might just like leave it for another year. Like people will start speculating about
00:09:45.260 what the sort of the real reason is for this association because we all on a gut level
00:09:50.060 understand that like having a piece of paper the marriage license is not extending your lifespan 0.99
00:09:56.300 this is also how you get that fucking garbage shit that's like it'll show up in like people 1.00
00:10:01.160 magazine it's like a pullout box that's like dog owners live five years longer than people who don't 1.00
00:10:06.560 own dogs people who own subcompacts have longer happier lives than people who drive suvs like
00:10:14.460 all of that utter garbage comes from either findings that are not great or people not
00:10:23.040 digging far enough into the findings before translating them into popular media and then 0.99
00:10:27.840 popular media running with that shit exactly no one looks at all these other associations 0.94
00:10:33.340 and immediately leaps to oh this is how we have to fix it like nobody is saying that like no matter 0.99
00:10:39.900 what you go into the doctor for, they should be like, uh, have you considered converting to
00:10:44.960 Judaism? Yeah. This isn't actually a difficult concept for people. What's difficult is that we
00:10:51.460 have been told this extremely simple story about obesity for so long that it's really hard to look
00:10:58.120 at it any other way. And the thing you have to constantly remind people of is that obesity is
00:11:03.260 not a behavior. Obesity is a characteristic. People become fat for all kinds of reasons,
00:11:09.040 and people live their lives as fat people with all kinds of habits. There are fat people who
00:11:14.780 don't get any exercise. There's fat people who get tons of exercise. There's skinny people who
00:11:18.840 get no exercise. There's skinny people who get tons of exercise. These are all overlapping circles.
00:11:24.120 And so the entire debate over obesity is where on the spectrum does it fall? So is obesity more
00:11:30.900 like smoking in that it's a behavior that we can very clearly link to causing all of these health
00:11:36.100 problems and shortening your life? Or is it more like having a master's degree? Where it's like,
00:11:41.120 okay, it's basically a cluster of other correlations. Yeah. I mean, first of all,
00:11:46.260 this is about to be maybe our methodology queeniest episode. I'm getting the sense.
00:11:51.400 Methodology Empress. Methodology Galactic Senate.
00:11:54.860 This is the project of creating and reinforcing an outgroup status. Yeah.
00:11:59.740 And making people believe that it is objective, that it is scientific, 0.97
00:12:03.060 that it is beyond reproach and that it is natural for you to feel like fat people are lazy and gross
00:12:08.600 because that's what they are because science tells you they are right like it's just bizarre that we
00:12:13.400 try and take this stuff in a vacuum and just parrot it out as gospel this is this is the
00:12:18.820 hardest thing about this is because you have to keep in mind issues outside of the data yeah so
00:12:23.880 one of the most famous findings from these is in the 1990s the journal of the american medical
00:12:29.120 association published a study that said that left-handed people die nine years younger than
00:12:35.620 right-handed people what this was actually something that i was told growing up because
00:12:39.840 i'm left-handed and like my whole adult life i've been thinking about like why like why why am i
00:12:45.940 gonna die so young there's these like crackpot theories about like driving what because like
00:12:50.720 the stick shift is with your right hand or something but also who drives stick anymore
00:12:55.180 i know like it makes no sense i don't know man but so what it actually turns out is essentially
00:13:00.000 the researchers looked at the population of 70 year olds and they said okay well there's you
00:13:05.380 know roughly 12 of the population is left-handed and then they look at 70 year olds and they're 0.81
00:13:10.000 like oh my fucking god only three percent of these people are left-handed so all of the left-handed 0.97
00:13:16.140 people must have died it's like they're dropping like flies and we're only left with this much 0.99
00:13:21.000 smaller population. But of course, these are like some of the most scathing letters to the editor
00:13:25.960 of a medical journal that I've ever read. All of these people write in and say, if you're looking
00:13:30.840 at a population who's in their 70s in the 1990s, these are people who grew up in the 20s and 30s.
00:13:36.960 And in the 20s and 30s, if you wrote with your left hand, they would make you write with your
00:13:41.200 right hand. So the reason why there are only 3% left handers in that age cohort isn't because
00:13:47.560 there's like a genocide of left-handed people it's because due to other social forces the rate of
00:13:53.520 left-handed people is like artificially low among old people i will tell you this my grandfather who 0.99
00:13:59.040 passed away 15 years ago ish was like an extreme left-handed pride oh yeah the level of vindication 0.64
00:14:09.500 that this guy is getting posthumously is through the roof well look aubrey all hands matter i don't
00:14:15.180 know why he was being so weird about it all hands matter all hands so i mean the left-handed thing
00:14:21.960 i mean first of all i just have kind of a chip on my shoulder as a left-handed person about this
00:14:25.200 but also it's just important that like you can't only look at the data that data about left-handed
00:14:31.400 people you can slice and dice it you can control for smoking status and like you can do all kinds
00:14:36.620 of statistical mumbo jumbo on that data but it's only going to drive you to like a slightly less
00:14:42.960 boneheaded conclusion so this leads us to meet our protagonist for this episode oh we have a
00:14:49.980 protagonist yes a woman named katherine flegal oh sure you know this oh my god mike i'm so excited
00:14:56.960 i forgot about this all right put put us in space and time who is katherine flegal and like what is
00:15:01.880 going on in the late 1990s so katherine flegal is a statistician who works at the cdc and i know
00:15:08.800 that she waited to tell this story until after she was retired yes so katherine fliegel is at
00:15:14.660 this time a statistical researcher for the cdc she got her bachelor's from uc berkeley she did
00:15:21.960 a phd in nutrition at cornell her work has always focused on obesity and she's one of the first
00:15:28.880 people to notice like hey americans are getting fatter like she's one of the first people to
00:15:34.300 start publishing on this. And she was actually in, I believe it was Geneva at that WHO meeting where
00:15:39.680 they changed the BMI categories. Oh, whoa. Holy shit. She is like deep in this field and like
00:15:45.660 one of the earliest people, like just trying to get the message out that like something is changing
00:15:51.320 in the American population and like people are getting fatter. So in the year 2000, she starts
00:15:56.540 working on a paper about obesity and health. She's been to the CDC for a while. There's this kind of
00:16:02.560 long-standing debate about like what does it actually mean for your health to be fat
00:16:06.420 and she decides to start looking into this in like a very concerted way like i'm gonna find
00:16:11.080 the best data i'm gonna like try to really answer this question once and for all what she does not
00:16:15.560 know is elsewhere at the cdc another team is also working on this so in 2004 we get the infamous
00:16:24.260 paper that is called years of life lost to obesity we've talked about we've debunked this like four
00:16:30.640 times on the show already it's basically this big cdc study that makes huge headlines everywhere
00:16:35.120 for saying that obesity causes 365 000 deaths per year yeah it also says that obesity is poised
00:16:43.780 to overtake smoking as the number one cause of death in america this is like a level of deaths
00:16:50.440 that like you would know someone who died just of fat that is a very large number of people to
00:16:57.220 just keel over because of their fatness, which is sort of the impression that certainly that
00:17:02.380 the general public is left with after the reporting on this, right?
00:17:06.800 So this paper comes out, gets a ton of media coverage.
00:17:09.880 Less than a year later, Catherine Flegel puts out her paper.
00:17:14.120 And in her paper, instead of showing that obesity causes 365,000 deaths a year, her
00:17:19.900 paper shows that obesity causes 112,000 deaths, but it also reduces deaths by 86,000 because
00:17:29.280 slightly overweight people are actually less likely to die.
00:17:33.460 Right. This is what they call the obesity paradox, which is my favorite thing to yell about
00:17:38.640 because it's only a fucking paradox if you can't imagine fat people living healthy lives. 0.92
00:17:44.660 Catherine Flegel has some like very salty papers about the obesity paradox. 0.97
00:17:47.700 really she's like it's not really a paradox it's just like maybe it's just not universally bad for
00:17:53.580 you like maybe it's complicated totally it's not really a paradox it's just you assumed this wasn't
00:17:59.700 possible then researchers actually looked into it and it is possible exactly like calm down and
00:18:05.940 sometimes you'll see scientists just kind of rejecting this out of hand they're like oh
00:18:10.160 there's no biological mechanism that would make fat people live a bit longer and it's like really
00:18:15.340 you don't see any reason why people who carry around extra energy on their bodies would live
00:18:22.340 longer from certain conditions? Like, I actually don't find this difficult to believe at all,
00:18:27.320 considering how many diseases cause you to sort of waste away as you get older. It makes sense
00:18:31.720 they would live longer from some of these conditions, and that gives them more time to
00:18:34.860 recover. Yeah. There's also actual studies on this. 80% of deaths in America are of people 70
00:18:41.220 the end above, right? So one of the most common causes of death and disability in old age is
00:18:46.980 falling down and breaking a bone, especially breaking a hip. One third of people who break
00:18:51.520 a hip die within a year. And so this sounds really facile, but fat people just have more
00:18:55.840 padding on themselves. So there's actual studies that find that the skinniest people are the most
00:19:01.980 likely to die after suffering a hip fracture. So once you get into the actual causes of death,
00:19:08.160 And you're actually curious about this phenomenon rather than simply rejecting it.
00:19:13.460 It's like, oh, yeah, it actually makes a lot of sense that under certain circumstances for certain people, being a little bit fatter is actually better for mortality.
00:19:21.760 Yeah.
00:19:21.920 So basically, I mean, the number one finding of her paper is that, like, people in the BMI overweight category are slightly less likely to die.
00:19:29.960 So a little bit of fat has, like, some protective effect on mortality rates.
00:19:34.360 the other big finding is that skinny people are more likely to die so in the fattest category
00:19:41.500 like the obese category she logs 26 000 deaths in the skinniest category she logs 33 000 deaths
00:19:49.640 yep so one of the quotes that goes around about this this is sort of how it ends up in the
00:19:54.000 mainstream media coverage of it is given current government guidelines it appears that the average
00:19:59.240 person is better off being 50 or even 75 pounds overweight than five pounds underweight this is a
00:20:06.280 thing that also gets sort of thorny right it's worth noting very thin people are more likely to
00:20:12.820 die than very fat people and there is no cause for you as a lay person to then start talking to 0.98
00:20:21.380 very fat people or very thin people about how they're gonna fucking die i like it when you 0.99
00:20:26.180 stand up for thin people like it won't someone think of the thin i like it when you say thin 0.97
00:20:32.660 rights there's all kinds of actually cohort studies that show this same pattern of like
00:20:38.320 this weird spike for thin people a little bit reduced mortality for people that are like a
00:20:42.760 little bit overweight and then a higher curve for people that are like fat they call it the u-shaped
00:20:47.940 curve even though it's like more like a nike swoosh but it's an extremely consistent finding
00:20:53.500 in this kind of research. So we now find ourselves in 2005. There's this 2004 paper that finds that
00:21:00.100 like obesity is like really bad for you. Overweight people are going to die. Fat people are like
00:21:04.360 totally going to die. It's just like really, really, really obvious deep line. And we've got 0.95
00:21:08.740 365,000 deaths caused by obesity every year. And we've also got Catherine Flegel's 2005 paper that
00:21:15.380 says that like, eh, it's really not that many people. Like once you subtract the lives that it
00:21:20.420 saves from the lives that it takes it's like 25 000 deaths a year due to obesity and there's this
00:21:26.080 weird thing with like skinny people being less healthy and like it seems actually to be like
00:21:31.000 good to be like five to ten percent overweight like there's now these two papers both of which
00:21:37.120 are from the cdc saying completely different things right and one estimate is 90 lower yes
00:21:44.720 So to this day, this is still framed as like a scientific debate and like two different ways to look at obesity data and like who can say.
00:21:53.800 But the important thing to know about these two estimates is that one of them is wrong.
00:21:58.380 Like wrong.
00:22:00.000 The first thing that happens with that 365,000 estimate is that people look into the numbers and there's errors.
00:22:06.620 Just like straight up transcription errors.
00:22:08.980 They put the deaths in like the wrong years.
00:22:12.060 human error in Microsoft Excel. Also, there's weird methodological stuff.
00:22:19.060 So remember how I said earlier that when you look at tobacco deaths, you can't just count up
00:22:23.420 the smokers that die because you have to control for all this other stuff because it's not a
00:22:27.800 representative sample of the population. What they did in this study, when they say obesity
00:22:32.440 is about to overtake tobacco, they adjusted the tobacco deaths downward because people who smoke
00:22:38.860 are more likely to be poor and like you know we we have to artificially make that number smaller
00:22:42.540 to make it more valid but they didn't do that with obesity deaths right struck by lightning
00:22:47.980 congratulations if you're fat you died of obesity this like this is literally like the sophistication
00:22:52.480 of the analysis basically right it's astonishing that this number clawed its way into the popular 0.99
00:22:59.200 imagination and still gets repeated in non-academic sources like all the fucking time no i know like 0.57
00:23:05.840 Like, I got a fact sheet from a lobbying organization in the last, like, three months 0.93
00:23:11.880 that was like, you know, 360,000 fat people die of obesity every year.
00:23:16.640 And I was like, yes, excuse me, what?
00:23:19.700 The earlier study also, I can't believe they did this.
00:23:22.280 Some of their cohort studies ended in the 1970s.
00:23:25.660 So, like, some of the deaths were in the 1970s, even though this paper is being published
00:23:29.600 in 2004.
00:23:30.240 but like the heart attack cardiovascular death rates in the 1970s were sky high. You're actually
00:23:37.320 much less likely to die of a heart attack as a fat person now than you were as a normal weight
00:23:43.820 person in the 50s and 60s. If you look at the death rates, they've all been declining for years,
00:23:48.140 even as the population has gotten fatter. Right. And a lot has happened in technology
00:23:52.160 and medicine since then. Right. The other problem with these studies is that they're
00:23:56.600 built exclusively around BMI categories. So every single person in these big cohort studies
00:24:03.380 is organized in like, normal way, 25 to 29. Wait, can that be your canonical BMI voice?
00:24:13.300 Every time I say a BMI number, I just like cringe inside. Also, every time I say like obese,
00:24:19.040 I also cringe. Like, I hope that it's clear from my general tone and personality that I'm putting
00:24:23.760 giant quote marks around these terms at all times. Yeah, normal weight. Like, I don't believe in
00:24:29.100 these things. But it's just gonna be much easier if I don't have to say quote, unquote, every single
00:24:33.180 time I use one of these adjectives. So I just like I feel deeply weird about like the way that I'm
00:24:39.140 speaking about this issue right now. I will also say many, many fat people experience the terms
00:24:44.860 quote, unquote, obese and obesity as slurs. Yeah, so bad. I've talked to one fat person ever
00:24:49.800 in my time of doing this project where I've talked to thousands of fat people who are like,
00:24:55.120 I feel neutral about the word obesity. Everyone else has said, I feel deep shame about it. I feel
00:25:01.020 like I'm being judged. I feel horrible. It makes me think about self-harm. I think there's this
00:25:06.380 belief that because it's a medical term, it can't hurt people. Lots and lots of medical terms hurt
00:25:12.740 people. But they're saying person with obesity now. Oh, fuck. Michael, someone corrected me
00:25:19.700 on Twitter and was like, actually, what you should be saying instead of fat person is person
00:25:25.080 with obesity. And I was like, get out of my mentions, you gremlin. The worst is person with 0.89
00:25:31.640 overweight. It's like, what? It's so weird. But so the problem with these categories for mortality
00:25:37.980 research is that a lot of those cohort studies rely on self-reported BMI. So you ask people their
00:25:45.700 weight, you ask people their height, and then you calculate that they're overweight or they're
00:25:49.520 normal weight or whatever. The problem is that when you do this, a huge number of people end up
00:25:55.660 in the wrong categories. Because people, I mean, I don't want to say lie about their weight because
00:26:01.620 a lot of people don't know accurate. I do not know my weight. I have not weighed myself in
00:26:05.960 like five years so like if somebody asked me my weight i would be wrong but then the problem with
00:26:10.540 self-reported data is skinny people will lie a little bit about their weight like if you're like
00:26:15.260 150 you'll say you're like 145 but if you weigh 400 pounds you'll say you're like 325 yeah the
00:26:22.360 larger you are the more weight you will take off but then there's also weird confounders that like
00:26:26.760 skinny men will add weight to themselves they'll say that they're 185 when they're actually 160
00:26:31.560 right what you're telling right now is the story of my first driver's license oh yeah were they
00:26:35.720 where they ask you your weight and they don't weigh you yeah totally and i was like 250 pounds
00:26:39.640 just leave me alone so let's say you have two people and they're both five foot eight and they
00:26:47.020 both say that they're 180 pounds so that's the data in your spreadsheet but in reality one of
00:26:53.040 them is 185 pounds and they're cutting a little bit of their weight off and the other one is 200
00:26:59.100 pounds and they're cutting a little bit more of their weight off in the actual numbers this isn't
00:27:03.620 actually that big of a deal and people who defend self-reported data they'll say like well when
00:27:09.280 people lie about their weight on average they only really lie by like whatever two five ten percent
00:27:14.660 something like that it's not that big of a deal but it's not actually about how many pounds they're
00:27:19.500 cutting off of themselves the problem with those two people is that the cutoff between overweight
00:27:25.000 and obese is 190 pounds so the person who's 200 pounds who says that he's 180 he just jumped from
00:27:33.440 one category to the other. Yeah. People have actually done studies where they compare data
00:27:38.180 that's self-reported and data that's actually measured. And some of these studies, the normal
00:27:43.420 weight participants, 30% of them should have been classified as overweight or obese. Yeah. And so
00:27:49.980 look, the BMI categories are bullshit. And like this, this says nothing about people's individual 0.99
00:27:54.680 health. And like, I don't want to make it sound like I'm giving any credibility to this. Yeah.
00:27:58.280 But the entire purpose of these studies is comparing people in different categories to
00:28:04.660 each other.
00:28:05.460 Right.
00:28:05.760 If you're comparing dogs to cats and you're like, oh, 30% of the cats in our sample are
00:28:12.440 actually dogs.
00:28:13.360 Yeah.
00:28:13.720 And we have no way of knowing which cats those are.
00:28:17.020 Like, which ones are actually dogs in disguise? 0.97
00:28:19.260 Your entire study is garbage because you're not actually comparing different categories. 0.99
00:28:23.920 Totally. 0.90
00:28:24.180 So another reason why Catherine's study is better than the earlier study is that she throws out all of the self-reported data.
00:28:33.000 You can't just mix bad data and good data and then say anything valid about a phenomenon.
00:28:38.540 Right.
00:28:38.840 Again, this is framed as like a scientific debate.
00:28:41.920 Like some people say 365,000 deaths.
00:28:44.480 And then like this wacky Catherine Flegel lady says it's like way smaller.
00:28:48.060 But the CDC corrected the 365,000 number.
00:28:51.620 they put out explicit guidance that they weren't going to use it. So in a press release, they say
00:28:56.400 like, yeah, we're no longer standing by this number. We are standing by Catherine Flegel's
00:28:59.840 work. It's just better data. Right. And she didn't write a bunch of conclusions based on typos.
00:29:06.720 Exactly. That's the thing that is like wild to me about this whole story. It's not like
00:29:12.200 not as methodologically sound. It's like total nonsense. This is like where I think the obesity 0.71
00:29:19.440 the epidemic as a moral panic is really instructive because the pattern that you see in moral panics
00:29:24.020 over and over again is what do we not need evidence to believe? Like what are the overall
00:29:29.140 societal narratives for which we will put aside these like pretty basic methodological
00:29:33.840 considerations, right? We won't apply as much scrutiny. Yeah. The CDC and like, you know,
00:29:39.000 the Journal of the American Medical Association and these like very high level public health
00:29:43.240 institutions were willing to print something that in the methodology says we are assuming
00:29:48.760 all deaths of fat people are because they are fat yeah they were just like yeah sounds good it's
00:29:53.000 astonishing no one caught or seemed concerned with this like very very bald statement of bias
00:30:02.520 like there's no other way to talk about that exactly that a statement of bias right so despite
00:30:07.940 the fact that catherine fliegel's paper is just like better than the previous paper there is huge
00:30:14.340 backlash so this is where we meet the antagonist of the episode are you familiar with somebody
00:30:21.760 named walter willett i mean i wasn't six months ago but i absolutely am now what do you what do
00:30:28.680 you know uh what i know about walter willett is that he worked does he still work at the harvard
00:30:35.340 school of public health he was at this time the chair of the nutrition department at the harvard
00:30:40.000 school of public health yeah and i feel like everything i know about him is a spoiler for
00:30:43.440 this story and through like the weird twitter explosion of people being like oh yeah i work
00:30:50.360 on a different issue and he did this to me too that was a good week on twitter it was a great
00:30:54.820 week on for like for you and me specifically like oh methodology fight you for like a month
00:31:02.940 this dude was my rachel hollis that's what i would say
00:31:05.960 so walker willett is like a extremely prominent figure in public health according to some accounts
00:31:16.800 he's the most cited nutritionist in the world he's also i think this is very important he also
00:31:22.180 came up with that generation of public health people who were like screaming about obesity
00:31:27.140 for years and nobody was listening and then once people started listening i think it left a lot of
00:31:33.460 people in that generation with like kind of a chip on their shoulder of like this is my issue
00:31:37.600 i was talking about this before anybody else was and like they really want to guide like the way
00:31:43.160 that this issue is framed for the public he also has like really bad attitudes about weight and
00:31:49.620 about fat people so during the events that are about to ensue he gives a bunch of interviews
00:31:54.760 you know one of the quotes he says next to whether you smoke the number that stares up at you from
00:31:59.860 your bathroom scale is the most important measure of your health he also has this weird thing he
00:32:06.060 tells npr the weight you should aim for is the weight you were at when you were 20 he says for
00:32:12.900 most people our ideal weight if we weren't seriously overweight is what we weighed when we
00:32:17.800 were 20 so it's like you should aim for what you weighed when you were 20 unless you were fat in
00:32:22.600 which case it should be lower presumably right i was gonna say like if his thing is like aim for
00:32:27.220 the weight you were when you were 20. Done and done. I win. Owned. Same size as when I graduated
00:32:35.900 from high school. Not quite, but close. It sounds like he's a thoughtful researcher in a lot of ways.
00:32:41.920 And when it comes to fat people, his brain goes into the mode of like, I've already decided what
00:32:48.180 I think of fat people. Exactly. So we don't really know the story from Walter's side. But 0.57
00:32:53.460 But from Catherine's side, after her study comes out in 2005, she starts getting calls
00:32:59.160 from journalists who are like, I'm calling to write about the criticism of your work.
00:33:04.340 And she's like, what criticism?
00:33:06.100 They're like, oh, well, you know, I got a call from this guy, Walter Willett at Harvard,
00:33:09.620 and he says you've published this study that's like methodologically flawed.
00:33:13.300 At one point, he calls her research naive, deeply flawed, and seriously misleading.
00:33:20.100 And she's like, what?
00:33:21.500 Yeah, which is like, that's mean.
00:33:23.460 Yeah. Like it feels like a way that professional men talk to professional women to dismiss them.
00:33:30.480 Exactly. Like, oh, that's naive. So I'm going to send you a little excerpt from the article that
00:33:35.680 Catherine Flegel writes about this in this summer, summer of 2021. Okay, great. Quote,
00:33:40.720 almost as soon as our article appeared, a symposium was scheduled at Harvard for the
00:33:44.640 express purpose of criticizing our article. The lineup consisted of a small number of vocal
00:33:50.380 critics, mostly from Harvard itself, all attacking our work and asserting that their previous
00:33:54.880 research somehow showed that our estimates should have been higher, although their previous research
00:34:00.000 had not even addressed the topic of estimating numbers of deaths. The presentations at the
00:34:04.640 symposium did not mention the multiple errors in the 2004 article. One speaker described us as
00:34:10.240 having no biomedical background, even though the four authors of our article were well-published
00:34:15.040 senior scientists all with doctoral degrees in nutrition or statistics and one with a medical
00:34:20.760 degree from harvard medical school jesus christ right seeking to maximize media coverage the
00:34:27.120 organizers arranged for the entire symposium to be webcast live and encouraged reporters to view
00:34:32.280 and report on it what i am learning from this quote is that all these well-published senior
00:34:38.180 scientists are up on their high horses with their longer life expectancy and their doctoral degrees
00:34:43.520 10% longer, man. None of these master's degree, these sickly people with master's degrees.
00:34:49.560 Chumps. 0.97
00:34:52.320 One of the things she mentions in this paper is that like she was invited to attend this 0.78
00:34:57.960 symposium, but was not invited to speak. That is such an aggressive move. 0.86
00:35:02.400 It's such an aggressive move. I also think on some level, like holding a symposium to
00:35:07.260 criticize a piece of work that you don't like, like, I honestly think like some of this is fine.
00:35:12.220 There are elements of this that is okay.
00:35:14.920 But what she finds over and over again is the actual rhetoric that they are putting out and that Walter Willett is putting out is this idea that the data is fundamentally flawed and the methods are fundamentally flawed.
00:35:26.880 So it's not like, well, she made these choices and we would make other choices and these are why we would make these other choices.
00:35:32.700 It's all this weird, bad faith stuff.
00:35:35.220 It also feels like, I'm not talking about this guy in particular, but like, often in the minds of like people with deep seated anti fat bias, any acknowledgement that fatness might not be as horrifically unhealthy as we have been led to believe, translates to, then people will think it's okay to be fat, then people will intentionally get fat or let themselves get fatter.
00:35:59.300 Exactly.
00:35:59.760 Then we'll be overrun with fat people, which is like a nightmare for them. 0.94
00:36:03.080 Exactly.
00:36:03.380 He constantly compares her work to the previous paper that found 365,000 deaths, but he doesn't mention the previous study had been corrected and that it has, like, errors in it.
00:36:15.840 Yeah.
00:36:15.920 He seems to talk about it as, like, this weird Kathy-Flegel joint where, like, she went out of her way to publish this, like, strangely ideological research when she's like, I work at the CDC.
00:36:27.660 After she got her data, she spent four months with the higher-ups checking her data and, like, stress testing it.
00:36:33.540 It feels very much like some interloper just shows up and like figures out that it's like
00:36:39.080 probably not going to play great for him to take aim at the CDC, which is where this research
00:36:45.300 originates. So he instead switches to like an ad hominem approach and is like, it's about this
00:36:52.360 woman. It's also so funny that to people outside of this world, he starts telling them random
00:36:59.080 reporters for like national publications he's like you know the real problem with katherine 0.96
00:37:03.760 fliegel's work is that the bmi is problematic oh fuck off like of course katherine fliegel is like 0.79
00:37:11.780 the other study also used bmi right you have shown no interest in your entire career in like 0.84
00:37:18.740 the ways in which the bmi is problematic like this is not work you have done you've not written about 0.93
00:37:23.040 this you haven't rejected other studies that use bmi because the entire fucking field uses bmi
00:37:27.300 But then I like, I'm the problem for using BMI. And you think we should use this other estimate
00:37:33.340 that also use the BMI, right? It feels like a little bit like the research equivalent of
00:37:38.940 Stockholm syndrome with the BMI, which is like, it's terrible. And it's sort of an open secret
00:37:44.560 that it's terrible for like a bunch of reasons. But also now in order to make research that you
00:37:51.260 get that can be in conversation with other research, it is the way that researchers talk
00:37:56.780 about size and health. It's like the measure that people use. So you can decide to opt out of BMI,
00:38:04.060 but then you're also kind of deciding to opt out of your research being like, considered in concert
00:38:09.460 with other more damaging research, right? Exactly. And like, this goes on for years.
00:38:15.500 Yeah. Blog posts start showing up on like the Harvard website that say that she's been demoted
00:38:22.120 from the cdc and her paper was retracted and she's like what like i won an award like i got
00:38:27.660 everything's fine i go to work here every day and people still like me and work with me what 0.78
00:38:35.720 are you talking about exactly and like fucking i mean there's no evidence whatsoever that walter 0.97
00:38:40.420 had anything to do with this but like someone updates the fucking wikipedia article for obesity 0.98
00:38:44.820 jesus christ it included her article and then someone said that it had been like discredited 0.88
00:38:49.760 what are you talking about there's just like this weird shadow campaign to make it seem as if she's 0.98
00:38:54.880 some like rogue statistician she's like looking around and she's like no everyone thinks this is 0.97
00:38:59.720 fine i get so exhausted by this as a fat person as a woman definitely and as a person who cares 0.99
00:39:06.980 about like having conversations in good faith totally exactly like it's such a fucking soap 1.00
00:39:12.700 opera it's weird hey you know how all you have all this like garbage shit to deal with at your 0.99
00:39:17.420 job and you've got like a co-worker who's like totally unreasonable and weird and like you can't 0.98
00:39:21.940 stand to be around them like so do people at the cdc yeah you're like oh got it the thing that i
00:39:28.440 have been taught to trust uncritically is from like deeply deeply deeply human sources and that
00:39:35.260 feels like yeah the headline of this story to me i mean yeah my headline for this is that academia 0.96
00:39:40.720 is basically the big brother house this little fight's going on it just sucks in there
00:39:46.500 it's the big brother house but like people are constantly running campaigns to get people
00:39:53.580 kicked out of the house but no one ever actually gets kicked out of the house
00:39:57.580 so fast forward to 2013 partly in response to the criticisms of her 2005 article
00:40:08.900 fliegel starts working on a much bigger meta-analysis so originally she just had these
00:40:14.080 like two data sets from american data but what she does is she looks for all other data sets like
00:40:19.880 you know there's there's hundreds of studies going on about like obesity and health at any
00:40:23.720 given time like it's it's kind of absurd you find these like random like norwegian cohort studies
00:40:28.300 and like the south korean nurse collaboration or whatever right india australia all around exactly
00:40:33.820 like this is a question that like the entire field of science wants to answer so katherine
00:40:38.560 fliegel does like a pretty normal meta-analysis where she looks at every single study that's been 0.51
00:40:42.280 published on this. And she publishes a meta-analysis that includes 97 studies and 2.9
00:40:49.260 million people. And she finds the same thing. Again, we have this spike of mortality for the
00:40:55.960 thinnest people. And then you have a reduction in mortality for the overweight people. And then it's
00:41:01.620 higher for fat people. So I interviewed Catherine Flegel for this. And one of the things that she 0.98
00:41:06.820 noticed is that oftentimes in these studies, they'll categorize people as like normal,
00:41:11.480 overweight obese but as we've talked about on the show many times there's actually a huge gradation
00:41:16.800 within fat people like you're talking about people that are like 210 pounds and you're talking about
00:41:21.220 people that are like 600 pounds that's a huge category right and so oftentimes what you find
00:41:25.900 in these studies is they'll say overweight people are fine but like obese people that's where all
00:41:30.340 the risk is but then what catherine notices once you start digging into their data if you break up
00:41:35.020 the fat people into like they call it again these names they call it class one class two and class
00:41:40.340 three obesity, the lowest class of obesity, like the thinnest fat people, oftentimes also don't
00:41:47.800 really have any health risks. So for her previous study, she found that if you're five foot eight,
00:41:52.240 up to 185 pounds, you don't really have any health risk. In this study, which is even bigger and has
00:41:57.700 more data, you can go up to 210 pounds and still not really have any elevated health risk.
00:42:02.680 Right. I mean, this is where it feels worthwhile to insert that until the late 90s, people who are now in class one obesity category were previously in overweight categories and people who were who are now in overweight categories used to be in quote unquote normal weight categories.
00:42:21.560 Oh, it's so fake, dude. It's so fake.
00:42:24.700 It's so fucking fake. 0.99
00:42:26.120 also one of the things that's amazing in because i've read a bunch of these like random ass studies 0.99
00:42:29.340 for this too one of the things that's incredible is they'll mention like almost like offhand they'll
00:42:34.300 be like oh yeah uh bmi for black people just like doesn't work like some of them will be like up to
00:42:38.040 like bmi of 35 so you can weigh like 280 pounds if you're black and like not have any of these 0.99
00:42:42.920 mortality statistics but some of them have also found that like it just doesn't fucking matter 0.97
00:42:46.400 for black people at all and it's again treated as this like weird footnote oh it's a paradox 0.95
00:42:50.500 no reason to dwell on it no reason to put it in the abstract no reason to like write an open letter 0.98
00:42:55.680 anything anyway back to the main anyway white people like jesus fucking christ this is the 0.99
00:43:03.220 fucking chickens coming home to roost when you made a like body measurement slash eventually 0.99
00:43:08.780 like moral assessment tool i know that was only built for white people so basically this massive 1.00
00:43:15.960 meta-analysis finds the same thing that her previous study found and that lots of other
00:43:21.460 studies around the world have found this like basic mechanism. And again, Walter Willett goes
00:43:27.420 on the attack. He shows up on NPR and says, this study is a pile of rubbish and no one should waste 0.98
00:43:33.880 their time reading it. Wait, is that a quote? Oh, yeah. Oh, Jesus Christ. He also tells, I believe 0.95
00:43:39.980 the Atlantic. He's like, Kathy Fleel just doesn't get it. And then Catherine's like, I go by
00:43:45.240 katherine i don't don't let's just keep it katherine walter that's as close as you'll get
00:43:51.820 in like academic research world to like get my name out of your mouth it's fully like smackdown
00:43:59.520 language it's great so speaking of unfair criticism i'm going to be very unfair to
00:44:05.580 walter right now so he gave because he calls all these journalists and like inserts himself in all
00:44:11.780 of these publications about this controversy, there's quotes from him everywhere. And so I made
00:44:16.700 like this little medley of like all of the inappropriate shit that he says about Catherine 0.99
00:44:21.360 during this period. And so I'm going to send it to you to read. Okay, quote, when Willett sees the 0.96
00:44:27.740 CDC study on obesity, his mind flashes to the picture of what he is sure this confusion will
00:44:31.840 lead to. Patients packing on more weight while their doctors become less willing to broach the
00:44:36.840 subject. People will get fatter, die sooner, and all the medical bills could cripple the
00:44:41.460 healthcare system. Willett feels he has no choice but to go on the attack. No choice. It's important
00:44:47.460 to push back strongly against the promotion of ideas and analyses that are based on faulty data,
00:44:52.620 he says. Willett claims that Flegel clearly has a point of view on this issue. After all,
00:44:58.640 she published a paper in 2005 that reached the same conclusion about the protective powers of
00:45:03.340 a few extra pounds. Quote, the Flegel paper was so flawed, so misleading, and so confusing to so
00:45:09.800 many people, we thought it really would be important to dig down more deeply, Willett says.
00:45:15.620 Studies such as Flegals are dangerous, Willett says, because they could confuse the public and 0.77
00:45:20.940 doctors and undermine public policies to curb rising obesity rates. According to Willett, 0.97
00:45:27.580 the papers Flegal published were woefully misleading and undermined dietary guidelines
00:45:32.740 that had been in place for several decades. Worse, he says, these findings can be hijacked
00:45:37.740 by powerful special interest groups such as the soft drink and food lobbies to influence
00:45:42.840 policymakers quote it's basically giving a green light to gaining weight and becoming overweight
00:45:48.120 we're talking about millions of lives lost every year due to obesity it's a death spiral as this
00:45:54.960 gets normalized as we look around and everyone's overweight end quote so other than agreeing
00:46:00.380 wholeheartedly what do you think um i mean this is so like here's what i think about this this is 1.00
00:46:06.100 genuinely the shit that trolls say online all the fucking time it's wild right this is the shit 0.99
00:46:11.220 right yeah and actually like i like i genuinely don't even know where to fucking start with this 0.99
00:46:16.800 like oh no this would undermine public policies to curb rising obesity rates which don't fucking
00:46:22.800 work i know exactly we've been doing for 20 years and have produced nothing but continually rising 0.84
00:46:29.100 obesity rates like oh no they go against dietary guidelines which change every five to ten 0.82
00:46:36.200 fucking years because we know very little about nutrition all told won't someone think of the
00:46:41.860 dietary guidelines everyone the beloved dietary guidelines that all americans know and love and 0.98
00:46:48.680 never change yes i can't jump inside his head i don't know what he's thinking what this communicates
00:46:53.560 to me as a fat person and what i hear with this stuff is like the ultimate terrible thing that
00:46:58.380 will happen as a result of this is that there will be slightly more people who look a little
00:47:02.320 bit more like me. I know. I also think an important thing that he says here is that he says, you know,
00:47:07.200 all this is going to undermine public health. And also it's going to be taken up by powerful
00:47:11.640 lobbies of like the soft drink makers and like the fast food companies. And on some level,
00:47:16.620 he's actually right. There are like really gross corporate lobbying firms, like these fake
00:47:22.480 grassroots companies like that pretend to be grassroots NGOs that like say the same shit 0.97
00:47:26.840 that we do yeah i sort of get that like yeah there's some really gross forces aligned with 0.96
00:47:31.620 like what we are saying on the show like that is worth acknowledging but also his argument is
00:47:35.780 aligned with pharmaceutical companies and the weight loss industry right you're not like the
00:47:40.720 lone david standing up to like the goliath of mcdonald's it's like you're sitting there with
00:47:45.880 novo nordisk dude yeah totally well and also just like listen corporations are going to use 0.96
00:47:52.180 whatever the fuck is in their reach to prove their point. And that can't and shouldn't drive
00:47:58.860 what research tells us, right? So like, his argument, I would say is like, even more dystopian,
00:48:05.200 which is like, we shouldn't research anything that corporations could then turn around and use to
00:48:10.520 defend their own self interest in their own bottom line. That's also not good. That's also
00:48:16.000 corporations exactly an undue influence on science and research that we shouldn't be encouraging
00:48:22.360 dude exactly it just feels like deeply not only half-baked but it's like leading him in like
00:48:27.560 unintentionally a more sinister direction and also i mean when i talked to katherine like i think this
00:48:32.960 was what really bothered her is that he's basically accusing her of like having ideological motives
00:48:38.500 with no evidence right there's one report of someone who was in a class of his and he said
00:48:44.920 that she holds these views she published these studies because she's a little bit plump herself
00:48:49.900 oh get fucked exactly so look this is a person on twitter like this this has not been confirmed 0.94
00:48:55.360 like i i want to be clear about like the source of this information right but what what bothers 0.98
00:49:00.700 me is like even if that isn't true what he's accusing her of is basically being an ideologue 0.99
00:49:05.660 right that like she wants a policy outcome she wants doctors to not talk to their patients and
00:49:11.700 And, like, he is skipping straight from the data to the implications of the data and accusing her of trying to bring about some sort of policy outcome.
00:49:20.700 Right.
00:49:20.840 And Catherine Flegel, like, I don't know much about her.
00:49:23.260 She's actually kept her personal life completely out of the public sphere, which seems totally fine to me.
00:49:28.300 But, like, even in our phone call, she would, like, correct me if I was like, yeah, I feel uncomfortable with, like, doctors talking to patients about their weight.
00:49:35.160 She'd be like, well, I don't know about that.
00:49:37.060 My study is about mortality rates.
00:49:38.720 All I'm talking about is mortality rates.
00:49:40.000 I don't really know what the implication of that is.
00:49:42.740 I don't know what the policy should be.
00:49:44.840 She's always been somebody that's like extremely temperate about everything she says.
00:49:50.220 And like you, I've seen other interviews with her, like you cannot get her to go beyond
00:49:54.880 the data.
00:49:55.480 Right.
00:49:55.620 And then we have Walter Willett, who's written four diet books.
00:49:59.080 Right.
00:49:59.320 He has told reporters from like every outlet in America that he thinks that it is dangerous
00:50:04.280 to be fat.
00:50:04.880 And he thinks that it is irresponsible to publish scientific research that even hints at the idea that it might not be as bad to be fat as he thinks.
00:50:15.420 Right.
00:50:15.580 And it's like, who's acting ideological here?
00:50:17.560 This is also such like every bad faith dude who I've ever worked with, who just sort of throws his weight around. Both this is a totally extraordinary story of science getting hijacked by like weird personal interactions and personal motives.
00:50:36.240 and also it feels totally unremarkable and common i know and like all of us especially
00:50:43.680 any of us who are not white dudes like know this fucking dude dude i know i love that it's like 0.85
00:50:49.300 this is an outrage and also wednesday and also and also staff meeting it's so remarkable and 0.97
00:50:59.020 it's so not all at once exactly and for like all of my close friends and family i can name you
00:51:05.500 three of these dudes in their lives yeah i have known those dudes and i've also been that dude
00:51:09.920 sometimes oh michael i know the bravery i know congratulate me congratulate me for admitting it
00:51:16.700 that's what i want i'm gonna mail you one cookie thank you that's all i want i don't have to
00:51:23.060 actually fix my behavior as long as i acknowledge it it's very good okay are you ready for the
00:51:30.740 methodology queen section of this episode i've never been readier mike so i've been saving this
00:51:36.900 because a lot of the arguments that walter makes against katherine are like technical
00:51:42.100 methodological things about how she has designed this study and i want to take them seriously so
00:51:47.720 what he says in his article that he eventually publishes like responding to her study
00:51:52.200 is he says we believe her study is flawed their normal weight group contains persons who are lean
00:51:59.080 and active heavy smokers frail and elderly and seriously ill with weight loss due to their
00:52:04.940 disease so he's making a specific point about her sampling he also says he also says as well 0.99
00:52:12.240 as asian populations historically undernourished and burdened by infectious diseases what the fuck 0.53
00:52:18.800 because i think she included some data sets from like south korea or japan what are you doing sir 0.56
00:52:23.820 Walter, we're talking about human biology, Walter.
00:52:27.800 And like, it's a normal thing to include studies from other countries when you're doing that.
00:52:32.440 I also like that he was like, my take doesn't seem garbage enough.
00:52:37.160 Put some sprinkles on that. 0.76
00:52:38.320 That has no sprinkles. 0.91
00:52:39.620 Burdened by undernourishment. 1.00
00:52:41.220 Like, what the fuck, man? 1.00
00:52:42.460 So he has a couple, there's a lot we have to unpack here. 1.00
00:52:45.800 The first argument that he levies against Catherine Flegel's work is the problem of smoking.
00:52:52.580 this is a real problem in these studies that like if you leave the smokers in it ends up
00:52:58.420 fucking with your mortality rates all over the place because like smoking is so bad like there's
00:53:02.900 statistical methods for sort of getting rid of smoking and making everybody comparable across 0.99
00:53:07.820 groups but walter thinks that that is not enough so what walter points out is that in the skinny
00:53:14.260 group right the reason why you have this weird spike among skinny people like skinny people have
00:53:21.020 really high mortality. The reason for that is because there's just like way more smokers in
00:53:25.720 that group. What? Because smoking makes you skinny. They're not at higher risk because
00:53:29.620 they're thin. They're at higher risk because they're smokers. So you have to remove all the
00:53:33.920 smokers and then you'll get cleaner data. Oh, Jesus. In studies that he does, he removes all
00:53:41.080 of the current smokers and everyone who has ever smoked. Great. Cool. That's how human health
00:53:48.060 works you could just remove an entire category i i don't know man i it just so yeah a couple
00:53:53.800 problems with this first of all current and former smokers is 40 of the population
00:53:59.260 so you're removing 40 of people from your analysis yeah they don't count there's also
00:54:06.280 this huge problem with the category of former smokers yeah because like my grandma is a former
00:54:11.580 smoker she smoked for like two years yeah when she was i think in her 30s and now she's 97 right
00:54:16.060 But also like there's also former smokers that smoke like a pack a day for like 50 years.
00:54:20.280 So it's like the group of former smokers, it's not clear what effect that is having
00:54:24.180 on their health or their weight, but you're just removing all of them as if like it's
00:54:29.080 so contaminated that you can't look at them at all. 0.51
00:54:31.480 Also, as you have noted, that's like basically like disproportionately removing poor people.
00:54:36.260 Exactly.
00:54:36.620 It's disproportionately removing like a bunch of marginalized communities. 0.71
00:54:40.460 And again, feel like a likely return to centering the bodies of like middle class white people, right? 0.65
00:54:48.060 Like if we're like getting down to it. 0.90
00:54:50.260 Well, this is the problem is that what you end up doing with removing all the smokers is you say that you're removing the effect of smoking, right?
00:54:57.840 You're like you want cleaner data.
00:54:59.120 But what you're actually doing is you're removing a bunch of poor, uninsured, unhealthy, thin people, right?
00:55:07.060 Because smokers are disproportionately thin and you're leaving in all the fat people that
00:55:11.800 have those bad health outcomes, right? 0.60
00:55:13.340 You're just removing all the sick, thin people and leaving the sick, fat people, right? 0.98
00:55:17.280 It's a, it, Oh God, I don't even like, I'm so fucking tired. 0.97
00:55:23.300 I like, I was about to say something about how like, yeah, heaven forbid we have like
00:55:27.720 actionable public health research that works for people who live at or below the poverty
00:55:32.720 line. 0.96
00:55:33.280 But then I was just like, I'm fucking tired. 0.97
00:55:35.420 it's all such a racket. I know. Great. So you're reverse engineering everything. So we're just 0.96
00:55:40.560 researching you. And then all of our public health mandates just come from people who live 1.00
00:55:44.900 the kind of life that you live. And they keep fucking only working for people who already have 0.78
00:55:49.540 a lot of like, wealth and privilege, then fucking what? Yeah. So the second thing that he says is 0.99
00:55:56.000 contaminating her work is that she's not removing sick people. So the idea basically is that like,
00:56:03.140 the reason why you have these higher death rates among super skinny people isn't because they're
00:56:07.980 like super skinny people that like post photos of themselves in bikinis. They're old people who are 1.00
00:56:12.680 like wasting away from some sort of pre-existing disease. So like my grandfather died of Parkinson's
00:56:18.060 and like in his last two, three years of life, like, yeah, I think he weighed like 85 pounds
00:56:22.680 when he died. In the data, he would count as a death among someone with a BMI of like something
00:56:29.120 like 17 or something that has nothing to do with his weight. It has to do with the fact that he
00:56:33.200 has this pre-existing illness that first made him thin and then killed him, right? So the spike in
00:56:39.120 mortality among thin people is because you're packing in all these people that have like
00:56:44.120 all kinds of diseases, right? Like various cancers, leukemia. If you're in the late stages of a
00:56:48.700 disease, you're going to have a very high mortality rate and you're going to be very thin. And that
00:56:54.080 also affects the normal weight category. And, you know, even that like slightly higher weights,
00:56:58.580 You still have all these people that basically have like wasting away due to disease.
00:57:02.260 Right.
00:57:02.420 That's what I was going to say is like, couldn't you say the same thing about like someone my size wasting away, quote unquote, could mean a loss of 100 pounds for me, right?
00:57:11.860 Like that would be like dramatic weight loss.
00:57:13.780 I would still be in the quote unquote obese category.
00:57:17.080 Well, this is this is the whole problem, right?
00:57:19.480 I think it's it's so interesting once you start talking to people about the higher rates of mortality among thin people.
00:57:27.080 then people are like very likely to die according to these studies and then you watch people's minds
00:57:32.880 go to like well that's not because they're thin right it's probably because like they have an
00:57:36.540 illness maybe they have like a really severe eating disorder or something like that you know
00:57:40.140 there's all these conditions that make you thin and then kill you and it's like wait until i tell
00:57:43.960 you about fat people yeah like there's also medical conditions that make you fat what do
00:57:50.720 you know how many like medications cause people to gain weight i think birth control pills it's
00:57:55.160 like typically like 15 pounds there's mental illness medications that cause you to gain like
00:57:59.740 50 pounds and like oftentimes those people have higher mortality rates too so it's like the same
00:58:04.420 thing is happening on the other end of the scale right there are also health conditions like
00:58:08.340 lipedema which is like swelling and like accumulation of fat in specific parts of bodies
00:58:15.020 if we're cutting people out and going okay it doesn't apply to this person it doesn't apply 0.86
00:58:19.360 to this kind of person it doesn't apply to this health condition it seems fully fucking bananas
00:58:24.280 to do research on fat mortality and not account for things like polycystic ovarian syndrome, 0.71
00:58:30.040 not account for things like eating disorders at any level.
00:58:33.780 Exactly.
00:58:34.200 I mean, this is so fascinating to me to watch people explain away the higher mortality rates 0.88
00:58:39.580 among thin people. 0.99
00:58:40.240 They're like, well, it's complicated.
00:58:41.280 But it's not for fat people.
00:58:42.560 Yeah.
00:58:42.800 So it's like, okay, obviously every fat person needs to lose weight because they're at like
00:58:45.940 a 40% higher mortality risk.
00:58:48.260 But why don't thin people need to gain weight?
00:58:50.280 It's like, oh, well, because it's complicated.
00:58:52.220 It's like, well, it's the same data.
00:58:53.520 it's literally the same table of the same study is saying that like these two groups have elevated
00:59:00.400 mortality and you're prescribing a change in weight for one of those groups and not for the
00:59:05.080 other group like that to me is so revealing it's like oh the thin people like oh this is a real
00:59:10.080 mystery that we need to get to the heart of okay it's also kind of a mystery with the fat people
00:59:14.280 too yeah guess what even knowing that and even noticing this pattern so frequently so commonly
00:59:21.440 that people will bend over backwards to talk about why being thin is always healthier than
00:59:27.400 being fat and being fat is always less healthy than being thin. You're just like, well, there's
00:59:32.900 your bias. And that still doesn't actually change anything in the mindset or behaviors of the person
00:59:37.980 who's doing it. So it's like really fucking frustrating as a fat person to see and hear 0.98
00:59:42.760 these conversations happening all the time and see and hear people just like showing their entire 0.99
00:59:48.360 ass dude yeah it's like look we don't we don't want to strip thin people of all of their humanity 0.98
00:59:53.980 like let's let's figure this out before we prescribe anything like wait a minute hey 0.99
00:59:58.160 thin people are more than a number dude like yeah you know what the fuck so like walter in his 0.96
01:00:05.620 studies recommends removing everybody who dies within the first five years after like their 0.94
01:00:11.360 height and weight are taken what he also says that like after people are diagnosed with an illness 0.96
01:00:16.020 they quote might become motivated for the first time to lose weight fuck off finally a reason to 1.00
01:00:24.700 lose weight oh good what a good motivation is your mortality what the fuck man but then it's it's it's 0.99
01:00:30.960 worth noting there's no actual evidence for anything that he is saying so katherine fliegel 0.99
01:00:36.660 writes like a very salty methodology paper about this that first of all the number of people who
01:00:42.700 are like wasting away like my grandfather with parkinson's is like actually pretty small yeah
01:00:48.080 like people in like the late stages of a disease and also those people are not very likely to
01:00:53.340 answer a survey for a longevity study they're not really contaminating your data all that much
01:01:00.220 and then this thing of like removing everyone who dies in the first five years like you can
01:01:06.100 actually check who are you removing and when you remove people you're mostly removing fat people
01:01:12.980 Walter is saying that you have to do this to remove all these like sick, thin people.
01:01:17.000 But then you actually end up removing a bunch of fat people.
01:01:19.520 And of course, Catherine runs the numbers on a bunch of these studies.
01:01:22.160 And it's like, when you do this, you just raise the mortality rates for fat people.
01:01:26.240 Right.
01:01:26.860 Intentionally or not, it's juking the stats, basically.
01:01:30.060 Exactly.
01:01:30.780 He talks about how you have to remove everybody with these pre-existing conditions
01:01:33.960 because they might be losing weight and then eventually die due to those conditions.
01:01:38.540 but there's just as many conditions where people gain weight after diagnosis. It's very typical
01:01:44.840 for people to gain weight after they're diagnosed with diabetes because you start taking insulin and
01:01:48.600 it spikes your appetite. There's a lot of other conditions that you gain weight once you're
01:01:54.380 diagnosed because you get on medication and you start getting better. So this idea that we have
01:01:58.820 to exclude everybody who has a medical diagnosis because they're becoming thinner, it just isn't
01:02:04.440 true right like you start winnowing out like okay we can't count the smokers and then we can't count
01:02:10.260 people who have who were previously smokers and then we can't count people who've had cancer or
01:02:15.380 parkinson's and then also now we can't count people who have mental illnesses and have been
01:02:20.540 treated for those but we can include the people who haven't been treated for like exactly where 0.86
01:02:25.140 i'm just like what the fuck is this weird ass patchwork that we're coming up with here and also
01:02:30.440 So this is, like, after you do all of these exclusions, right, you're excluding everybody 0.98
01:02:35.220 with a pre-existing condition, everybody who's ever smoked, and everybody who dies within
01:02:40.680 the first five years, Catherine finds a bunch of articles that were written by Walter and
01:02:45.180 his colleagues where they're removing 90% of the deaths.
01:02:48.760 What?
01:02:49.540 90% of the data is gone. 1.00
01:02:52.240 Get fucked. 0.99
01:02:53.720 The central gaslightiness of these statistical methods is that, like, they are doing this 1.00
01:02:58.820 to remove bias officially, right? They're like, well, Catherine Flegel's numbers are biased
01:03:04.760 because she left all these people in. And she's like, Punk, you're removing 90% of the people?
01:03:10.400 And then you're going to tell me that like, it's less biased? Right. And like, does this look like
01:03:15.240 the US population at all? People who've never smoked? People with no pre-existing conditions?
01:03:20.320 Right. Again, this is a place where it would be helpful to start the conversation not from a place
01:03:25.420 of, is research biased or is it not? But where does bias lie in all research that we have?
01:03:32.140 Exactly. So like, among the population, around 40% of people die to something related to
01:03:38.700 cardiovascular disease, like 40% of deaths in the US. But then when she looks at these
01:03:43.200 samples that they're using, only 20% of the deaths are due to cardiovascular disease.
01:03:48.200 This isn't a normal pattern. 0.90
01:03:50.200 Part of the reason why it's so hard to get to this fucking number is something that it feels 0.94
01:03:53.760 like none of the data is really like engaging with. It turns out fat people have other health 0.98
01:03:58.900 conditions and so do thin people than just the size of their fucking bodies. There's like lots 1.00
01:04:04.140 and lots of factors that impact people's health that are beyond this like bizarre world that so
01:04:10.500 much of the research wants to imagine, which is just like a bunch of fat people started out as
01:04:14.840 thin people and then just decided to keep eating. And we're just going to measure the impact of that
01:04:20.740 on their health. Like the number of assumptions built into that are astonishing. I mean,
01:04:25.960 this gets us to like the final chapter of the story, which is basically none of this shit 0.99
01:04:30.640 matters. Yeah. I'm just going to go ahead and undermine everything we've already talked about. 0.99
01:04:35.040 So it's now 2016. Catherine ends up resigning from the CDC. She's now at Stanford. Walter,
01:04:41.900 meanwhile, has basically put together this entire like consortium of I think it's like 500 researchers
01:04:47.520 or something like that, like this big global BMI collaboration, basically just to debunk
01:04:54.480 Catherine's work.
01:04:55.980 So he publishes in 2016, this like massive study that he says is like to settle the debate.
01:05:03.300 Her meta-analysis had like 3 million people in it.
01:05:05.660 His meta-analysis has like 10 million people in it.
01:05:08.660 Okay.
01:05:09.100 And it's like, all right, Catherine, this is it, the final word on obesity and health.
01:05:13.960 and they published this study and wouldn't you know it normal weight people have the lowest
01:05:19.060 mortality and then overweight people have more and then obese people have like even more yeah
01:05:23.880 sorry katherine that's just the science i do like the idea that the dude who like started this whole
01:05:30.740 weird quote-unquote debate that is really just seems like an attempt to torpedo this lady's
01:05:35.600 credibility is like guys i got it i'm gonna settle the debate that i started
01:05:41.640 no okay i also love that in this study even all of the manipulations that we'll get into
01:05:48.980 thin people are still more likely to die yeah like he wasn't he wasn't able to get rid of that 0.51
01:05:53.640 there it is the funniest fucking thing about this is that he's like captain methodology like i am 0.99
01:05:59.040 so disappointed with katherine's methods right and this study is fucking garbage oh katherine 0.97
01:06:06.340 legal writes like a series of responses that are like some of those scathing like in academic 0.98
01:06:11.320 language but like very scathing language about how bad the study is so first of all his whole thing
01:06:16.560 is of course like he wants to remove like everyone with a pre-existing illness everyone who's ever
01:06:22.580 smoked etc the problem is among these studies that he's looking at there's 239 studies that
01:06:29.300 they're looking at only 28 of them even have data on people with pre-existing diseases yeah so he's
01:06:36.560 like how dare katherine fliegel not remove these people and and she's like only a tiny bit of your
01:06:42.120 data allows you to remove those people so functionally you're leaving them in two and then
01:06:46.560 my favorite shit is that in all of their analyses of north america because it's a global study 94 0.92
01:06:53.460 percent of the deaths came from studies with self-reported weight and the other thing i got 0.99
01:06:59.940 totally obsessed with is that in this study walter is cutting out everybody who dies in the first
01:07:04.780 five years. And he says that you have to do this to make your data clean. And anybody who doesn't
01:07:09.240 do this is publishing flawed, naive, misleading data, whatever. The question I kept coming back
01:07:14.280 to is like, well, wait a minute. Why five years? Why not four years? Why not two years? And when
01:07:20.440 Catherine looks into this, Walter Willett himself has published studies that cut out everybody who
01:07:25.900 died in the first four years. A lot of his co-authors on this same study have published
01:07:30.700 data where they cut out people who die in the first two years. So it's like existentially
01:07:36.360 important to do this because it totally contaminates your data. But you yourself are not
01:07:41.040 doing it consistently. Like, it's weird to me that we're all supposed to see it as some sort
01:07:46.360 of coincidence that he happens to choose the methodologies that produce the highest mortality
01:07:53.000 rates for fat people. Like, we're all supposed to look at this and be like, oh, well, all he's
01:07:57.760 doing is what he considers in a content neutral way to be the best methodology for these studies.
01:08:02.740 And would you look at that? This aligns with exactly the policy preferences that he has been
01:08:09.600 telling NPR reporters and everybody else who will listen in public. He thinks that it is dangerous
01:08:15.220 for you to be fat. He thinks that you should weigh what you weighed when you were 20. And he happens
01:08:19.980 to believe, total coincidence, that the methodologies that produce that result are the
01:08:24.760 right methodologies. It just all reads to me as people grasping at straws to justify what they
01:08:31.120 already believe. Yes. There's a little bit of an imperfect parallel, but a parallel to journalism
01:08:37.660 and the idea of objectivity and journalism, right? Yeah. Scientifically, we know that like
01:08:43.220 objectivity doesn't actually exist, right? That all of us are influenced by cultural biases all
01:08:48.980 of the time, implicit or explicit. So like, what would it look like to build journalism or build
01:08:54.380 research around acknowledging those biases and working to counterbalance them rather than
01:09:00.240 insisting that they don't exist and still doing what you were going to do anyway.
01:09:04.800 I mean, this is the whole, this is the whole thing. I mean, this is like,
01:09:08.520 this is like my grand conclusion, like what I want to end with. It's important to break
01:09:12.920 this whole dilemma into two separate parts. So for the vast majority of people who are in
01:09:20.020 either the overweight category or like class one obesity like low grade fat people that's like the
01:09:27.040 majority of the people we're really talking about in america and like the the relative risk mortality 0.61
01:09:33.960 ratios they're so fucking small yeah if you even in like the walter willett like p hacked within
01:09:39.460 an inch of its life study it's like 40 higher for like grade one obesity i think again unmarried 0.97
01:09:46.920 people have a 230 increased mortality rate guys i'm going to die because i'm single not because 0.99
01:09:55.120 i'm fat this is the thing it's like in these like fake stupid categories of like overweight and 0.95
01:10:00.840 class one obesity the mortality risk rates are so weird and like tiny and conditional on these 0.93
01:10:06.600 weird statistical methods at those weights yep that like i i honestly just think that like it's
01:10:12.160 it's bullshit to tell somebody in those categories that like they must lose weight like the data does
01:10:17.720 not support that i think straightforwardly right and then there are these categories of much fatter 0.93
01:10:23.620 people once you get into bmis above like 35 or 40 it's like somewhere between like two and five
01:10:29.660 percent of americans right very consistent correlations that like those people have like
01:10:33.540 very elevated mortality rates they have much higher rates of heart disease like i'm not going
01:10:38.420 to deny that like those correlations are extremely consistent. But then the question with much fatter
01:10:43.600 people is, first of all, what is causing that? And second of all, what do we do about it?
01:10:48.760 Yes, it's a small number of people. It's it's me. I'm in that category. Hi, everybody. I'm in that
01:10:53.980 category. And that two to 5% of people are people who are already facing like untold levels of
01:11:01.820 stigma. Yes, we are being encouraged at every turn to treat that group of people like shit. 1.00
01:11:06.640 So, like, bad faith actors start to fucking activate and form, like, whatever their, like, Voltron shitty anti-fat shit. 1.00
01:11:15.200 Yeah. 1.00
01:11:15.400 This is already a group of people who are getting shit on left, right, and center. 0.99
01:11:19.000 Exactly. 0.99
01:11:19.580 If we go to those people, when you look at those statistics, it's dire.
01:11:23.820 It's like they're half as likely to be college graduates.
01:11:27.320 One quarter of them are earning less than $20,000 per year.
01:11:31.300 They're twice as likely to be on Medicaid.
01:11:33.120 the group with the highest prevalence of grade three obesity is black women who didn't complete
01:11:39.860 high school? Yeah, there it is. Is this a group that like we're confident in saying it's the 1.00
01:11:45.960 weight that's doing this and that like what they need is to lose weight? Yeah. Is that the biggest
01:11:51.900 need that we've identified here? Also, have we listened to poor fat black women about what they 1.00
01:11:57.300 want and need exactly instead we're focused on chiding fat people for what we assume their 0.99
01:12:04.280 behaviors are there's all kinds of studies on like fat people do not get pap smears fat people
01:12:09.320 do not get like prostate cancer checks if you want to say that it's like the adipose tissue
01:12:13.900 that is causing those health consequences you know what fine i'm probably not going to talk
01:12:18.960 you out of that opinion but it's like what can we do like what are the steps that we can take
01:12:24.260 for that group that will actually make a difference. Because telling them to lose weight,
01:12:28.500 we know that it doesn't work. Medical institutions define successful weight loss as losing 10% of
01:12:33.840 your body weight. So if you're 300 pounds, you're going to be a 270 pound person. That's the best
01:12:39.840 we can deliver. And not even like very reliably, but like, there are some programs that have been
01:12:45.020 shown to deliver that for like some number of people. Great. We now have a 270 pound person.
01:12:50.460 what does that person need right the idea that any of this data doesn't account for medical bias
01:12:57.400 and the ways in which doctors will like diagnose illnesses differently the ways in which doctors
01:13:02.060 will like and health care providers of all stripes will have like shorter visits with fat people
01:13:06.820 they'll order fewer tests they'll consider fewer options and then fat people understandably will
01:13:14.220 postpone health care as long as they possibly can. I have a family friend who was like put on
01:13:20.700 prednisone by her doctor and put on weight because that's what happens when you go on prednisone,
01:13:25.340 right? This is one of many medications where this happens and is now having significant,
01:13:30.200 significant cardiovascular issues. And her doctor is saying, well, it's because you didn't lose
01:13:34.340 weight and you need to lose weight. And she's like, I need to go into the hospital. These are
01:13:39.600 like very clear cut cases of fat people clearly not getting the care that thin people are getting
01:13:46.260 it just feels utterly bananas that we have a cultural conversation that is dominated by thin
01:13:52.340 people and thin people's imagining of fat people's health and not the like extremely overwhelming
01:13:59.800 very clear experiences of fat people on this front one things i found in one of the older papers
01:14:05.140 one of the few papers actually that actually investigated the causes of death among fat
01:14:10.380 people like cancer deaths are much higher among fat people this is like a pretty well established
01:14:14.420 thing but they noticed that one of the reasons for that was that fat people couldn't fit into
01:14:19.900 the machines that they needed yeah they also a really interesting one was they mentioned that
01:14:24.300 like fat people were much more likely to get gallstones right it's some sort of side effect
01:14:28.580 of like what you're eating but when you look into gallstones it's also a side effect of dieting
01:14:33.400 If you stop, especially fat consumption, but like calorie consumption, your liver just like goes nuts. So like 30% of people who get gastric bypass surgery and stop eating very quickly, they get gallstones, 30%. So it's like another one of those things that is like, it's seen as like, well, fat people are at higher risk of gallstones. And like, okay, but is it the fat doing that? Is it their diet? Or is it dieting? And those are actually like very different phenomena as a society that we would need to look into.
01:15:01.300 Quite a bit of the things that we think of as being like the cost of being fat are actually the costs of weight cycling, right? 0.94
01:15:09.060 Guess who's under the greatest pressure to diet all the fucking time? 0.99
01:15:13.820 Surprise, it's fat people. 0.99
01:15:15.980 Again, like there are so many confounding factors and the fact that we spend zero time talking about any of them feels like a real tell.
01:15:25.300 Wait, can I read you something?
01:15:26.700 Yes.
01:15:27.160 okay so i do think that like the effect of dieting and people basically being in like a starvation
01:15:33.400 state for potentially years is like vastly under covered yes but the studies are trash oh really 0.98
01:15:41.260 oh my god a lot of them are on fucking rats and they do show that like raising and lowering their 0.99
01:15:45.420 rats weight is like really bad for them which like fine but also rat studies yeah yeah a lot of them 0.99
01:15:49.720 are just slicing and dicing the same data and like people aren't gathering weight cycling data so
01:15:54.900 So, like, these big cohort studies, one of the huge weaknesses of them is you're relying on, like, what were they asking people in, like, the 1990s?
01:16:02.440 You can't go back and ask people the same questions.
01:16:04.920 So, they're using the data that they have.
01:16:07.180 So, I found a study on weight cycling, like, a study that shows that weight cycling is bad. 0.98
01:16:11.920 But look at this fucking methodology. 0.99
01:16:13.680 So, they're looking only at men, of course, because these cohorts are terrible. 1.00
01:16:16.580 Strong start.
01:16:17.480 Exactly.
01:16:17.780 So it says, at the time of their first examination, these men were asked to recall their weights at age 20, 25, 30, 35, and 40.
01:16:27.700 I know you would love this.
01:16:28.980 Weight cycling was defined as a gain of 10% of body weight in one five-year interval and a loss of 10% in another five-year interval.
01:16:37.360 What?
01:16:37.940 That's not how weight cycling works. 1.00
01:16:39.900 My mom weight cycled my entire fucking upbringing. 1.00
01:16:42.700 Yes. 0.99
01:16:43.140 It was not on five-year cycles, dude.
01:16:45.380 It was on like three months.
01:16:46.940 one year your mom and like many many many moms yeah like oh the moms this is what's so fascinating
01:16:55.920 to me is like there's so little interest in figuring out any explanation for these health
01:17:01.240 conditions other than obesity yeah there is also like i would imagine a pretty significant impact
01:17:07.620 on the health of people who engage in very severe dieting but don't meet that definition of weight
01:17:16.580 cycling exactly or that don't lose or gain weight particularly at all exactly that's the thing you
01:17:23.640 know and we've talked on this show about like people can be 250 pounds and eating so little
01:17:29.160 that they're not getting their period yeah there's lots of people that are like at some version of
01:17:34.940 that at some level of like a starvation state in their body and just maintaining that for years so
01:17:39.580 they wouldn't show up in the study i'd just like to say it again there's not a single method of
01:17:45.340 weight loss that is non-surgical, that meets the standards of being an evidence-based treatment.
01:17:51.080 Exactly. 0.96
01:17:51.680 So like, what the fuck are we doing here? If we're all, if all of our research that we're 0.97
01:17:57.260 conducting is reverse engineered to lead back to the same fucking mandate about weight loss, 0.86
01:18:03.860 why are we doing that if we don't know how to produce weight loss? And frankly, 0.97
01:18:09.140 even if we did know how to make people lose weight, it is intensely troubling that we have
01:18:14.860 massive societal institutions that are all geared toward forcing people to look one way yeah and to
01:18:22.020 have one kind of body that's fucking terrifying so can we end with like a little epilogue yes
01:18:28.120 please so in 2021 katherine fliegel wrote a article about like all of this that is where 0.96
01:18:35.600 all of this is coming from this chronological retelling of like i did this work and uh this
01:18:40.500 guy walter willett was like a real weirdo about it yeah walter has always been like totally clear
01:18:45.920 that like he regrets nothing so when he was contacted by journalists after catherine wrote
01:18:51.460 this essay he said and i quote her description is mostly correct and she shouldn't have been
01:18:58.020 surprised i did describe her paper as rubbish and i do stand by it wow the captain's going down with
01:19:04.480 the ship just like she says it really sucks the way that i treated her and like i feel fine about 0.89
01:19:09.720 it wow you know what your paper was bad yep she's right i did it i'd do it again so i have been
01:19:15.560 trying for a couple weeks to get an interview with walter willett because i feel like if i'm
01:19:19.200 gonna be mean to you on a podcast like i should give you a chance to respond yeah we scheduled
01:19:23.920 one and then he had to cancel something moved around but then he said look sorry i had to
01:19:29.320 cancel our interview but i've written a response to katherine fliegel and like here it is the
01:19:34.720 The response is called evidence does not support benefit of being overweight on mortality.
01:19:40.300 So just like right back to like, she's wrong.
01:19:44.480 Back to square one.
01:19:45.380 So he says her meta analysis did not deal adequately with the fact that smokers have
01:19:51.760 lower BMIs, but high mortality.
01:19:54.000 And it's like, Walter, the whole debate was about whether this was an adequate way to
01:19:58.580 deal with that.
01:19:59.420 Right.
01:19:59.660 That's the debate, Walter.
01:20:01.020 Yeah.
01:20:01.280 On the one hand, I appreciate his like straightforwardness of just being like, 0.98
01:20:06.460 yeah, I fucking did it. 0.97
01:20:07.400 I do it again. 0.99
01:20:08.060 What about it? 1.00
01:20:08.560 She sucks. 1.00
01:20:09.280 Yeah. 1.00
01:20:09.500 She sucks. 1.00
01:20:10.280 She's right. 1.00
01:20:11.420 I did say that she sucks. 1.00
01:20:12.600 I do believe that she sucks. 1.00
01:20:14.280 And on the other hand, it feels so disingenuous to be like, actually, this debate was about 1.00
01:20:18.400 this other thing where I'm like, no, it fucking wasn't. 0.90
01:20:20.580 And it's all in writing.
01:20:21.560 What are you doing?
01:20:22.320 He also says that his 2016 garbage paper firmly refuted her own work, which like, no, it's
01:20:29.740 Just, you did it differently than she did.
01:20:32.080 He also says, this is my favorite line.
01:20:35.040 He says, these are not statistical issues, but rather the reality of human biology.
01:20:40.240 Great.
01:20:40.740 And require detailed analyses that could not be conducted in a meta-analysis of previously published data.
01:20:47.760 Good.
01:20:48.060 Walter, you also did a meta-analysis that was based on previously published data.
01:20:53.240 Like, you're now against meta-analyses?
01:20:56.240 It's real wild that there's like no engagement with the substance of the critique here.
01:21:01.380 It's incredible.
01:21:01.880 There's no point at which he feels, he appears to feel compelled to defend his own methodology.
01:21:07.140 Exactly. 0.99
01:21:07.800 He just goes back to, anyway, she sucks and I'm right. 0.99
01:21:10.700 Exactly. 0.99
01:21:11.760 Which I'm like, what?
01:21:12.860 I have drafted an email to him so many times because he emailed me this paper and I read
01:21:18.360 it like a gog.
01:21:19.680 Yeah.
01:21:19.960 I wanted to be like, Walter, do you not, do you not see it?
01:21:24.660 this you're not engaging with this debate at all or like even acknowledging that it's a debate
01:21:29.780 right it is i imagine reading that email was not unlike reading the one sentence where they're 0.99
01:21:35.720 like we just assumed every fat person died too fat where you're like you just fucking said it 0.93
01:21:40.140 and published it and then a bunch of people were like this science is good like what i don't want 0.98
01:21:46.440 to both sides of this and say that like they're both doing the same thing because i think walter
01:21:50.360 willett is like much worse and what he's doing is like less methodologically defensible
01:21:54.860 But also, Catherine Flegel is probably also guided by bias in this too.
01:21:58.940 The process of making science is a series of judgment calls.
01:22:02.020 And there's nothing wrong with that.
01:22:03.060 There is a very strong association between fat people and worse health outcomes.
01:22:07.660 But I think we have been told a specific way to interpret that for most of our lives.
01:22:12.840 And I don't think that we've been told all of the judgment calls that made that conclusion.
01:22:17.940 Yes.
01:22:18.140 And if you're creating all this research, if you're creating this industry around reducing
01:22:22.520 the number of fat people. This is what Alicia Mundy calls Obesity Inc. You are going to be
01:22:27.420 training generations of healthcare providers in reinforcing their existing cultural biases
01:22:33.540 using science, right? So it's like introducing this incredibly moralizing view of patients who
01:22:40.980 have no choice but to trust you or just not get healthcare. And that's like part of what's
01:22:46.700 happening with fat people, right? Yes. Oh, yeah. And also, I mean, the whole idea of boiling
01:22:51.460 somebody down to their like statistical mortality risk that that's a gross thing to do regardless
01:22:58.260 like i i have had this conversation with so many people where i'm like uh yeah you probably should
01:23:02.360 just like not treat people according to like what you think their you know health risks are
01:23:06.800 i don't go around like scolding smokers and then every once in a while you'll get people who are
01:23:12.740 like well i do scold smokers i do yell at smokers in my life and it's like congratulations on like 0.98
01:23:18.660 being a weird dick like great good for you I mean like genuinely like all of this shit about like 0.99
01:23:24.500 I'm concerned about your health all of this shit about burdening health care systems all of this 1.00
01:23:30.240 stuff to me as a fat person reads as a fig leaf for I want to talk about how uncomfortable I am 0.99
01:23:37.540 with fat people and how much I dislike them and how much they gross me out and how much I don't
01:23:42.380 want to be fat and all of these other things that are not even remotely related to the health of
01:23:47.220 people. Yeah. It's not like people learned about the population level hazard ratios of adipose
01:23:54.160 tissue and then decided to dislike fat people. It seems extremely obvious that people disliked
01:24:00.120 fat people and then went looking for the hazard ratios. They were looking for a reason.
01:24:05.540 It's such a challenging thing to have people doing all this shit in the name of quote unquote, 0.98
01:24:10.320 the science. And then when you look at quote unquote, the science, it's like very human, 0.97
01:24:14.700 very flawed very unreliable and very disputed and also the science is pretty clear that like being 0.95
01:24:20.920 mean to people is like bad for them right crystal clear it's crystal clear that being a dick to 0.99
01:24:27.600 people is bad for everybody's health so maybe knock that shit off but how bad 0.99
01:24:32.220 aubrey i need to uh i need a cohort find me 70 000 nurses 0.99
01:24:36.720 I love you.