January 2, 2025 • 46 mins
Here at ParentData, we talk a lot about panic headlines. You know, the headlines that cycle through your feed about coffee and wine and sleep and lead and the causes of autism, many of which contradict the last panic headline, and almost all of them turning out to be not nearly as bad as they seem. But in the moment, they feel so scary and urgent. And if you're a parent just trying to follow the science, do what's best for your kid, sometimes it feels like you're being absolutely and really nonsensically bombarded with the wrong things to do.
Today on ParentData, we've invited Dr. Bapu Jena to help us stay sane. Bapu is an economist and a medical doctor who specializes in natural experiments, which means observing human behavior in naturally existing behavior (as opposed to a randomized trial). This makes him an ideal person to talk about the uses and abuses of data, and how curious nerds conduct research that makes its long and winding way into a headline that almost feels like it's designed to scare the crap out of parents. We talk about the complicated relationship between causality and correlation, the academic and popular incentives to publish these kinds of headlines, and also who decides what research is worth sharing with the world.
This is on the face a conversation about research, but really it's about reassurance - there are a lot of reasons behind publishing a story about lead in Cheerios that have nothing to do with you or how dangerous Cheerios actually are or whether you're a good parent who cares about the health and wellbeing of your kids. You are and you do. Don't throw out your Cheerios, but do explore the journey with us.
Subscribe to ParentData.org for free access to new articles every week on data-driven pregnancy and parenting.
This episode is generously supported by:
See omnystudio.com/listener for privacy information.
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:01):
I'm Emily Oster. This is parent Data, and today.
Speaker 2 (00:05):
New tests reveal popular cereals and snacks could have potentially
dangerous levels of the same chemical found in weed killer.
Speaker 3 (00:17):
Here at parent Data, we talk a lot about panic headlines,
you know, the headlines.
Speaker 1 (00:23):
That cycle through your feed about coffee.
Speaker 3 (00:25):
And wine and sleep and lead and the causes of autism.
And many of the panic headlines contradict the last panic headline.
Some of them end up debunked, or maybe there was
a math error. They're not as scary as they originally seem,
but in the moment these feel so scary and urgent.
Speaker 1 (00:45):
There is new research that impacts you if you're a
coffee drinker.
Speaker 4 (00:48):
Can more screen make your child autistic? Nearly forty percent
of parents use white noise machines to help their children
sleep sounder and longer. However, some health experts are concerned
about the effects that these machines could be having on
a child's brain development.
Speaker 5 (01:03):
More and more women are actually throwing out their tampons,
opting out of using tampons for fears of toxic chemicals
that could be harmful, according to.
Speaker 1 (01:11):
A recent Consumer Reports start a news study from.
Speaker 4 (01:13):
Japan study on rats show that those raised in continuous way. Today,
researchers in Japan say the more time your child is
exposed to it, the more likely they could land on
the autism spectrum.
Speaker 1 (01:25):
The study was conducted between one.
Speaker 3 (01:30):
If you're a parent just trying to follow the science,
do what's best for your kid. Sometimes it feels like
you're being absolutely and really nonsensically bombarded with the wrong
things to do. And when it all feels like too much,
I like to call doctor Babu. Jenna Babu is an
economist like me, but unlike me, he's also a medical
(01:53):
doctor and he loves natural experiments, which means an experiment
he can't can observing people existing in naturally occurring circumstances.
For example, he has some excellent work on flu shots
showing that kids whose birthdays fall in the fall are
more likely to get their flu shots because they can
(02:15):
get them at their yearly well visits. And then he's
able to study the impact on flu shots and find,
for example, the kids are less likely to get the flu, and.
Speaker 1 (02:25):
So are their elderly relatives.
Speaker 3 (02:27):
This isn't a randomized trial that you can set up yourself.
This is finding something clever in the data that can
allow you to generate a causal estimate even though it
hasn't been explicitly randomized. If you've read Freakonomics, you're familiar
with this kind of research. And for me, I love
(02:50):
this stuff and I love BAPU, And the reason we
brought them back today is because we wanted to talk
about how research can be abused on it makes its
way into the world. I wanted to talk about how
we get to that splashy headline that warns you that
there's lead in your cheerios, or you have to throw
out all your black plastic cooking utensils, or how screens
(03:13):
give your kids autism?
Speaker 1 (03:15):
How do we get there?
Speaker 3 (03:17):
How does the research that informs that get published If
it's not the researchers aim to make us worry, they're
just curious data nerds. Why does it feel like they're
colluding with the newspapers to scare the crap out of parents.
So that's what PAPU and I are going to talk about,
what the researchers are actually aiming to do, and then
(03:38):
what happens to that research when it arrives in your
headlines to freak you out. We talk about the complicated
relationship between causality and correlation, the academic and popular incentives
to publish these kinds of headlines, and also who decides
what research is worth sharing with the world. This is
(03:59):
on the fase conversation about research, but really it's about
reassurance that there are a lot of reasons behind publishing
a story about lead and cheerios that have nothing to
do with you, or how dangerous cheerios actually are, or
whether you're a good parent who cares about the health
and well being of your kids. You are, and you
do don't throughout your cheerios. But also, if you're curious,
(04:21):
listen to the journey that headline chok to land in
your lap.
Speaker 1 (04:26):
After the break, Doctor Babu.
Speaker 3 (04:28):
Jenna, Babu Jenna, thanks for joining me.
Speaker 2 (04:39):
Thank you for having me.
Speaker 3 (04:42):
You've actually been on the Parent Data podcast before, but
for people who weren't listening, like you know, two and
a half years ago or whenever.
Speaker 1 (04:50):
I had you. Could you introduce yourself?
Speaker 2 (04:53):
First of all, it feels like yesterday, but that's right.
Speaker 1 (04:55):
Oh, it just feels it's so such a treat. You're
always in my.
Speaker 5 (04:58):
Head except yeah, my name is Babu Jenna. I'm an economist,
physician and professor at Harvard Medical School. I have a
book called Random acx of Medicine with Chris Worsham, and
a podcast called Freaconomics MD.
Speaker 1 (05:16):
I think it's important to emphasize that although you.
Speaker 3 (05:17):
Start with being an economist, you are, unlike me, also
an actual doctor.
Speaker 2 (05:24):
That's true. I'm a doctor doctor.
Speaker 1 (05:26):
Yeah, you see patients.
Speaker 2 (05:28):
I do.
Speaker 5 (05:29):
I was actually just on service for a couple of
weeks at Mass General in Boston.
Speaker 1 (05:34):
In the emergency room or no, I work.
Speaker 2 (05:37):
In the general medical floor.
Speaker 5 (05:38):
So I spent about two weeks there in a stretch,
and so I just finished up yesterday.
Speaker 3 (05:44):
Do you feel like you get a lot of ideas
for your economics work from your doctoring?
Speaker 5 (05:50):
Yeah?
Speaker 2 (05:50):
I do.
Speaker 3 (05:51):
Yeah.
Speaker 5 (05:51):
I mean there's like all sorts of random things that
you see in the hospital, things that patients will say,
or problems that you'll encounter, which I think certainly have
you given me ideas in the past. A couple of
ideas came from things I saw in the hospital.
Speaker 1 (06:05):
I'm just always curious about that.
Speaker 3 (06:07):
You seem like a person who gets many of your
research ideas from just walk like walking around in the
world or things that happened in your own life.
Speaker 5 (06:15):
Yeah, and chat GPT is feels quite good at this.
I don't know if you've tried it for coming up
with ideas, but it's not bad.
Speaker 3 (06:22):
Were you just write in like, I need a research idea.
I'm a crazy doctor who likes natural experiments. Give me
a research idea.
Speaker 2 (06:29):
Yeah.
Speaker 5 (06:29):
So, for example, I had some dinner with some students
at Harvard the other night, and we were talking about
chat GPT, and so I said to the bart you know,
suppose I want to study whether or not playing football
as a youth impacts.
Speaker 2 (06:43):
Your long term health.
Speaker 5 (06:45):
And so I asked, chat GPT, that's what I want
to know. Give me some natural experiments to try to
figure this out. And CHATCHEPT came up with a couple
of good ideas. It says, all right, well, let's look
at whether or not Pop Warner programs come into certain
at certain times. Let's look at age cutoffs, let's look
at changes in policies towards contact versus no contact football.
(07:10):
You know, all very plausible ideas. And I don't think
anybody has studied this question before, and so it was
remarkable that it was able to come up with ideas
that all made a lot of sense. None of them
I would pursue, but not bad.
Speaker 2 (07:23):
Not bad.
Speaker 3 (07:25):
I feel like my use of GPT is more limited.
Yesterday I had it make me an image of Elf
on the shelf, which was actually very was actually very good. Okay,
But the purpose of this call is not to advertise
the products of open AI, which is not a not
a sponsor, but to talk about a topic that I
(07:48):
think you and I both care a lot about, which
is why so much of the research that people hear
about in the world is just correlation and not causation.
Speaker 1 (08:03):
And you know, we've I talk about this a.
Speaker 3 (08:05):
Lot, and I guess that the I want to sort
of set the stage for the problem that we're talking about,
But then I actually want to talk about like why
this happens, which for me is the more interesting, the
more interesting question. So, uh, the canonical example of this
problem is, you know, the sort of studies show that
you know, new studies shows that you know, coffee like
(08:29):
generates makes you live, you know, generates higher longevity, or
new study shows that coffee makes you die sooner, or
you know new A lot of this stuff happens in nutrition,
and it's kind of.
Speaker 1 (08:40):
Either specific or it can be very general.
Speaker 3 (08:42):
You know, new study shows that, like people who consume
more ultra processed foods, you die sooner. And and these
are places where often when people send them to me,
then I say, you know it's correlation and not causation,
but can you give me your answer?
Speaker 1 (08:59):
You know, I'm your page.
Speaker 3 (09:00):
I come and I say, I saw this new study
that says blah blah blah about coffee, Like, how do
you help me understand what is the problem with that?
What are some of your concerns about that study?
Speaker 5 (09:11):
Well, so, typically if someone going to ask me that
kind of question, I'd say, all right, Well, if their
goal is to understand whether taking this medication, taking this supplement,
drinking coffee is going to actually improve the quality or
length of their life, how would you study it? And
I'd say to them, the only way to really study
(09:31):
this is to do a randomized trial where you take
a bunch of people and you randomize some of them
to getting.
Speaker 2 (09:40):
The product.
Speaker 5 (09:40):
In this case, it could be coffee, it could be medication,
And you take an otherwise similar group of people and
you randomize them to not getting that, and then you
study differences in whatever outcome you care about. It could
be blood pressure, could be weight, could be how long
you live. And if you see something there, then you
can say it's because of the thing that you did,
(10:02):
in this case drinking coffee. In the real world, though,
you don't have that randomization for the most part, and
so you have people who drink a lot of coffee
and people who don't. And I would explain to them
that people who drink a lot of coffee are different
than people who don't. And because of that, you don't
know if it's the coffee that's leading to this health
outcome or everything else that is different in those people
(10:24):
who drink a lot of coffee. And I think most
most people would understand that idea if told in that way.
Speaker 3 (10:31):
I mean, the other thing I will sometimes tell people
is think about this choice and think about whether it
seems like people are making it randomly. So people say,
you know, there's a relationship between family dinner, you know,
having a family dinner, homegook, family meal every night, and
outcomes for kids and your school outcomes.
Speaker 1 (10:53):
Just ask them to think about, you know.
Speaker 3 (10:55):
Once you accept the best way to do this would
be randomization, you would ask sort of compare groups who
do it and who do not. How close is that
to randomization? Like how much randomness do you think there
is in this choice? And with many of these choices
like what I eat or whether I have family dinner
or whatever it is, it doesn't really seem very random.
(11:16):
It doesn't seem like there's a very important component of
randomization there. That's like the world is not doing a
good job randomizing for you.
Speaker 5 (11:23):
Yeah, and another way to put it, and we have
a technical term that we use in our studies called
falsification tests. I would say to them, all right, suppose
that I showed you that people who drive expensive cars
live longer. Would you say to me that buying an
expensive car will make you live longer? And most people
say no, absolutely not. And I'd say, well, what if
(11:45):
it's the case that people who drink a lot of
coffee drive fancy your cars or something like that. Then
they would sort of get Okay, I get it. There's
like there's there are other factors that are going to
be correlated with some outcome, but could not plausibly be
so in a ca.
Speaker 3 (12:01):
My favorite thing I ever did about this was a
time when I took every food in the National Health
and Nutrition Examination Survey, and I correlated it with the weight,
and I showed that like, for example, dandelion greens are
associated with a lower weight, but iceberg lettuce is associated
with having a higher weight, And.
Speaker 1 (12:20):
Like you know, chemical based.
Speaker 3 (12:22):
Sugar substitutes are associated with higher weight, but plant based
sugar substitutes are associated with lower weight, which for me
was like such an illustration of like, you know, who's
eating iceberg lettuce?
Speaker 1 (12:31):
Who's eating what kind of idiots eating dandelion greens? You know,
it's like my dad and he's also doing every other thing.
Speaker 3 (12:37):
And I think that kind of thing can make this
quite vivid when you're just like, you know, well, actually,
what do you.
Speaker 1 (12:44):
Think is going on?
Speaker 3 (12:45):
And as soon as you point that out to someone,
the fancy car is a good example, it's like, oh yeah, okay,
Like it's not that this is not that complicated a
point to make two people I find.
Speaker 5 (12:57):
Yeah, yeah, but I mean, just to be fair, they're
pretty high quality evidence about dan line greens and dimension
that's not true, there's not Can I just ask a
clarifying question when you say dandelion greens, are those the
same thing as those fancy little microgreens that you get.
Speaker 3 (13:14):
No, a dan lion green is literally a part of
a dandelion. It's like instead of eating, you don't eat
the flower, but.
Speaker 1 (13:21):
You eat the greens.
Speaker 3 (13:23):
Yeah, microgreens are a delicious treat associated with uh, you know,
higher rates of IQ.
Speaker 5 (13:30):
Or I'll say the following though, that's like, you know,
the astute listener will know that my name is A
is an Indian name, and so it's not uncommon for
Indian people to take flowers and like like fry them
in some like flour like do it comes out delicious?
Speaker 2 (13:46):
I don't think it's healthy though.
Speaker 3 (13:48):
Right, the flowers might be good for you, but the
frying somewhat negates It's actually that people are frying their
iceberg lettice.
Speaker 2 (13:57):
That's exactly exactly.
Speaker 4 (14:01):
So.
Speaker 3 (14:01):
A thing I find puzzling about this space is I
feel like if you explained this, and I have to
my thirteen year old, this concept that you know, correlation
is not causation and that you would really want to
worry quite a lot about the other things that might
(14:25):
be driving these relationships.
Speaker 1 (14:26):
I feel like that is a.
Speaker 3 (14:27):
Very intuitive idea, and you know, not that like everybody
sees it immediately, but that it isn't very hard for
people to understand. Even in a you know, a short conversation,
you can kind of get people to see it. I mean,
I think, you know, I spend a lot of time
on Instagram short form Instagram videos.
Speaker 1 (14:43):
Explaining this, and I think some of the time it hits.
Speaker 3 (14:46):
The question I find more puzzling is why are there
so many papers which do this, which are written by
people who haven't just watched a forty five second Instagram
video or had one conversation with you, but who have
spent literally many, many years taking classes on these exact
(15:06):
topics and then writing papers. And these people are, you know,
professors or like, they have a lot of education, They're smart,
and I just fundamentally don't understand why so much of
this literature keeps happening, and whether the people who are
writing it think that they are uncovering something causal or not.
Speaker 1 (15:32):
Do you have any insight?
Speaker 5 (15:34):
I have answers, I don't know if they're insightful, But well,
let me just say the following. First of all, and
at some point later day, I'm going to tell you
about a paper that we have coming out in probably
a couple of weeks. It will be December, so I
don't know when this will air, but it will fall
into this category of a very weird sort of correlation.
Speaker 2 (15:54):
And the.
Speaker 5 (15:57):
Way that we would describe it in the way that
we do justscribed as finding, which I know I'm making
it sort of sound very sort of like wow, what's
what's he going to drop on me in a moment?
Is it's just a hypothesis. So I would have no
problem if people would do the kinds of studies that
you're describing, which are purely correlational studies, no sort of experiment,
(16:18):
serving as the backbone behind the idea and saying, look,
here's this interesting observation. We don't know what to make
of it. It could be causal for X, y Z
reason like there's a there's a channel by which this thing,
in this case dandelions, could have an effect on some
health outcome and it would work by this mechanism. You know,
(16:41):
this flower affects this protein in the body. This protein
in the body affects these cells. These cells affect the
development dementia something like that. Totally fine with that, because
then a reader could say, all right, well, you know
I want to investigate this or not investigate and investigate
this any further. But that's not the way the studies
come out. The studies come out as you know, danne
greens are you know, strongly associated with dementia, and therefore
(17:04):
you should eat you know fewer or more of these
day online greens. The question that is to your questions
like why do you see this? And among educated researchers,
I think it's two things. Probably one is I think
that they might at their core believe.
Speaker 2 (17:20):
This to be true.
Speaker 5 (17:22):
And this is something I've been thinking a lot about now,
which is sort of how do your beliefs impact the
way that you do science, the way that you do research.
I mean, certainly, if you have a belief that something
is right or correct true, you might do research in
a way that validates that, and you might sort of
interpret the findings in a way that that validate that.
So I think that's one problem, and I think that
(17:44):
was something that we've maybe saw a lot of actually
during the COVID nineteen pandemic. And then the other problem
is I think that the incentives aren't there to do
something more like, you know, it takes more work to
do something clever and creative to solve a problem in
a more rigorous way. It takes more effort and more
thinking and more creativity, and if the incentives aren't there
(18:06):
to do that, then you're not going to do that.
And I think in general, medical journals don't require that
as a standard. You know, you write a lot of
economics papers. Economics papers do require that as a standard,
so you would not typically see those kinds of papers
published in the best economics journals, whereas you know, almost
every week in the best medical journals you'll see a
(18:27):
paper like that come out.
Speaker 1 (18:29):
Yeah, I mean, I think that's a that's all for me.
Speaker 3 (18:33):
Like that is an answer which is almost certainly right,
which is, you know, my incentive as a researcher is
to publish my papers in the best you know in JAMA.
And if JAMA is happy to see a correlation, then
I guess I then I then that's much easier paper
to produce than than you know, something else, And so
maybe I do that. That of course pushes the question
(18:55):
down the line to you know what, why don't the
editors of JAMA understand the difference between correlation and causality?
Speaker 1 (19:02):
A question I asked myself frequently, and.
Speaker 3 (19:07):
I get the impression when I read these papers that
people think that the kinds of empirical adjustments they do
are sufficient.
Speaker 1 (19:18):
I mean, this is like when you read these papers.
Speaker 3 (19:19):
So just to sort of back up for people who
don't spend their days in papers about coffee and longevity,
you know, it's very common in research like this for
researchers to talk about adjusting for variables. So they say,
you know, in the data, I see not only do
you eat dandelion greens, and you know how long do
you live? Or do you have dementia. Let's say, let's
(19:41):
keep with our dementia thing. So not only do I
see your dandelion greens and your dementia, but I also
see you know, whether you went to high school, and
you know, some categories of income, and your sex, and
your age and maybe your race, and maybe I see
more whatever it is, and they adjust for those They
use a statistical method to effectively try to match people
(20:05):
with similar values of those variables. It's not usually quite
that technique, but something where I basically said, I'm going
to try to hold constant these other features so I
can be closer to being causal and my view informed
by a lot of the research I've done is that
most of the time that's completely insufficient. And then in fact,
there's tons of things we don't see about people which
(20:27):
are very important for their choices of, say, what to eat,
that are not summarized. A full component of somebody's interest
in their health is not summarized by two categories of
education and three categories of income, and that these are
just really insufficient. But I wonder how much.
Speaker 1 (20:42):
Of it is that I just have a particular, I.
Speaker 3 (20:46):
Think informed, but maybe unusual view about just how important
unobservables are, and that most of the public health profession
has a different view about how important unobservables are.
Speaker 1 (20:57):
I mean, how much do you think it's that?
Speaker 5 (21:02):
I think that's part of it. But I'll tell you, you know,
as someone who spent most of my life writing the
kinds of studies where we're trying to find experiments and
publishing them mostly in medical journals, there's a lot of
papers that I've sent to journals where they come back
to me and the paper gets rejected and they'll explain
(21:25):
why they think this is not causal. And I'm always
thinking to myself, Wow, this is like definitely causal.
Speaker 2 (21:33):
It may not be important, but it's definitely.
Speaker 5 (21:36):
Causal, and we've gone through great lengths to show that
there's a causal relationship here. And it's not as if
the editors don't have an understanding of these ideas, which
is always puzzling to me, because they reject pay And
it's not just me, it's other people who I work
with who are submitting papers that are using these sort
of causal designs, and they'll get rejected. And one of
(21:59):
the reasons they'll get rich is because they'll say, you know,
we're not sure this is causal because of X, Y
and z. But then another paper which is not at
all causal will go through. And the only way I
can sort of rationalize this is to say that I
think everybody knows that these studies that we're talking about
(22:19):
are just can't be true. Like the only way this
makes sense is like they just can't be true, and
so you hold a different standard to them. Even though
the language is there which is sort of close to causal.
There's these disclaimers at the end about how the study
is not causal or you can't interpret anything about the
cause relationships here based on the study design. But I
think that they have an understanding of the issues, but
(22:40):
they are applied differently, And the question might be why
are they applied differently?
Speaker 1 (22:47):
Do you think the reason they're implied?
Speaker 3 (22:48):
First of all, that's an incredibly cynical view, although perhaps true.
But do you think the reason they're applied differently is
because you, as a journal editor, know that like if
you publish, not that your work is intimportant, but some
of the things that you work on, I think are not.
Some of the things you work on are very headline friendly,
but some of the things you work on are not
(23:08):
very headline friendly. And so you know, like you publish
a Bapuo paper, you know that it's causal, but maybe
the New York Times doesn't care you publish something that says,
you know, ultra processed foods caused dementia, or dandelion greens
don't cause dementia or whatever, like the New York Times,
well section they're calling, they're calling you, yeah, And I mean,
(23:29):
how much do you think that matters?
Speaker 5 (23:32):
I think it matters. I'm and you know, as you're
saying that, my mind is sort of racing, like how
could you how could you sort of start to study that?
I mean, I wonder if there are any shocks to
media interest in these topics or something like that where
you could really show that when the New York Times
or something like that becomes interested in a topic, that
(23:54):
you start to see more papers being published in the
top journals on that or something to that, because I
do think that's part of it that the journals are
responding to. I mean, ultimately, the job of the journals
is to create scientific information that shapes the scientific record.
That's your certainly one goal, I think, but the other
is to sort of make sure people read their journal
(24:15):
to be to feel like they're being relevant, Like that's
the core part of their business model. So if they're
not publishing things that people don't want to read, that's
a problem. And one of the things that people have
an appetite for there's a demand for is this kind
of science. Now why that is I don't know, Or
why there's a demand for this low quality science, I
(24:37):
don't know. I mean, I certainly understand why there's an
interest in understanding whether or not coffee has an effect
in your life or all these things that we are
sort of do all the time.
Speaker 1 (24:46):
Yeah, I mean, I think people.
Speaker 3 (24:50):
Part of the reason for the demand is like if
you as the now we're like on one step you
know below this, But like the New York Times would
like people to click on their things and read them.
And something that says, you know, coffee has this effect,
or you know, plastics have or whatever. It is like
something that sort of causes people to worry about a
(25:12):
topic that is relevant for their lives. That is a
great thing to get people to click on. That's like
popular click clicker location.
Speaker 2 (25:22):
Yeah, I mean there's a question of sort of what.
Speaker 5 (25:24):
I don't know that the right word here is research integrity,
But you know, I will sometimes see things where people
will landbast a journalist for talking about a study that's
clearly not causal. And my view is, like, why are
we picking on the journalists. The journalist wasn't trained to
do this, right, We should be picking on the journal itself,
the journal editors, or even the researchers who decided to
(25:47):
do that, like they really should know better, and yet
they're they're doing that. And I think that, like, again,
we can always quibble about what's an important study or
a creative study that I would have no problem debating that,
But I think that most people can agree that certain
studies are just not likely to be right, or you
(26:08):
have no idea whether they're right or wrong. And I
think people would agree with that.
Speaker 3 (26:12):
I mean, I think that's I find that very cynical,
and I think maybe a different way to say it
is I think it goes back to this thing you
said at the beginning about sort of people believe something
to be true, you know, if I believe it to
be true for whatever reason based on the outside, you know,
like based on other things I know about the world,
maybe from better studies, maybe something. If I really have
(26:33):
a core belief that like eating more ultra process foods
is bad for you, which I think is a belief
that many people have and may be a reasonable belief,
and then I publish a study in which there's a
correlation between you know, the share of your food that's
ultra processed and your health outcome, of which there are
many studies, even though almost all of those studies really
(26:55):
it's very hard to know how much of that affect
is the food and how much is all of the
other differences. And really maybe you would dig into people
and they would say, well, yeah, you know, there's probably
some other confounds that the fact that they they're more
comfortable saying to a journalist, you know, people should eat
less ulster processed foods, because they're coming in with the
belief that that is right, and that this is this
(27:18):
paper reinforces something which they really already thought it was true.
And even if you said, well, this particular paper didn't
really add that much to my knowledge, I was already
very sure that was true. And so I'm comfortable saying
that it should change people's behavior, not so much because
this particular piece of evidence that we've produced, but just
because I generally think that that's the thing people should do,
and this is an opportunity to express Then.
Speaker 5 (27:39):
Yeah, I totally agree. It's such a hard question to study.
I mean, the way I've thought about this is like,
are there particular topics where you could reasonably show that
a researcher would have a particular view in mind, and
to see whether or not their studies sort of align
with those views.
Speaker 2 (28:00):
You know.
Speaker 5 (28:01):
The example that I sometimes tell my undergraduate students is,
you know, think about sort of evidence on masking. There's
not like a ton of studies, and certainly not a
ton of high quality studies on this, But if you
were to see that individuals who published research that showed
that masks are beneficial for the prevention of COVID nineteen
(28:23):
and you look at researchers who found the opposite. If
we looked at those researchers on social media prior to
ever publishing any studies on this topic, would those who
were in the first category be more likely to be
wearing masks, to be tweeting.
Speaker 2 (28:36):
About masks things like that.
Speaker 5 (28:38):
Like, I think that there are ways to sort of
show that our researchers views about the world. And I
won't even call it ideology because there's no ideology about
coffee and longevity. It's sort of I think it could
be true. There's an incentive for me to publish this,
so I'm going to do it. But thinking about how
the incentives, whether it's career incentives, financial incentives, ideological incentives,
(29:03):
how much the role of incentives play in the questions
that people ask and the way they study them and
the answers that they find. I think is an untapped
area for research, but could be really interesting.
Speaker 3 (29:18):
Yeah, And I think, I mean there's so many pieces
of the incentives. There's my personal incentives, there's you know,
how aligned am I with some you know, with an
existing set of beliefs that that other people have. You know,
I think there is some work on like whether it
is you know, how much more difficult it is to
publish something that goes again some existing scientific paradigm, which,
of course, in some ways it should be, because if
(29:39):
you know, everyone says the sky is blue, and you
write a paper that says the sky is green, like, probably.
Speaker 1 (29:45):
We shouldn't publish your paper.
Speaker 3 (29:46):
But yeah, you know, but that kind of thing in
settings in which is less clear, can can result in
you basically sort of stasis in a in a field
until sort of something surprising happens.
Speaker 5 (30:02):
Yeah, you know what conversations I've never had. I've never
asked a journal editor.
Speaker 2 (30:06):
Why they publish these studies.
Speaker 5 (30:07):
I would love to know the answer that question.
Speaker 3 (30:10):
I would love to know the answer to that question too.
I've never managed to find someone to discuss it with me.
I mean, the other thing it's very interesting is probably
like two in the weeds for here, but like an
aspect that's totally so. In economics, like, the peer review
process is very very different than medicine, and economics like,
the peer review process takes an extremely long time, and
there are.
Speaker 1 (30:29):
Many people who have means.
Speaker 3 (30:31):
It's too extreme in some ways in my view, but
it's you know, you'll have five different people writing you
three pages about how terrible your paper is, and every
single line is the worst. You'll revise and resubmit, and
it's just like a lot of commenting and a lot
of like people reading things in great detail. The referee
process in medicine is much faster, but it also is
(30:54):
way much more about like, here is this paper, do
I like it or not?
Speaker 1 (30:59):
More or less as is? And then you know.
Speaker 3 (31:02):
That that kind of allows for less space for an
expert reviewer to fix something.
Speaker 2 (31:08):
Yeah, yeah, you know, we have this paper.
Speaker 5 (31:12):
I'll tell you the finding.
Speaker 1 (31:15):
Are you going to tell me the finding of the paper? Now?
Speaker 2 (31:17):
I'm going to tell you that.
Speaker 5 (31:17):
Oh no, I just I want to tell you another
because it does illustrate some of the things we're talking
about now.
Speaker 2 (31:21):
But then I want to do.
Speaker 1 (31:22):
That and I want to hear about this.
Speaker 5 (31:23):
Yeah, yeah, so we have this paper that shows so
a couple of years ago we had this paper in
the New Aternal Medicine that showed that kids who are
born in August have higher rates of ADHD diagnosis than
kids who were born in September, and the idea was
that in this other economists had shown this as well.
But the idea is that kids were born in August
are often the youngest kids in their class, and so
(31:46):
when they behave differently, that's sort of perceived to be
reflective of ADHD, as opposed to the recognition that these
kids are just younger for their their grade. And that
all stems from the idea that ADHD is a subjective diagnosis.
You don't see the same thing for asthma, you don't
see the same thing for diabetes, because those are more
objectively determined. And now fast forward a couple years later,
(32:08):
we have this paper now which shows that the diagnosis
of ADHD goes up on Halloween. And so we look
at millions of visits to pediatricians on Halloween versus all
the surrounding days. We definitely account for the possiblity that
parents who go to the doctor on Halloween take their
(32:30):
kid there might be different and we show that we
don't think that that's what's going on, and we show
that the diagnosis rate of ADHD is higher, and what
we think is going on is that there are some
children who the provider was thinking about this as a
possibility of a diagnosis ADHD, but they never sort of
pull the trigger on the diagnosis. But on that day,
(32:52):
the behavior the child is different for obvious reasons, right,
Like they're excited about Halloween, right, So they're just less attentive,
they're not answering questions, and so that diagnosis gets made.
Speaker 2 (33:02):
It's not a huge.
Speaker 5 (33:02):
Effect, but it's it's clearly measurable. You can see it
in the data. And you know, we write this paper
up and we don't know whether this is underdiagnosis or overdiagnosis.
Could be it could be overdiagnosis, right like something as
arbitrary as Halloween leads you to make a diagnosis that
you otherwise wouldn't have made. But it could also be underdiagnosis,
like these children could be. This could be what we
(33:24):
call a stress test, Like you do a cardiac stress
test to look for heart disease. This could be an
ADHD stress test where if that child responds in such
a way on Halloween that you now think about the
diagnosis of ADHD more firmly, that might be indicative of
something that needs to be addressed in that child. So
that could be underdiagnosis. So we write this paper and
(33:46):
we send it to a lot of places, medical journals,
all good places, and you know, most of the reviews
are actually generally like on the methods, they don't have
anything to say. But from a number of places we've
gotten feedback back, both from the editors and from sometimes
reviewers about how this finding is highly stigmatizing to the disease,
(34:06):
and we've had an extraordinarily difficult time to get it published.
Now we're trying going a different route, but you know,
my pushback has been, well, look, you know we're not
calling it. We're not saying that this over diagnoses or
this these are frivolous diagnoses. In fact, if this is
under diagnosis, you know, what does it mean to a
parent that the for their child, the only way that
(34:28):
they would have gotten diagnosed with ADHD is if by
chance they happened to go on Halloween, and otherwise it
wouldn't have been diagnosed with a condition that their child
might have. Like that arbitrariness and the diagnosis I think
should should be concerning to people. It might be inevitable,
but it should certainly be something that flags your attention.
Speaker 2 (34:48):
And we've had a really.
Speaker 5 (34:49):
Difficult time getting this thing published because, for lack of
better words, what I think are the optics. I mean,
that's what we're that's the feedback that we're getting.
Speaker 3 (34:57):
That's very frustrating because, of course, if you sort of
have the very high minded view we're all supposed to
have of science, which is like, we're going to learn things,
and we're going to get data, and then we're going
to use it to make decisions and we're going to
you know, put truth out in the world to sort
of say, well, here's the thing, Like I believe it
to be true, but I feel that people shouldn't hear
it because it will make them feel bad, or it
(35:20):
will you know, the variety of reasons that I that's very.
Speaker 1 (35:24):
That must be very frustrating.
Speaker 5 (35:26):
It is, but we keep keep watching on. But tell
me when you're ready for this other random finding.
Speaker 2 (35:31):
Yes, I'm ready.
Speaker 3 (35:32):
Okay, So, so I have one other thing I want
to I want to close this out, and then I
want to hear your random finding.
Speaker 1 (35:37):
Okay.
Speaker 3 (35:38):
So when I once asked my undergraduate this like sort
of sums up this entire conversation. I once asked my
undergraduate class we had gone through one of these papers
that I was complaining about, and then I asked them
like this sort of question we're have here, just like
sort of whose fault is this? And there was a
kid who kind of thought about it for a while
and then he basically was like, well, I think it's
(36:01):
my fault. And he said, you know, I clicked on
that headline, and that's why they put up the headline,
and that's why the person wrote the story, and that's
why the journal took the article, and that's why the
researcher wrote the article so the journal would take it,
and the journal took it so the guy would write
about it, and the person wrote about it because they
(36:21):
knew it would get a good headline, and they put
the headline on so I would click on it, and
so this is my fault. And I was like, yeah,
I guess it is. It's like, thank you for the lesson.
So that's why twenty year olds are smart.
Speaker 2 (36:36):
Yeah, that's almost like reader shaming. I don't know. I wouldn't.
I don't think it's his fault. I don't think it's
his fault at all.
Speaker 3 (36:50):
More parenting it including the importance of debunking panic headlines
for the sake of parents' mental health, and we finally
get to Bapu's groundbreaking research after the break, just to
(37:15):
sort of close it because it's a parenting podcast. I
think that so much of what I spend my time
doing is sort of dialing down people's panic around these
kind of headlines which come up all the time in parenting.
Speaker 1 (37:28):
Because if you had to.
Speaker 3 (37:30):
Pick a space in the world to write panicky headlines,
parenting would be it, because that's really what people like
to click and I guess just to like not to
put too fine a point on it, but I think
actually in the parenting space it can be quite bad
in the sense of, you know, we talk about this,
you and I talk about this like oh, like of
(37:51):
course nobody thinks that this is true, and like, you know,
researchers know it's not.
Speaker 1 (37:55):
But here's why.
Speaker 3 (37:55):
Here's the sort of interesting academic thing. But I think
for many people who who click on these adlines, who
see them, actually it does make them feel really bad,
and it does make them worry about behaviors which are
which they're already taking. You know, I let my kid
watch a screen and is it my fault that you know?
Speaker 1 (38:15):
They have ADHD?
Speaker 3 (38:16):
It's such a common question, and so it's not free
these things. This is not just an academic debate once
it gets out into the world.
Speaker 5 (38:25):
So your your statement makes me think of a like
a research question, which is, you know, I think the
journals and the people who write these articles probably view
these articles as being costless, meaning that there's no cost
imposed on society from generating this kind of knowledge. But
what if you did a study where you showed that
(38:46):
if there is a paper that comes out that shows
some purported link between autism or ADHD or something, you know,
where we don't have a good firm understanding of what's
causing it, and it links it to a very common
behavior or common thing that people might do. I wonder
if you see, like in the weeks or months following that,
(39:08):
you see you know, increased diagnoses of anxiety or depression,
or increased prescription fills among mothers versus of young children
who have that condition versus mothers of young children who
do not have that condition, or something like that. You
have to find something that was really splashy that got
people like, oh my goodness, could this possibly be true?
(39:31):
But if your sort of hypothesis is right, which I
totally think it is. I mean, you see studies like
this and you think, oh, what did I do wrong?
You didn't want to do anything wrong? Yeah, what I
do wrong? It's a paralyzing feeling, I think for a parent.
So it would be great to do that, to show
that these studies have impact. It's not you know, maybe
coffee and you know, heart attacks is not a good
(39:53):
one because no one's really going to feel bad, but
they're probably.
Speaker 3 (39:56):
Tail and all and autism Thailand. I mean, it's like
a bunch of these things where I hear about them
a lot.
Speaker 2 (40:03):
Or vaccine related things. Yeah, you know, like you know,
you know what you know?
Speaker 5 (40:07):
What is the cause of effect of vaccines on various
different or COVID nineteen vaccine of various different outcomes? If
your child had some outcome and you ascribed it to
a vaccine, your view of vaccines might be different. Your
views of the federal government might be different after that,
for reasons that were sort of totally unfounded and would
never have entered your mind had that kind of study
(40:29):
not been published.
Speaker 1 (40:32):
Okay, to close, can you please tell us your result.
I'm dying.
Speaker 5 (40:35):
Wow, Now I kind of feel bad because you're going
to see there's a link between what I literally just
said and what I'm about to say.
Speaker 2 (40:41):
All right, but that's life, all right, So let me
just frame it about it.
Speaker 5 (40:45):
There was a study I don't know, maybe ten years ago,
a long time ago, which looked at taxi cab drivers
in London and it showed that a particular part of
the brain called the hippocampus is enlarged in an enhancing
way in these drivers. And the reason why is because
(41:07):
they had to memorize all these streets they're driving around.
There is actually a test that they had to take
called the Knowledge and the idea was that because of
the hippocampus is involved in spatial processing and spatial recognition,
that that part of the brain becomes stronger, for lack
of better words, just like if you exercise certain parts
of your body, they'll become stronger. Same intuition. The hippocampus
(41:28):
has also been implicated in Alzheimer's disease. Now the study
that's coming out and it's coming out in the British
Medical Journal in a.
Speaker 2 (41:36):
Couple of weeks.
Speaker 5 (41:38):
We use this data that you're probably familiar with, where
death certificates in the United States have recently been linked
to occupation. So we're looking at these data and what
kind of questions can we answer? And I think about
this study in London taxicab drivers a long time ago,
and I'm like, well, I wonder what if we look
at taxicab drivers in the United States. What is you know,
(42:01):
what do rates of Alzheimer's dementia look like in taxicab
drivers as a cause of death compared to every other occupation.
And so we look at something like four hundred and
fifty occupations, and taxicab drivers and ambulance drivers not like EMTs,
but ambulance drivers who are just literally doing the same thing.
They have to drive all different places, they don't know
(42:21):
where they're going, they have to memorize these roads. They
have the two lowest rates of Alzheimer's related dementia contributing
to mortality than all other occupations, and we show that
they have No, they're not lower in terms of other
forms of dementia that are not thought to be related
to sort of hippocampus or what we call Alzheimer's disease.
(42:43):
And so it's a very strange finding. And of course
we're not going to come out and say you should
become a taxi cab driver. We're not going to come
out and say you should stop using Google Maps.
Speaker 3 (42:55):
I think the headline here is is, like, you know,
using Google Maps dementia.
Speaker 1 (43:01):
That's the headline I would go with.
Speaker 5 (43:02):
And I've already shortaged stocks for Google or alphabet whatever.
Speaker 2 (43:05):
It's called an anticipation of this.
Speaker 5 (43:07):
But like you know, we literally one of the last
lines to say is you know, we view this as
just generating a hypothesis. It's an idea. Think more about
how you might study this in a more causal way.
Speaker 2 (43:19):
We don't.
Speaker 5 (43:19):
We're not arguing that this is causal. We're not using
causal language. We're just saying it's an idea, a really
strange idea. And so I think that, like there is
space for correlational studies that are a answering interesting questions that
have not been already published on fifty times before in
the last two years. So there is space for those
(43:40):
studies if they're a answering or looking for interesting relationships
that haven't previously been studied, and b not claiming more
than what they're claiming and saying, look, here's what here's
what it makes us think about, here's how we should
look at this further. I think those kinds of things
are useful.
Speaker 3 (43:57):
Yeah, And I think from this there's like a million
sort of quite interesting studies one could think about doing
that would dial more into causality.
Speaker 1 (44:06):
Like the thing the obvious thing that comes to mind.
Speaker 3 (44:07):
For me is, you know, randomly in like basically fMRI,
put people in an fMRI machine and then you know,
have them do two months in which they do some
training on you know, different kinds of maps and sort
of like linking this to some of these this research
questions around you know, can we encourage people to stay
(44:28):
undamented by like pushing their brain in various ways?
Speaker 5 (44:32):
Yeah, yeah, looking at that and then maybe doing memory
assessments of these individuals. I mean, there's all sorts of
ways that you could do this, and to be honest,
probably none of them would work, but that that's totally fine.
But it's just sort of at least it's a different
idea than what we see.
Speaker 1 (44:51):
I love it, Thank you, Bapu. I really love having you.
Speaker 2 (44:55):
I love being here. Thank you. It feels like just yes, sir.
Speaker 3 (45:11):
Parent Data is produced by Tamar Avishai with support from
the Parent Data Team and PI Rex. If you have
thoughts on this episode. Please join the conversation on my
Instagram at prof Emily Oster, and if you want to
support the show, become a subscriber to the parent Data
newsletter at parentdata dot org, where I write weekly posts
on everything to do with parents and data to help
(45:33):
you make better, more informed parenting decisions. And just for fun,
I'd encourage you to go to our website and just
search for panic headlines. The articles that come up run
the gamut from whether you should worry about lead in
your tampons or white noise or breast pump bacteria or microplastics,
and after digging into the data, the answer is almost
(45:54):
always no. Cut yourself some slack at parent data dot org.
There are a lot of ways you can help people
find out about us. Leave a rating or a review
on Apple podcasts, Text your friend about something you learned
from this episode. Debate your mother in law about the
merits of something parents do now that is totally different
from what she did. Closet a story to your Instagram,
(46:16):
demunking a panic headline of your own. Just remember to
mention the podcast too, Write penelpe right, mom, We'll
Speaker 1 (46:23):
See you next time.
I'm Emily Oster. This is parent Data, and today.
Speaker 2 (00:05):
New tests reveal popular cereals and snacks could have potentially
dangerous levels of the same chemical found in weed killer.
Speaker 3 (00:17):
Here at parent Data, we talk a lot about panic headlines,
you know, the headlines.
Speaker 1 (00:23):
That cycle through your feed about coffee.
Speaker 3 (00:25):
And wine and sleep and lead and the causes of autism.
And many of the panic headlines contradict the last panic headline.
Some of them end up debunked, or maybe there was
a math error. They're not as scary as they originally seem,
but in the moment these feel so scary and urgent.
Speaker 1 (00:45):
There is new research that impacts you if you're a
coffee drinker.
Speaker 4 (00:48):
Can more screen make your child autistic? Nearly forty percent
of parents use white noise machines to help their children
sleep sounder and longer. However, some health experts are concerned
about the effects that these machines could be having on
a child's brain development.
Speaker 5 (01:03):
More and more women are actually throwing out their tampons,
opting out of using tampons for fears of toxic chemicals
that could be harmful, according to.
Speaker 1 (01:11):
A recent Consumer Reports start a news study from.
Speaker 4 (01:13):
Japan study on rats show that those raised in continuous way. Today,
researchers in Japan say the more time your child is
exposed to it, the more likely they could land on
the autism spectrum.
Speaker 1 (01:25):
The study was conducted between one.
Speaker 3 (01:30):
If you're a parent just trying to follow the science,
do what's best for your kid. Sometimes it feels like
you're being absolutely and really nonsensically bombarded with the wrong
things to do. And when it all feels like too much,
I like to call doctor Babu. Jenna Babu is an
economist like me, but unlike me, he's also a medical
(01:53):
doctor and he loves natural experiments, which means an experiment
he can't can observing people existing in naturally occurring circumstances.
For example, he has some excellent work on flu shots
showing that kids whose birthdays fall in the fall are
more likely to get their flu shots because they can
(02:15):
get them at their yearly well visits. And then he's
able to study the impact on flu shots and find,
for example, the kids are less likely to get the flu, and.
Speaker 1 (02:25):
So are their elderly relatives.
Speaker 3 (02:27):
This isn't a randomized trial that you can set up yourself.
This is finding something clever in the data that can
allow you to generate a causal estimate even though it
hasn't been explicitly randomized. If you've read Freakonomics, you're familiar
with this kind of research. And for me, I love
(02:50):
this stuff and I love BAPU, And the reason we
brought them back today is because we wanted to talk
about how research can be abused on it makes its
way into the world. I wanted to talk about how
we get to that splashy headline that warns you that
there's lead in your cheerios, or you have to throw
out all your black plastic cooking utensils, or how screens
(03:13):
give your kids autism?
Speaker 1 (03:15):
How do we get there?
Speaker 3 (03:17):
How does the research that informs that get published If
it's not the researchers aim to make us worry, they're
just curious data nerds. Why does it feel like they're
colluding with the newspapers to scare the crap out of parents.
So that's what PAPU and I are going to talk about,
what the researchers are actually aiming to do, and then
(03:38):
what happens to that research when it arrives in your
headlines to freak you out. We talk about the complicated
relationship between causality and correlation, the academic and popular incentives
to publish these kinds of headlines, and also who decides
what research is worth sharing with the world. This is
(03:59):
on the fase conversation about research, but really it's about
reassurance that there are a lot of reasons behind publishing
a story about lead and cheerios that have nothing to
do with you, or how dangerous cheerios actually are, or
whether you're a good parent who cares about the health
and well being of your kids. You are, and you
do don't throughout your cheerios. But also, if you're curious,
(04:21):
listen to the journey that headline chok to land in
your lap.
Speaker 1 (04:26):
After the break, Doctor Babu.
Speaker 3 (04:28):
Jenna, Babu Jenna, thanks for joining me.
Speaker 2 (04:39):
Thank you for having me.
Speaker 3 (04:42):
You've actually been on the Parent Data podcast before, but
for people who weren't listening, like you know, two and
a half years ago or whenever.
Speaker 1 (04:50):
I had you. Could you introduce yourself?
Speaker 2 (04:53):
First of all, it feels like yesterday, but that's right.
Speaker 1 (04:55):
Oh, it just feels it's so such a treat. You're
always in my.
Speaker 5 (04:58):
Head except yeah, my name is Babu Jenna. I'm an economist,
physician and professor at Harvard Medical School. I have a
book called Random acx of Medicine with Chris Worsham, and
a podcast called Freaconomics MD.
Speaker 1 (05:16):
I think it's important to emphasize that although you.
Speaker 3 (05:17):
Start with being an economist, you are, unlike me, also
an actual doctor.
Speaker 2 (05:24):
That's true. I'm a doctor doctor.
Speaker 1 (05:26):
Yeah, you see patients.
Speaker 2 (05:28):
I do.
Speaker 5 (05:29):
I was actually just on service for a couple of
weeks at Mass General in Boston.
Speaker 1 (05:34):
In the emergency room or no, I work.
Speaker 2 (05:37):
In the general medical floor.
Speaker 5 (05:38):
So I spent about two weeks there in a stretch,
and so I just finished up yesterday.
Speaker 3 (05:44):
Do you feel like you get a lot of ideas
for your economics work from your doctoring?
Speaker 5 (05:50):
Yeah?
Speaker 2 (05:50):
I do.
Speaker 3 (05:51):
Yeah.
Speaker 5 (05:51):
I mean there's like all sorts of random things that
you see in the hospital, things that patients will say,
or problems that you'll encounter, which I think certainly have
you given me ideas in the past. A couple of
ideas came from things I saw in the hospital.
Speaker 1 (06:05):
I'm just always curious about that.
Speaker 3 (06:07):
You seem like a person who gets many of your
research ideas from just walk like walking around in the
world or things that happened in your own life.
Speaker 5 (06:15):
Yeah, and chat GPT is feels quite good at this.
I don't know if you've tried it for coming up
with ideas, but it's not bad.
Speaker 3 (06:22):
Were you just write in like, I need a research idea.
I'm a crazy doctor who likes natural experiments. Give me
a research idea.
Speaker 2 (06:29):
Yeah.
Speaker 5 (06:29):
So, for example, I had some dinner with some students
at Harvard the other night, and we were talking about
chat GPT, and so I said to the bart you know,
suppose I want to study whether or not playing football
as a youth impacts.
Speaker 2 (06:43):
Your long term health.
Speaker 5 (06:45):
And so I asked, chat GPT, that's what I want
to know. Give me some natural experiments to try to
figure this out. And CHATCHEPT came up with a couple
of good ideas. It says, all right, well, let's look
at whether or not Pop Warner programs come into certain
at certain times. Let's look at age cutoffs, let's look
at changes in policies towards contact versus no contact football.
(07:10):
You know, all very plausible ideas. And I don't think
anybody has studied this question before, and so it was
remarkable that it was able to come up with ideas
that all made a lot of sense. None of them
I would pursue, but not bad.
Speaker 2 (07:23):
Not bad.
Speaker 3 (07:25):
I feel like my use of GPT is more limited.
Yesterday I had it make me an image of Elf
on the shelf, which was actually very was actually very good. Okay,
But the purpose of this call is not to advertise
the products of open AI, which is not a not
a sponsor, but to talk about a topic that I
(07:48):
think you and I both care a lot about, which
is why so much of the research that people hear
about in the world is just correlation and not causation.
Speaker 1 (08:03):
And you know, we've I talk about this a.
Speaker 3 (08:05):
Lot, and I guess that the I want to sort
of set the stage for the problem that we're talking about,
But then I actually want to talk about like why
this happens, which for me is the more interesting, the
more interesting question. So, uh, the canonical example of this
problem is, you know, the sort of studies show that
you know, new studies shows that you know, coffee like
(08:29):
generates makes you live, you know, generates higher longevity, or
new study shows that coffee makes you die sooner, or
you know new A lot of this stuff happens in nutrition,
and it's kind of.
Speaker 1 (08:40):
Either specific or it can be very general.
Speaker 3 (08:42):
You know, new study shows that, like people who consume
more ultra processed foods, you die sooner. And and these
are places where often when people send them to me,
then I say, you know it's correlation and not causation,
but can you give me your answer?
Speaker 1 (08:59):
You know, I'm your page.
Speaker 3 (09:00):
I come and I say, I saw this new study
that says blah blah blah about coffee, Like, how do
you help me understand what is the problem with that?
What are some of your concerns about that study?
Speaker 5 (09:11):
Well, so, typically if someone going to ask me that
kind of question, I'd say, all right, Well, if their
goal is to understand whether taking this medication, taking this supplement,
drinking coffee is going to actually improve the quality or
length of their life, how would you study it? And
I'd say to them, the only way to really study
(09:31):
this is to do a randomized trial where you take
a bunch of people and you randomize some of them
to getting.
Speaker 2 (09:40):
The product.
Speaker 5 (09:40):
In this case, it could be coffee, it could be medication,
And you take an otherwise similar group of people and
you randomize them to not getting that, and then you
study differences in whatever outcome you care about. It could
be blood pressure, could be weight, could be how long
you live. And if you see something there, then you
can say it's because of the thing that you did,
(10:02):
in this case drinking coffee. In the real world, though,
you don't have that randomization for the most part, and
so you have people who drink a lot of coffee
and people who don't. And I would explain to them
that people who drink a lot of coffee are different
than people who don't. And because of that, you don't
know if it's the coffee that's leading to this health
outcome or everything else that is different in those people
(10:24):
who drink a lot of coffee. And I think most
most people would understand that idea if told in that way.
Speaker 3 (10:31):
I mean, the other thing I will sometimes tell people
is think about this choice and think about whether it
seems like people are making it randomly. So people say,
you know, there's a relationship between family dinner, you know,
having a family dinner, homegook, family meal every night, and
outcomes for kids and your school outcomes.
Speaker 1 (10:53):
Just ask them to think about, you know.
Speaker 3 (10:55):
Once you accept the best way to do this would
be randomization, you would ask sort of compare groups who
do it and who do not. How close is that
to randomization? Like how much randomness do you think there
is in this choice? And with many of these choices
like what I eat or whether I have family dinner
or whatever it is, it doesn't really seem very random.
(11:16):
It doesn't seem like there's a very important component of
randomization there. That's like the world is not doing a
good job randomizing for you.
Speaker 5 (11:23):
Yeah, and another way to put it, and we have
a technical term that we use in our studies called
falsification tests. I would say to them, all right, suppose
that I showed you that people who drive expensive cars
live longer. Would you say to me that buying an
expensive car will make you live longer? And most people
say no, absolutely not. And I'd say, well, what if
(11:45):
it's the case that people who drink a lot of
coffee drive fancy your cars or something like that. Then
they would sort of get Okay, I get it. There's
like there's there are other factors that are going to
be correlated with some outcome, but could not plausibly be
so in a ca.
Speaker 3 (12:01):
My favorite thing I ever did about this was a
time when I took every food in the National Health
and Nutrition Examination Survey, and I correlated it with the weight,
and I showed that like, for example, dandelion greens are
associated with a lower weight, but iceberg lettuce is associated
with having a higher weight, And.
Speaker 1 (12:20):
Like you know, chemical based.
Speaker 3 (12:22):
Sugar substitutes are associated with higher weight, but plant based
sugar substitutes are associated with lower weight, which for me
was like such an illustration of like, you know, who's
eating iceberg lettuce?
Speaker 1 (12:31):
Who's eating what kind of idiots eating dandelion greens? You know,
it's like my dad and he's also doing every other thing.
Speaker 3 (12:37):
And I think that kind of thing can make this
quite vivid when you're just like, you know, well, actually,
what do you.
Speaker 1 (12:44):
Think is going on?
Speaker 3 (12:45):
And as soon as you point that out to someone,
the fancy car is a good example, it's like, oh yeah, okay,
Like it's not that this is not that complicated a
point to make two people I find.
Speaker 5 (12:57):
Yeah, yeah, but I mean, just to be fair, they're
pretty high quality evidence about dan line greens and dimension
that's not true, there's not Can I just ask a
clarifying question when you say dandelion greens, are those the
same thing as those fancy little microgreens that you get.
Speaker 3 (13:14):
No, a dan lion green is literally a part of
a dandelion. It's like instead of eating, you don't eat
the flower, but.
Speaker 1 (13:21):
You eat the greens.
Speaker 3 (13:23):
Yeah, microgreens are a delicious treat associated with uh, you know,
higher rates of IQ.
Speaker 5 (13:30):
Or I'll say the following though, that's like, you know,
the astute listener will know that my name is A
is an Indian name, and so it's not uncommon for
Indian people to take flowers and like like fry them
in some like flour like do it comes out delicious?
Speaker 2 (13:46):
I don't think it's healthy though.
Speaker 3 (13:48):
Right, the flowers might be good for you, but the
frying somewhat negates It's actually that people are frying their
iceberg lettice.
Speaker 2 (13:57):
That's exactly exactly.
Speaker 4 (14:01):
So.
Speaker 3 (14:01):
A thing I find puzzling about this space is I
feel like if you explained this, and I have to
my thirteen year old, this concept that you know, correlation
is not causation and that you would really want to
worry quite a lot about the other things that might
(14:25):
be driving these relationships.
Speaker 1 (14:26):
I feel like that is a.
Speaker 3 (14:27):
Very intuitive idea, and you know, not that like everybody
sees it immediately, but that it isn't very hard for
people to understand. Even in a you know, a short conversation,
you can kind of get people to see it. I mean,
I think, you know, I spend a lot of time
on Instagram short form Instagram videos.
Speaker 1 (14:43):
Explaining this, and I think some of the time it hits.
Speaker 3 (14:46):
The question I find more puzzling is why are there
so many papers which do this, which are written by
people who haven't just watched a forty five second Instagram
video or had one conversation with you, but who have
spent literally many, many years taking classes on these exact
(15:06):
topics and then writing papers. And these people are, you know,
professors or like, they have a lot of education, They're smart,
and I just fundamentally don't understand why so much of
this literature keeps happening, and whether the people who are
writing it think that they are uncovering something causal or not.
Speaker 1 (15:32):
Do you have any insight?
Speaker 5 (15:34):
I have answers, I don't know if they're insightful, But well,
let me just say the following. First of all, and
at some point later day, I'm going to tell you
about a paper that we have coming out in probably
a couple of weeks. It will be December, so I
don't know when this will air, but it will fall
into this category of a very weird sort of correlation.
Speaker 2 (15:54):
And the.
Speaker 5 (15:57):
Way that we would describe it in the way that
we do justscribed as finding, which I know I'm making
it sort of sound very sort of like wow, what's
what's he going to drop on me in a moment?
Is it's just a hypothesis. So I would have no
problem if people would do the kinds of studies that
you're describing, which are purely correlational studies, no sort of experiment,
(16:18):
serving as the backbone behind the idea and saying, look,
here's this interesting observation. We don't know what to make
of it. It could be causal for X, y Z
reason like there's a there's a channel by which this thing,
in this case dandelions, could have an effect on some
health outcome and it would work by this mechanism. You know,
(16:41):
this flower affects this protein in the body. This protein
in the body affects these cells. These cells affect the
development dementia something like that. Totally fine with that, because
then a reader could say, all right, well, you know
I want to investigate this or not investigate and investigate
this any further. But that's not the way the studies
come out. The studies come out as you know, danne
greens are you know, strongly associated with dementia, and therefore
(17:04):
you should eat you know fewer or more of these
day online greens. The question that is to your questions
like why do you see this? And among educated researchers,
I think it's two things. Probably one is I think
that they might at their core believe.
Speaker 2 (17:20):
This to be true.
Speaker 5 (17:22):
And this is something I've been thinking a lot about now,
which is sort of how do your beliefs impact the
way that you do science, the way that you do research.
I mean, certainly, if you have a belief that something
is right or correct true, you might do research in
a way that validates that, and you might sort of
interpret the findings in a way that that validate that.
So I think that's one problem, and I think that
(17:44):
was something that we've maybe saw a lot of actually
during the COVID nineteen pandemic. And then the other problem
is I think that the incentives aren't there to do
something more like, you know, it takes more work to
do something clever and creative to solve a problem in
a more rigorous way. It takes more effort and more
thinking and more creativity, and if the incentives aren't there
(18:06):
to do that, then you're not going to do that.
And I think in general, medical journals don't require that
as a standard. You know, you write a lot of
economics papers. Economics papers do require that as a standard,
so you would not typically see those kinds of papers
published in the best economics journals, whereas you know, almost
every week in the best medical journals you'll see a
(18:27):
paper like that come out.
Speaker 1 (18:29):
Yeah, I mean, I think that's a that's all for me.
Speaker 3 (18:33):
Like that is an answer which is almost certainly right,
which is, you know, my incentive as a researcher is
to publish my papers in the best you know in JAMA.
And if JAMA is happy to see a correlation, then
I guess I then I then that's much easier paper
to produce than than you know, something else, And so
maybe I do that. That of course pushes the question
(18:55):
down the line to you know what, why don't the
editors of JAMA understand the difference between correlation and causality?
Speaker 1 (19:02):
A question I asked myself frequently, and.
Speaker 3 (19:07):
I get the impression when I read these papers that
people think that the kinds of empirical adjustments they do
are sufficient.
Speaker 1 (19:18):
I mean, this is like when you read these papers.
Speaker 3 (19:19):
So just to sort of back up for people who
don't spend their days in papers about coffee and longevity,
you know, it's very common in research like this for
researchers to talk about adjusting for variables. So they say,
you know, in the data, I see not only do
you eat dandelion greens, and you know how long do
you live? Or do you have dementia. Let's say, let's
(19:41):
keep with our dementia thing. So not only do I
see your dandelion greens and your dementia, but I also
see you know, whether you went to high school, and
you know, some categories of income, and your sex, and
your age and maybe your race, and maybe I see
more whatever it is, and they adjust for those They
use a statistical method to effectively try to match people
(20:05):
with similar values of those variables. It's not usually quite
that technique, but something where I basically said, I'm going
to try to hold constant these other features so I
can be closer to being causal and my view informed
by a lot of the research I've done is that
most of the time that's completely insufficient. And then in fact,
there's tons of things we don't see about people which
(20:27):
are very important for their choices of, say, what to eat,
that are not summarized. A full component of somebody's interest
in their health is not summarized by two categories of
education and three categories of income, and that these are
just really insufficient. But I wonder how much.
Speaker 1 (20:42):
Of it is that I just have a particular, I.
Speaker 3 (20:46):
Think informed, but maybe unusual view about just how important
unobservables are, and that most of the public health profession
has a different view about how important unobservables are.
Speaker 1 (20:57):
I mean, how much do you think it's that?
Speaker 5 (21:02):
I think that's part of it. But I'll tell you, you know,
as someone who spent most of my life writing the
kinds of studies where we're trying to find experiments and
publishing them mostly in medical journals, there's a lot of
papers that I've sent to journals where they come back
to me and the paper gets rejected and they'll explain
(21:25):
why they think this is not causal. And I'm always
thinking to myself, Wow, this is like definitely causal.
Speaker 2 (21:33):
It may not be important, but it's definitely.
Speaker 5 (21:36):
Causal, and we've gone through great lengths to show that
there's a causal relationship here. And it's not as if
the editors don't have an understanding of these ideas, which
is always puzzling to me, because they reject pay And
it's not just me, it's other people who I work
with who are submitting papers that are using these sort
of causal designs, and they'll get rejected. And one of
(21:59):
the reasons they'll get rich is because they'll say, you know,
we're not sure this is causal because of X, Y
and z. But then another paper which is not at
all causal will go through. And the only way I
can sort of rationalize this is to say that I
think everybody knows that these studies that we're talking about
(22:19):
are just can't be true. Like the only way this
makes sense is like they just can't be true, and
so you hold a different standard to them. Even though
the language is there which is sort of close to causal.
There's these disclaimers at the end about how the study
is not causal or you can't interpret anything about the
cause relationships here based on the study design. But I
think that they have an understanding of the issues, but
(22:40):
they are applied differently, And the question might be why
are they applied differently?
Speaker 1 (22:47):
Do you think the reason they're implied?
Speaker 3 (22:48):
First of all, that's an incredibly cynical view, although perhaps true.
But do you think the reason they're applied differently is
because you, as a journal editor, know that like if
you publish, not that your work is intimportant, but some
of the things that you work on, I think are not.
Some of the things you work on are very headline friendly,
but some of the things you work on are not
(23:08):
very headline friendly. And so you know, like you publish
a Bapuo paper, you know that it's causal, but maybe
the New York Times doesn't care you publish something that says,
you know, ultra processed foods caused dementia, or dandelion greens
don't cause dementia or whatever, like the New York Times,
well section they're calling, they're calling you, yeah, And I mean,
(23:29):
how much do you think that matters?
Speaker 5 (23:32):
I think it matters. I'm and you know, as you're
saying that, my mind is sort of racing, like how
could you how could you sort of start to study that?
I mean, I wonder if there are any shocks to
media interest in these topics or something like that where
you could really show that when the New York Times
or something like that becomes interested in a topic, that
(23:54):
you start to see more papers being published in the
top journals on that or something to that, because I
do think that's part of it that the journals are
responding to. I mean, ultimately, the job of the journals
is to create scientific information that shapes the scientific record.
That's your certainly one goal, I think, but the other
is to sort of make sure people read their journal
(24:15):
to be to feel like they're being relevant, Like that's
the core part of their business model. So if they're
not publishing things that people don't want to read, that's
a problem. And one of the things that people have
an appetite for there's a demand for is this kind
of science. Now why that is I don't know, Or
why there's a demand for this low quality science, I
(24:37):
don't know. I mean, I certainly understand why there's an
interest in understanding whether or not coffee has an effect
in your life or all these things that we are
sort of do all the time.
Speaker 1 (24:46):
Yeah, I mean, I think people.
Speaker 3 (24:50):
Part of the reason for the demand is like if
you as the now we're like on one step you
know below this, But like the New York Times would
like people to click on their things and read them.
And something that says, you know, coffee has this effect,
or you know, plastics have or whatever. It is like
something that sort of causes people to worry about a
(25:12):
topic that is relevant for their lives. That is a
great thing to get people to click on. That's like
popular click clicker location.
Speaker 2 (25:22):
Yeah, I mean there's a question of sort of what.
Speaker 5 (25:24):
I don't know that the right word here is research integrity,
But you know, I will sometimes see things where people
will landbast a journalist for talking about a study that's
clearly not causal. And my view is, like, why are
we picking on the journalists. The journalist wasn't trained to
do this, right, We should be picking on the journal itself,
the journal editors, or even the researchers who decided to
(25:47):
do that, like they really should know better, and yet
they're they're doing that. And I think that, like, again,
we can always quibble about what's an important study or
a creative study that I would have no problem debating that,
But I think that most people can agree that certain
studies are just not likely to be right, or you
(26:08):
have no idea whether they're right or wrong. And I
think people would agree with that.
Speaker 3 (26:12):
I mean, I think that's I find that very cynical,
and I think maybe a different way to say it
is I think it goes back to this thing you
said at the beginning about sort of people believe something
to be true, you know, if I believe it to
be true for whatever reason based on the outside, you know,
like based on other things I know about the world,
maybe from better studies, maybe something. If I really have
(26:33):
a core belief that like eating more ultra process foods
is bad for you, which I think is a belief
that many people have and may be a reasonable belief,
and then I publish a study in which there's a
correlation between you know, the share of your food that's
ultra processed and your health outcome, of which there are
many studies, even though almost all of those studies really
(26:55):
it's very hard to know how much of that affect
is the food and how much is all of the
other differences. And really maybe you would dig into people
and they would say, well, yeah, you know, there's probably
some other confounds that the fact that they they're more
comfortable saying to a journalist, you know, people should eat
less ulster processed foods, because they're coming in with the
belief that that is right, and that this is this
(27:18):
paper reinforces something which they really already thought it was true.
And even if you said, well, this particular paper didn't
really add that much to my knowledge, I was already
very sure that was true. And so I'm comfortable saying
that it should change people's behavior, not so much because
this particular piece of evidence that we've produced, but just
because I generally think that that's the thing people should do,
and this is an opportunity to express Then.
Speaker 5 (27:39):
Yeah, I totally agree. It's such a hard question to study.
I mean, the way I've thought about this is like,
are there particular topics where you could reasonably show that
a researcher would have a particular view in mind, and
to see whether or not their studies sort of align
with those views.
Speaker 2 (28:00):
You know.
Speaker 5 (28:01):
The example that I sometimes tell my undergraduate students is,
you know, think about sort of evidence on masking. There's
not like a ton of studies, and certainly not a
ton of high quality studies on this, But if you
were to see that individuals who published research that showed
that masks are beneficial for the prevention of COVID nineteen
(28:23):
and you look at researchers who found the opposite. If
we looked at those researchers on social media prior to
ever publishing any studies on this topic, would those who
were in the first category be more likely to be
wearing masks, to be tweeting.
Speaker 2 (28:36):
About masks things like that.
Speaker 5 (28:38):
Like, I think that there are ways to sort of
show that our researchers views about the world. And I
won't even call it ideology because there's no ideology about
coffee and longevity. It's sort of I think it could
be true. There's an incentive for me to publish this,
so I'm going to do it. But thinking about how
the incentives, whether it's career incentives, financial incentives, ideological incentives,
(29:03):
how much the role of incentives play in the questions
that people ask and the way they study them and
the answers that they find. I think is an untapped
area for research, but could be really interesting.
Speaker 3 (29:18):
Yeah, And I think, I mean there's so many pieces
of the incentives. There's my personal incentives, there's you know,
how aligned am I with some you know, with an
existing set of beliefs that that other people have. You know,
I think there is some work on like whether it
is you know, how much more difficult it is to
publish something that goes again some existing scientific paradigm, which,
of course, in some ways it should be, because if
(29:39):
you know, everyone says the sky is blue, and you
write a paper that says the sky is green, like, probably.
Speaker 1 (29:45):
We shouldn't publish your paper.
Speaker 3 (29:46):
But yeah, you know, but that kind of thing in
settings in which is less clear, can can result in
you basically sort of stasis in a in a field
until sort of something surprising happens.
Speaker 5 (30:02):
Yeah, you know what conversations I've never had. I've never
asked a journal editor.
Speaker 2 (30:06):
Why they publish these studies.
Speaker 5 (30:07):
I would love to know the answer that question.
Speaker 3 (30:10):
I would love to know the answer to that question too.
I've never managed to find someone to discuss it with me.
I mean, the other thing it's very interesting is probably
like two in the weeds for here, but like an
aspect that's totally so. In economics, like, the peer review
process is very very different than medicine, and economics like,
the peer review process takes an extremely long time, and
there are.
Speaker 1 (30:29):
Many people who have means.
Speaker 3 (30:31):
It's too extreme in some ways in my view, but
it's you know, you'll have five different people writing you
three pages about how terrible your paper is, and every
single line is the worst. You'll revise and resubmit, and
it's just like a lot of commenting and a lot
of like people reading things in great detail. The referee
process in medicine is much faster, but it also is
(30:54):
way much more about like, here is this paper, do
I like it or not?
Speaker 1 (30:59):
More or less as is? And then you know.
Speaker 3 (31:02):
That that kind of allows for less space for an
expert reviewer to fix something.
Speaker 2 (31:08):
Yeah, yeah, you know, we have this paper.
Speaker 5 (31:12):
I'll tell you the finding.
Speaker 1 (31:15):
Are you going to tell me the finding of the paper? Now?
Speaker 2 (31:17):
I'm going to tell you that.
Speaker 5 (31:17):
Oh no, I just I want to tell you another
because it does illustrate some of the things we're talking
about now.
Speaker 2 (31:21):
But then I want to do.
Speaker 1 (31:22):
That and I want to hear about this.
Speaker 5 (31:23):
Yeah, yeah, so we have this paper that shows so
a couple of years ago we had this paper in
the New Aternal Medicine that showed that kids who are
born in August have higher rates of ADHD diagnosis than
kids who were born in September, and the idea was
that in this other economists had shown this as well.
But the idea is that kids were born in August
are often the youngest kids in their class, and so
(31:46):
when they behave differently, that's sort of perceived to be
reflective of ADHD, as opposed to the recognition that these
kids are just younger for their their grade. And that
all stems from the idea that ADHD is a subjective diagnosis.
You don't see the same thing for asthma, you don't
see the same thing for diabetes, because those are more
objectively determined. And now fast forward a couple years later,
(32:08):
we have this paper now which shows that the diagnosis
of ADHD goes up on Halloween. And so we look
at millions of visits to pediatricians on Halloween versus all
the surrounding days. We definitely account for the possiblity that
parents who go to the doctor on Halloween take their
(32:30):
kid there might be different and we show that we
don't think that that's what's going on, and we show
that the diagnosis rate of ADHD is higher, and what
we think is going on is that there are some
children who the provider was thinking about this as a
possibility of a diagnosis ADHD, but they never sort of
pull the trigger on the diagnosis. But on that day,
(32:52):
the behavior the child is different for obvious reasons, right,
Like they're excited about Halloween, right, So they're just less attentive,
they're not answering questions, and so that diagnosis gets made.
Speaker 2 (33:02):
It's not a huge.
Speaker 5 (33:02):
Effect, but it's it's clearly measurable. You can see it
in the data. And you know, we write this paper
up and we don't know whether this is underdiagnosis or overdiagnosis.
Could be it could be overdiagnosis, right like something as
arbitrary as Halloween leads you to make a diagnosis that
you otherwise wouldn't have made. But it could also be underdiagnosis,
like these children could be. This could be what we
(33:24):
call a stress test, Like you do a cardiac stress
test to look for heart disease. This could be an
ADHD stress test where if that child responds in such
a way on Halloween that you now think about the
diagnosis of ADHD more firmly, that might be indicative of
something that needs to be addressed in that child. So
that could be underdiagnosis. So we write this paper and
(33:46):
we send it to a lot of places, medical journals,
all good places, and you know, most of the reviews
are actually generally like on the methods, they don't have
anything to say. But from a number of places we've
gotten feedback back, both from the editors and from sometimes
reviewers about how this finding is highly stigmatizing to the disease,
(34:06):
and we've had an extraordinarily difficult time to get it published.
Now we're trying going a different route, but you know,
my pushback has been, well, look, you know we're not
calling it. We're not saying that this over diagnoses or
this these are frivolous diagnoses. In fact, if this is
under diagnosis, you know, what does it mean to a
parent that the for their child, the only way that
(34:28):
they would have gotten diagnosed with ADHD is if by
chance they happened to go on Halloween, and otherwise it
wouldn't have been diagnosed with a condition that their child
might have. Like that arbitrariness and the diagnosis I think
should should be concerning to people. It might be inevitable,
but it should certainly be something that flags your attention.
Speaker 2 (34:48):
And we've had a really.
Speaker 5 (34:49):
Difficult time getting this thing published because, for lack of
better words, what I think are the optics. I mean,
that's what we're that's the feedback that we're getting.
Speaker 3 (34:57):
That's very frustrating because, of course, if you sort of
have the very high minded view we're all supposed to
have of science, which is like, we're going to learn things,
and we're going to get data, and then we're going
to use it to make decisions and we're going to
you know, put truth out in the world to sort
of say, well, here's the thing, Like I believe it
to be true, but I feel that people shouldn't hear
it because it will make them feel bad, or it
(35:20):
will you know, the variety of reasons that I that's very.
Speaker 1 (35:24):
That must be very frustrating.
Speaker 5 (35:26):
It is, but we keep keep watching on. But tell
me when you're ready for this other random finding.
Speaker 2 (35:31):
Yes, I'm ready.
Speaker 3 (35:32):
Okay, So, so I have one other thing I want
to I want to close this out, and then I
want to hear your random finding.
Speaker 1 (35:37):
Okay.
Speaker 3 (35:38):
So when I once asked my undergraduate this like sort
of sums up this entire conversation. I once asked my
undergraduate class we had gone through one of these papers
that I was complaining about, and then I asked them
like this sort of question we're have here, just like
sort of whose fault is this? And there was a
kid who kind of thought about it for a while
and then he basically was like, well, I think it's
(36:01):
my fault. And he said, you know, I clicked on
that headline, and that's why they put up the headline,
and that's why the person wrote the story, and that's
why the journal took the article, and that's why the
researcher wrote the article so the journal would take it,
and the journal took it so the guy would write
about it, and the person wrote about it because they
(36:21):
knew it would get a good headline, and they put
the headline on so I would click on it, and
so this is my fault. And I was like, yeah,
I guess it is. It's like, thank you for the lesson.
So that's why twenty year olds are smart.
Speaker 2 (36:36):
Yeah, that's almost like reader shaming. I don't know. I wouldn't.
I don't think it's his fault. I don't think it's
his fault at all.
Speaker 3 (36:50):
More parenting it including the importance of debunking panic headlines
for the sake of parents' mental health, and we finally
get to Bapu's groundbreaking research after the break, just to
(37:15):
sort of close it because it's a parenting podcast. I
think that so much of what I spend my time
doing is sort of dialing down people's panic around these
kind of headlines which come up all the time in parenting.
Speaker 1 (37:28):
Because if you had to.
Speaker 3 (37:30):
Pick a space in the world to write panicky headlines,
parenting would be it, because that's really what people like
to click and I guess just to like not to
put too fine a point on it, but I think
actually in the parenting space it can be quite bad
in the sense of, you know, we talk about this,
you and I talk about this like oh, like of
(37:51):
course nobody thinks that this is true, and like, you know,
researchers know it's not.
Speaker 1 (37:55):
But here's why.
Speaker 3 (37:55):
Here's the sort of interesting academic thing. But I think
for many people who who click on these adlines, who
see them, actually it does make them feel really bad,
and it does make them worry about behaviors which are
which they're already taking. You know, I let my kid
watch a screen and is it my fault that you know?
Speaker 1 (38:15):
They have ADHD?
Speaker 3 (38:16):
It's such a common question, and so it's not free
these things. This is not just an academic debate once
it gets out into the world.
Speaker 5 (38:25):
So your your statement makes me think of a like
a research question, which is, you know, I think the
journals and the people who write these articles probably view
these articles as being costless, meaning that there's no cost
imposed on society from generating this kind of knowledge. But
what if you did a study where you showed that
(38:46):
if there is a paper that comes out that shows
some purported link between autism or ADHD or something, you know,
where we don't have a good firm understanding of what's
causing it, and it links it to a very common
behavior or common thing that people might do. I wonder
if you see, like in the weeks or months following that,
(39:08):
you see you know, increased diagnoses of anxiety or depression,
or increased prescription fills among mothers versus of young children
who have that condition versus mothers of young children who
do not have that condition, or something like that. You
have to find something that was really splashy that got
people like, oh my goodness, could this possibly be true?
(39:31):
But if your sort of hypothesis is right, which I
totally think it is. I mean, you see studies like
this and you think, oh, what did I do wrong?
You didn't want to do anything wrong? Yeah, what I
do wrong? It's a paralyzing feeling, I think for a parent.
So it would be great to do that, to show
that these studies have impact. It's not you know, maybe
coffee and you know, heart attacks is not a good
(39:53):
one because no one's really going to feel bad, but
they're probably.
Speaker 3 (39:56):
Tail and all and autism Thailand. I mean, it's like
a bunch of these things where I hear about them
a lot.
Speaker 2 (40:03):
Or vaccine related things. Yeah, you know, like you know,
you know what you know?
Speaker 5 (40:07):
What is the cause of effect of vaccines on various
different or COVID nineteen vaccine of various different outcomes? If
your child had some outcome and you ascribed it to
a vaccine, your view of vaccines might be different. Your
views of the federal government might be different after that,
for reasons that were sort of totally unfounded and would
never have entered your mind had that kind of study
(40:29):
not been published.
Speaker 1 (40:32):
Okay, to close, can you please tell us your result.
I'm dying.
Speaker 5 (40:35):
Wow, Now I kind of feel bad because you're going
to see there's a link between what I literally just
said and what I'm about to say.
Speaker 2 (40:41):
All right, but that's life, all right, So let me
just frame it about it.
Speaker 5 (40:45):
There was a study I don't know, maybe ten years ago,
a long time ago, which looked at taxi cab drivers
in London and it showed that a particular part of
the brain called the hippocampus is enlarged in an enhancing
way in these drivers. And the reason why is because
(41:07):
they had to memorize all these streets they're driving around.
There is actually a test that they had to take
called the Knowledge and the idea was that because of
the hippocampus is involved in spatial processing and spatial recognition,
that that part of the brain becomes stronger, for lack
of better words, just like if you exercise certain parts
of your body, they'll become stronger. Same intuition. The hippocampus
(41:28):
has also been implicated in Alzheimer's disease. Now the study
that's coming out and it's coming out in the British
Medical Journal in a.
Speaker 2 (41:36):
Couple of weeks.
Speaker 5 (41:38):
We use this data that you're probably familiar with, where
death certificates in the United States have recently been linked
to occupation. So we're looking at these data and what
kind of questions can we answer? And I think about
this study in London taxicab drivers a long time ago,
and I'm like, well, I wonder what if we look
at taxicab drivers in the United States. What is you know,
(42:01):
what do rates of Alzheimer's dementia look like in taxicab
drivers as a cause of death compared to every other occupation.
And so we look at something like four hundred and
fifty occupations, and taxicab drivers and ambulance drivers not like EMTs,
but ambulance drivers who are just literally doing the same thing.
They have to drive all different places, they don't know
(42:21):
where they're going, they have to memorize these roads. They
have the two lowest rates of Alzheimer's related dementia contributing
to mortality than all other occupations, and we show that
they have No, they're not lower in terms of other
forms of dementia that are not thought to be related
to sort of hippocampus or what we call Alzheimer's disease.
(42:43):
And so it's a very strange finding. And of course
we're not going to come out and say you should
become a taxi cab driver. We're not going to come
out and say you should stop using Google Maps.
Speaker 3 (42:55):
I think the headline here is is, like, you know,
using Google Maps dementia.
Speaker 1 (43:01):
That's the headline I would go with.
Speaker 5 (43:02):
And I've already shortaged stocks for Google or alphabet whatever.
Speaker 2 (43:05):
It's called an anticipation of this.
Speaker 5 (43:07):
But like you know, we literally one of the last
lines to say is you know, we view this as
just generating a hypothesis. It's an idea. Think more about
how you might study this in a more causal way.
Speaker 2 (43:19):
We don't.
Speaker 5 (43:19):
We're not arguing that this is causal. We're not using
causal language. We're just saying it's an idea, a really
strange idea. And so I think that, like there is
space for correlational studies that are a answering interesting questions that
have not been already published on fifty times before in
the last two years. So there is space for those
(43:40):
studies if they're a answering or looking for interesting relationships
that haven't previously been studied, and b not claiming more
than what they're claiming and saying, look, here's what here's
what it makes us think about, here's how we should
look at this further. I think those kinds of things
are useful.
Speaker 3 (43:57):
Yeah, And I think from this there's like a million
sort of quite interesting studies one could think about doing
that would dial more into causality.
Speaker 1 (44:06):
Like the thing the obvious thing that comes to mind.
Speaker 3 (44:07):
For me is, you know, randomly in like basically fMRI,
put people in an fMRI machine and then you know,
have them do two months in which they do some
training on you know, different kinds of maps and sort
of like linking this to some of these this research
questions around you know, can we encourage people to stay
(44:28):
undamented by like pushing their brain in various ways?
Speaker 5 (44:32):
Yeah, yeah, looking at that and then maybe doing memory
assessments of these individuals. I mean, there's all sorts of
ways that you could do this, and to be honest,
probably none of them would work, but that that's totally fine.
But it's just sort of at least it's a different
idea than what we see.
Speaker 1 (44:51):
I love it, Thank you, Bapu. I really love having you.
Speaker 2 (44:55):
I love being here. Thank you. It feels like just yes, sir.
Speaker 3 (45:11):
Parent Data is produced by Tamar Avishai with support from
the Parent Data Team and PI Rex. If you have
thoughts on this episode. Please join the conversation on my
Instagram at prof Emily Oster, and if you want to
support the show, become a subscriber to the parent Data
newsletter at parentdata dot org, where I write weekly posts
on everything to do with parents and data to help
(45:33):
you make better, more informed parenting decisions. And just for fun,
I'd encourage you to go to our website and just
search for panic headlines. The articles that come up run
the gamut from whether you should worry about lead in
your tampons or white noise or breast pump bacteria or microplastics,
and after digging into the data, the answer is almost
(45:54):
always no. Cut yourself some slack at parent data dot org.
There are a lot of ways you can help people
find out about us. Leave a rating or a review
on Apple podcasts, Text your friend about something you learned
from this episode. Debate your mother in law about the
merits of something parents do now that is totally different
from what she did. Closet a story to your Instagram,
(46:16):
demunking a panic headline of your own. Just remember to
mention the podcast too, Write penelpe right, mom, We'll
Speaker 1 (46:23):
See you next time.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
Kingdom of Fraud
It’s the unlikeliest of criminal partnerships: a devout polygamist from an insular Utah sect joining forces with a shadowy Armenian tycoon from LA. The result - a billion dollar fraud conspiracy. In Kingdom of Fraud, investigative reporter Michele McPhee traces the origins of the extraordinary alliance between Jacob Kingston and Levon Termendzhyan. Together, the two men trigger the largest tax investigation in American history and weave around themselves a web of dirty cops, influential political relationships and transnational money laundering. All this is set against the backdrop of Jacob Kingston’s clan – The Order. A powerful and secretive polygamist organization in Salt Lake City. To whom Jacob is desperate to prove his worth. Kingdom of Fraud is produced by Novel for iHeart Podcasts. For more from Novel, visit https://novel.audio/. You can listen to new episodes of Kingdom of Fraud completely ad-free and 1 week early with an iHeart True Crime+ subscription, available exclusively on Apple Podcasts. Open your Apple Podcasts app, search for “iHeart True Crime+, and subscribe today!