July 27, 2025 • 18 mins
🎙️ Episode 88: Stable Heritability of Type 1 Diabetes Over Three Decades in Sweden
🧬 In this episode of PaperCast Base by Base, we explore a Swedish nationwide register-based study that assesses whether the heritability of childhood-onset type 1 diabetes has changed from 1982 to 2010 and how environmental factors have contributed to the rising incidence.
🔍 Study Highlights:
Leveraging data from nearly three million children born in Sweden between 1982 and 2010, the authors estimated heritability and non-shared environmental contributions to type 1 diabetes variance using sibling-based liability threshold models.
They found that the heritability of childhood-onset type 1 diabetes remained consistently high at around 0.83 across birth cohorts.
Age-specific analyses revealed higher genetic contributions in younger-onset cases and stable heritability across age groups.
Environmental exposures including maternal smoking during pregnancy and childhood adiposity collectively explained less than 10% of the observed increase in incidence between 1982 and 2000.
🧠 Conclusion:
These results underscore the enduring role of genetic factors in type 1 diabetes susceptibility and highlight the need to identify novel environmental triggers to explain the rising incidence.
📖 Reference:
Wei Y, Andersson T, Liu S, Feychting M, Kuja-Halkola R, Carlsson S. Stable heritability of type 1 diabetes in a Swedish nationwide cohort study. Nat Commun. 2025;16:5327. doi:10.1038/s41467-025-60813-2.
📜 License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:14):
Welcome to Base by Base, the AERcast that brings genomics to
you, wherever you are. Today, we're embarking on a deep
dive into, well, a really groundbreaking scientific study.
It challenges some, some pretty deeply held assumptions about a
common childhood illness. OK, picture this.
A common childhood condition, right?
One that's been steadily increasing, going up and up for
(00:35):
decades now. What do you think is driving
that rise? You know, is it our environment,
changes in our genes, or maybe something else entirely?
We're talking about childhood onset type 1 diabetes.
Its incidence has been climbing worldwide since like the 1950s,
increasing by maybe 3-4 percent every year.
That's not a small shift. It's a really significant global
(00:56):
trend, and it's frankly perplexed researchers.
For years now. Conventional wisdom often points
the finger at, you know, our changing environment, our diets,
maybe early life exposures, things like that.
But what if the story is much more complex, maybe even
counterintuitive? What if the thing we thought was
driving this trend? What if it's actually been
remarkably stable all along? Today, we're going to unpack a
study that sheds new light on this exotic question.
(01:18):
It challenges some long held beliefs and honestly forces us
to reconsider where we should belooking.
But before we get into the the fascinating details, we really
want to recognize the incredibleminds behind this work.
Today we celebrate the work of Yuksi Away, Thomas Anderson,
Sheng Sinlu, Maria Fakting, Ralph Kuja Halkola and Sophia
Carlson from the Karolinska Institute in Sweden.
(01:40):
They've really advanced our understanding of the well, the
changing influences on childhoodonset type 1 diabetes.
OK, let's unpack the central scientific debate here.
We know type one that diabetes incidence is rising globally,
but what's been the main hypothesis explained well, the.
Core problem has been, it's beenpretty unclear, is this global
increase in childhood T1D because of rising environmental
(02:02):
risk factors, non genetic ones? Or is it maybe increasing
genetic susceptibility in the population?
Or, you know, perhaps some complex mix of both?
Right. And when you see a disease
increase that rapidly, it feels natural to look at environmental
changes first. But why was that the dominant
idea for SO? Long that's true.
Environmental explanations have generally been favored.
(02:23):
And the reasoning? It seems pretty sound on the
surface. It just seems highly unlikely
that the genetic makeup of a whole population would change so
drastically, so quickly. You know, in just a few decades
plus, you see these big differences in T1D incidences
even between populations that are genetically quite similar.
Think about Finland having a much higher rate than, say,
(02:44):
Estonia, even though they're geographically and ethnically
pretty close. That points to external factors.
Right. Sure, that makes sense.
So what specific environmental factors were kind of on the
radar, potential culprits or maybe even things that protect
against it? Yeah, several stood out.
For instance, the rise in T1D happened alongside the rise in
childhood obesity that's often proposed as a risk factor,
(03:05):
though the causal link is still debated.
Other early life things often discussed include maternal
smoking, which interestingly hassometimes been suggested as
potentially protective, believe it or not.
Also higher maternal agent delivery that's seen as a risk,
and breastfeeding, generally thought to be protective.
The big question has always been, you know, how much have
changes in these factors actually contributed to the
(03:26):
overall rise in T1D? And key to the study is the idea
of heritability. Could you break that down for
us? What does it mean here and why
is it so crucial? Right, heritability.
In simple terms, it's the proportion of the variation we
see in a disease like T1D. How much of that variation can
be explained by all the genetic factors combined.
It's not about whether you specifically will get it based
(03:47):
on genes, but about how much of the difference between people
who get it and people who don't within a population is down to
genetic differences. For T1D, studies have estimated
heritability to be pretty high, actually, somewhere between say
50% and 88% in different groups.A recent Swedish study put it at
81% in kits. Now this is crucial because,
(04:08):
well, if environmental factors are really the main drivers
pushing incidents up, then you'dexpect their relative
contribution to increase over time, and logically heritability
should decrease. So the key question this study
asked was, has T1D heritability actually changed over the last
few decades? Because as far as they knew,
nobody had really looked at thiscomprehensively before.
That's a massive question to tackle.
And Speaking of genetics, I remember reading about HLA
(04:31):
genotypes. HLA human leukocyte antigen
genes right key for the immune system.
Certain versions are known risksfor autoimmune diseases like
T1D. Did those earlier HLA studies
give any clues? They did, yeah.
Yeah, Some intriguing hints. A study in Australia and similar
findings came out of Finland, Sweden, the USUK.
They found something kind of counterintuitive.
(04:52):
The proportion of T1D cases withthe highest risk HLA types
actually decreased from the 80s to the 2000s.
Meanwhile, the proportion with intermediate risk HLA types went
up. So it suggests the HLA
contribution shifted, but didn'tnecessarily lessen the overall
genetic risk. It just changed which genetic
profiles were most affected. Maybe.
And it's important to remember, HLA genes only account for about
(05:14):
half the total genetic risk for T1D.
There's still a lot more genetics involved.
So moving to their methods, whatreally makes the study stand out
is how they approach this. They built their analysis on an
incredibly solid foundation. Their goal was clear, figure out
if environmental and genetic influences on childhood T1D,
specifically kids aged zero to 18, have changed in Sweden over
(05:35):
the last 30 years 30. Years.
That's a really comprehensive long term view.
What made their data so special for this?
Oh, the data set is genuinely remarkable.
It was a nationwide study using Swedish registers.
They linked data from, get this,almost 3,000,000 children born
between 1982 and 2010. It's actually the world's
largest nationwide multi generation register and this
(05:58):
allowed them to link full siblings.
Ah, siblings. OK.
Yeah, which gives you this incredible ability to control
for shared family factors, both genetic and environmental.
It's a goldmine for this kind ofresearch, truly.
Wow, OK, so how did they actually track the T1D diagnosis
in such a huge group? How do they make sure they
caught everything? They're really meticulous.
Kids were followed from birth until age 19 or until they
(06:19):
emigrated or passed away. With data going up 2020.
They use multiple sources to identify cases, the National
Patient Register, the National Diabetes Register, even the
first prescription for glucose lowering drugs.
Very robust and they categorize T1D by age at diagnosis
206-712-1318 years to see if patterns differed by age.
(06:39):
Okay. And calculating that crucial
herability number across all those people over time, what was
their method? They use something called an AE
model that stands for additive genetic factors plus non shared
Environmental factors. It's powerful because it tries
to isolate the direct genetic contributions, separating them
from environmental things that siblings don't necessarily
(07:01):
share. Gives a cleaner look at
heritability. And they base this model on an
incredible number of sibling pairs, over one and a half
million pairs of full siblings. Yeah.
And they specifically looked forwhat they called moderation
effects of birth year. Basically, did heritability
change depending on when you were born?
That's key to their main question.
Makes sense? And beyond the genetics, what
(07:23):
environmental factors did they pull from those registers?
They looked at a whole range of factors already linked or
suspected to be linked to T1D. Things like maternal smoking
during pregnancy, maternal BMI, any infections the mother had
during pregnancy. Also gestational age, birth
weight, C-section versus vaginalbirth, even parental education
level. They cast a pretty wide net.
(07:44):
That sounds incredibly thorough.You mentioned the sibling
comparison design. That seems really important for
teasing things apart. Absolutely.
The sibling comparison is reallypowerful here because it helps
control for all those unmeasuredthings, genetic factors,
environmental factors that siblings share because they grow
up in the same family. It gives you a much clearer
picture of the direct link between an individual exposure
(08:06):
like maybe maternal smoking and T1D risk separate from all that
family background stuff. They also use standard Cox
models for tracking incidence trends over time and these
things called causal mediation models.
OK. What do those?
Do they help quantify how much of the increase in T1D could
actually be explained by changesin how common those specific
(08:27):
environmental factors became? And they even did a simulation
study. They asked, OK, what should
heritability look like if this entire rise in T1D was only due
to environmental changes? That helped put their actual
findings in context. Right, the what if scenario.
OK, so they laid all this groundwork, used this incredible
data set. What did they actually find?
Let's start with the incidence trend itself.
(08:47):
Did it match what we thought we knew?
Pretty much, yeah. The incidence of T1D nearly
doubled in Sweden for kids born between 1982 and 2000.
So for those born in 82, the cumulative incidents by age 19
was about half a percent, .50%. But for kids born in 2000, it
jumped up to .93%. That's a huge increase in less
(09:08):
than 20 years. It really is.
Did that steep climb just keep going or did something change?
That's the interesting part. It didn't keep climbing at that
rate. For children born between 2001
and 2010, the incidents actuallyplateaued leveled off.
This increasing trend, followed by the plateau, was seen across
all the age groups they looked at, though maybe a bit less
pronounced in the oldest teenagers.
(09:29):
And that pattern matches other studies from Sweden and places
like Finland too. OK, so the incidents doubled,
then flattened. Now the big one, the
heritability results. This is the core finding, right?
The potential game changer? Exactly.
This is where it gets really interesting.
First off, the overall heritability of T1D.
Across the whole study period, they estimated it to be very
high. Point 83.
(09:50):
That's 83%. But here's the kicker, the most
striking, maybe counterintuitivefinding that heritability
remained remarkably stable, thatHubbard right around .8 across
the entire observation period. Yeah.
So for example, was .80 back in 1982.83 in 2000 and still .83 in
2010. Hold on, let me make sure I've
(10:12):
got this. The number of cases nearly
doubled, right? But the proportion of risk
explained by genetics, the heritability stayed basically
the same, around 80%. That's exactly what they found.
That completely goes against theidea that changing environmental
factors were the main thing driving the increase, because if
they were, heritage should have dropped, right?
Precisely. It directly contradicts that
expectation. If new environmental factors
(10:33):
were piling on risk, the relative importance of genetics
should have gone down. But it didn't.
They also found heritability wasa bit higher in the youngest
kids. 06 years old is .85 there.But crucially, even within those
age groups it stayed stable overtime.
And remember that simulation? It showed that if the rise in
case is about 40 extra cases per10,000 kids between 82 and 2000,
(10:55):
if that was all due to environment, heritability should
have plummeted to around .59, maybe 59%.
Wow. But what they observed was
stable around .83. That stable high number really
challenges the environment. Only explanation.
OK, so if heritability stayed stable, what about those
specific environmental factors they tracked?
Did they find any link? Did their changes explain any of
(11:17):
the rise? They did find some associations,
yes. In the full cohort analysis,
maternal smoking during pregnancy seem protective.
Higher maternal age of delivery was linked to higher risk.
Also maternal bacterial infections during pregnancy,
gestational age, mode of delivery, extremely low birth
weight. Those showed associations too.
However, that sibling analysis, the really powerful one.
(11:39):
Right, controlling for family factors.
Exactly. It suggested that the link they
saw with maternal BMI during pregnancy might actually be due
to what we call familial confounding meaning, meaning it
might not be the BMI by itself, but other factors, genetic or
environmental, that are shared within the family that are
really driving that particular association.
(12:00):
OK, that's a critical difference.
So bottom line, when they lookedat the changing prevalence of
these known factors over time, like fewer people smoking,
mothers being older, how much ofthat near doubling of T1D
incidents could they actually account for?
And this is the crucial point. Very little.
When they quantified it, the changing rates of maternal
smoking and maternal age explained only a tiny fraction
(12:21):
of the increase, specifically about 3.2% from smoking changes
and only .8% from changes in maternal age.
Childhood overweight and obesitycontributed a bit more.
Still small numbers, maybe two 4% in boys, one 2% in girls,
depending on the estimate. Adding it all up.
Adding it all up, these known environmental factors, even the
(12:42):
ones whose prevalence changed significantly over those
decades, explained less than 10%of the massive rise in T1D they
observed. Less than 10%.
Wow. Yeah.
It really points towards something else driving the bulk
of that increase. So if you connect this to the
bigger picture, these findings are really quite profound.
That consistently high in stableheritability happening at the
(13:03):
same time as incidents doubled. It directly challenges that
widely held idea that the known environmental factors are the
main drivers. It strongly suggests that the
proportions of T1D risk explained by genetics versus
environment has stayed pretty much the same in Sweden over the
last 30 years. That's a big shift in thinking.
It really is, Yeah. So if the genetic contributions
(13:25):
stayed stable proportion wise and the usual environmental
suspects explain less than 10% of the rise, we have this
paradox. Incidents went way up, but our
standard explanations don't cut it.
The paper proposes this really interesting idea, this decreased
combined load hypothesis. What exactly does that mean?
How does it try to square this circle?
It's a really intriguing hypothesis.
Essentially, they suggest that the total combined load, the sum
(13:48):
of genetic risk and environmental risk needed for a
person to actually develop T1D, might have decreased over time.
So like the threshold for getting the disease got lower?
Exactly. The bar has been lowered, so to
speak, even though the overall genetic free disposition in the
population hasn't fundamentally changed in its relative
importance. OK, so how could that happen?
(14:09):
Is the idea that something in our modern environment is making
it easier for people with maybe just moderate genetic risk to
tip over over into developing T1D?
That's precisely the implication.
The hypothesis is that the penetrance of moderate risk
genes, particularly certain HLA genotypes, might have increased
over time. Pentrance being the likelihood
(14:30):
the gene actually causes the disease if you have.
It correct. So this increased penetrance is
thought to be driven by an increasingly diabetogenic
environment, whatever that involves.
This means kids who have maybe only a moderate genetic risk,
kids who perhaps wouldn't have developed T1D in the environment
of, say, the 1970s, are now developing it because today's
environment somehow makes it easier for that modern genetic
(14:52):
risk to manifest as disease. It points strongly towards gene
environment interactions being key.
That makes a certain kind of sense, but it definitely feels
like there is still big questionmarks.
What were the main limitations the researchers acknowledged?
Yeah, definitely no studies perfect.
A key limitation was the lack ofreally detailed individual level
data for some things, like childhood overweight.
(15:15):
Obesity measures weren't available for everyone
individually, and they didn't have specific genetic data like
precise HLA genotypes for the whole massive cohort.
That limited their ability to directly test those specific
gene environment interaction hypotheses.
They also couldn't distinguish certain infections like
enteroviruses, which are strongly suspected in T1D, from
(15:36):
just general infection categories in the registers.
And of course, this is all done in Sweden.
So while it's a huge and powerful study, we have to be
cautious about generalizing these exact findings, the stable
herability, the specific factor contributions to other
populations with different genetic backgrounds and
environments. OK.
So given those limitations and this really provocative finding,
(15:57):
what's the biggest unanswered question here?
What's the most important next step for research?
Well, the clearest message is that there must be other
currently unidentified environmental factors playing a
major role, especially factors that are not shared by siblings
because the analysis showed the shared environment explained
very little. So we need to look beyond the
usual suspects. The most urgent next step I
(16:19):
think is funding and conducting large scale studies that collect
really comprehensive data. We've the individual level
genetics, detailed environmentalexposures may be looking at
viruses, microbiome, diet, chemicals measured prospectively
following kids over time. Only then can we really hope to
pinpoint these missing environmental treasures and
truly understand how they're interacting with genetic risk to
(16:40):
drive this disease. We need to map out that
diabetogenic environment. This really does shift the
perspective, doesn't it? It suggests the focus shouldn't
just be on the known environmental factors, but
really on uncovering the unknowns and understanding how
genes in this changing environment interact in ways
where maybe just starting to grasp.
It feels like a new chapter for two 1D research, so a wrap up
(17:01):
our deep dive today. The key take away is that the
heritability of childhood onset type 1 diabetes in Sweden stayed
remarkably high and stable over 3 decades, even while the number
of cases nearly doubled. This strongly implies that the
environmental factors we've traditionally focused on, while
they might play some role, are not the major drivers behind
that dramatic increase. Instead, the evidence points
(17:23):
towards a complex interplay, a stable underlying genetic
predisposition interacting with still to be identified
environmental triggers, and these interactions seem to be
lowering the overall threshold needed to develop the disease.
It leaves us with a big question.
What does this mean for how we approach prevention, how we
think about genetic counseling, and crucially, where we hunt for
those elusive environmental culprits in the future of type 1
(17:45):
diabetes? This episode was based on an
Open Access article under the CCBY 4 Point O license.
You can find a direct link to the paper and the license in our
episode description. If you enjoyed this analysis,
the best way to support Base by Base is to subscribe or follow
in your favorite podcast app andleave us a five star rating.
It only takes a few seconds but makes a huge difference in
(18:06):
helping others discover the show.
Thanks for listening and join usnext time as we explore more
science base by Base.
Welcome to Base by Base, the AERcast that brings genomics to
you, wherever you are. Today, we're embarking on a deep
dive into, well, a really groundbreaking scientific study.
It challenges some, some pretty deeply held assumptions about a
common childhood illness. OK, picture this.
A common childhood condition, right?
One that's been steadily increasing, going up and up for
(00:35):
decades now. What do you think is driving
that rise? You know, is it our environment,
changes in our genes, or maybe something else entirely?
We're talking about childhood onset type 1 diabetes.
Its incidence has been climbing worldwide since like the 1950s,
increasing by maybe 3-4 percent every year.
That's not a small shift. It's a really significant global
(00:56):
trend, and it's frankly perplexed researchers.
For years now. Conventional wisdom often points
the finger at, you know, our changing environment, our diets,
maybe early life exposures, things like that.
But what if the story is much more complex, maybe even
counterintuitive? What if the thing we thought was
driving this trend? What if it's actually been
remarkably stable all along? Today, we're going to unpack a
study that sheds new light on this exotic question.
(01:18):
It challenges some long held beliefs and honestly forces us
to reconsider where we should belooking.
But before we get into the the fascinating details, we really
want to recognize the incredibleminds behind this work.
Today we celebrate the work of Yuksi Away, Thomas Anderson,
Sheng Sinlu, Maria Fakting, Ralph Kuja Halkola and Sophia
Carlson from the Karolinska Institute in Sweden.
(01:40):
They've really advanced our understanding of the well, the
changing influences on childhoodonset type 1 diabetes.
OK, let's unpack the central scientific debate here.
We know type one that diabetes incidence is rising globally,
but what's been the main hypothesis explained well, the.
Core problem has been, it's beenpretty unclear, is this global
increase in childhood T1D because of rising environmental
(02:02):
risk factors, non genetic ones? Or is it maybe increasing
genetic susceptibility in the population?
Or, you know, perhaps some complex mix of both?
Right. And when you see a disease
increase that rapidly, it feels natural to look at environmental
changes first. But why was that the dominant
idea for SO? Long that's true.
Environmental explanations have generally been favored.
(02:23):
And the reasoning? It seems pretty sound on the
surface. It just seems highly unlikely
that the genetic makeup of a whole population would change so
drastically, so quickly. You know, in just a few decades
plus, you see these big differences in T1D incidences
even between populations that are genetically quite similar.
Think about Finland having a much higher rate than, say,
(02:44):
Estonia, even though they're geographically and ethnically
pretty close. That points to external factors.
Right. Sure, that makes sense.
So what specific environmental factors were kind of on the
radar, potential culprits or maybe even things that protect
against it? Yeah, several stood out.
For instance, the rise in T1D happened alongside the rise in
childhood obesity that's often proposed as a risk factor,
(03:05):
though the causal link is still debated.
Other early life things often discussed include maternal
smoking, which interestingly hassometimes been suggested as
potentially protective, believe it or not.
Also higher maternal agent delivery that's seen as a risk,
and breastfeeding, generally thought to be protective.
The big question has always been, you know, how much have
changes in these factors actually contributed to the
(03:26):
overall rise in T1D? And key to the study is the idea
of heritability. Could you break that down for
us? What does it mean here and why
is it so crucial? Right, heritability.
In simple terms, it's the proportion of the variation we
see in a disease like T1D. How much of that variation can
be explained by all the genetic factors combined.
It's not about whether you specifically will get it based
(03:47):
on genes, but about how much of the difference between people
who get it and people who don't within a population is down to
genetic differences. For T1D, studies have estimated
heritability to be pretty high, actually, somewhere between say
50% and 88% in different groups.A recent Swedish study put it at
81% in kits. Now this is crucial because,
(04:08):
well, if environmental factors are really the main drivers
pushing incidents up, then you'dexpect their relative
contribution to increase over time, and logically heritability
should decrease. So the key question this study
asked was, has T1D heritability actually changed over the last
few decades? Because as far as they knew,
nobody had really looked at thiscomprehensively before.
That's a massive question to tackle.
And Speaking of genetics, I remember reading about HLA
(04:31):
genotypes. HLA human leukocyte antigen
genes right key for the immune system.
Certain versions are known risksfor autoimmune diseases like
T1D. Did those earlier HLA studies
give any clues? They did, yeah.
Yeah, Some intriguing hints. A study in Australia and similar
findings came out of Finland, Sweden, the USUK.
They found something kind of counterintuitive.
(04:52):
The proportion of T1D cases withthe highest risk HLA types
actually decreased from the 80s to the 2000s.
Meanwhile, the proportion with intermediate risk HLA types went
up. So it suggests the HLA
contribution shifted, but didn'tnecessarily lessen the overall
genetic risk. It just changed which genetic
profiles were most affected. Maybe.
And it's important to remember, HLA genes only account for about
(05:14):
half the total genetic risk for T1D.
There's still a lot more genetics involved.
So moving to their methods, whatreally makes the study stand out
is how they approach this. They built their analysis on an
incredibly solid foundation. Their goal was clear, figure out
if environmental and genetic influences on childhood T1D,
specifically kids aged zero to 18, have changed in Sweden over
(05:35):
the last 30 years 30. Years.
That's a really comprehensive long term view.
What made their data so special for this?
Oh, the data set is genuinely remarkable.
It was a nationwide study using Swedish registers.
They linked data from, get this,almost 3,000,000 children born
between 1982 and 2010. It's actually the world's
largest nationwide multi generation register and this
(05:58):
allowed them to link full siblings.
Ah, siblings. OK.
Yeah, which gives you this incredible ability to control
for shared family factors, both genetic and environmental.
It's a goldmine for this kind ofresearch, truly.
Wow, OK, so how did they actually track the T1D diagnosis
in such a huge group? How do they make sure they
caught everything? They're really meticulous.
Kids were followed from birth until age 19 or until they
(06:19):
emigrated or passed away. With data going up 2020.
They use multiple sources to identify cases, the National
Patient Register, the National Diabetes Register, even the
first prescription for glucose lowering drugs.
Very robust and they categorize T1D by age at diagnosis
206-712-1318 years to see if patterns differed by age.
(06:39):
Okay. And calculating that crucial
herability number across all those people over time, what was
their method? They use something called an AE
model that stands for additive genetic factors plus non shared
Environmental factors. It's powerful because it tries
to isolate the direct genetic contributions, separating them
from environmental things that siblings don't necessarily
(07:01):
share. Gives a cleaner look at
heritability. And they base this model on an
incredible number of sibling pairs, over one and a half
million pairs of full siblings. Yeah.
And they specifically looked forwhat they called moderation
effects of birth year. Basically, did heritability
change depending on when you were born?
That's key to their main question.
Makes sense? And beyond the genetics, what
(07:23):
environmental factors did they pull from those registers?
They looked at a whole range of factors already linked or
suspected to be linked to T1D. Things like maternal smoking
during pregnancy, maternal BMI, any infections the mother had
during pregnancy. Also gestational age, birth
weight, C-section versus vaginalbirth, even parental education
level. They cast a pretty wide net.
(07:44):
That sounds incredibly thorough.You mentioned the sibling
comparison design. That seems really important for
teasing things apart. Absolutely.
The sibling comparison is reallypowerful here because it helps
control for all those unmeasuredthings, genetic factors,
environmental factors that siblings share because they grow
up in the same family. It gives you a much clearer
picture of the direct link between an individual exposure
(08:06):
like maybe maternal smoking and T1D risk separate from all that
family background stuff. They also use standard Cox
models for tracking incidence trends over time and these
things called causal mediation models.
OK. What do those?
Do they help quantify how much of the increase in T1D could
actually be explained by changesin how common those specific
(08:27):
environmental factors became? And they even did a simulation
study. They asked, OK, what should
heritability look like if this entire rise in T1D was only due
to environmental changes? That helped put their actual
findings in context. Right, the what if scenario.
OK, so they laid all this groundwork, used this incredible
data set. What did they actually find?
Let's start with the incidence trend itself.
(08:47):
Did it match what we thought we knew?
Pretty much, yeah. The incidence of T1D nearly
doubled in Sweden for kids born between 1982 and 2000.
So for those born in 82, the cumulative incidents by age 19
was about half a percent, .50%. But for kids born in 2000, it
jumped up to .93%. That's a huge increase in less
(09:08):
than 20 years. It really is.
Did that steep climb just keep going or did something change?
That's the interesting part. It didn't keep climbing at that
rate. For children born between 2001
and 2010, the incidents actuallyplateaued leveled off.
This increasing trend, followed by the plateau, was seen across
all the age groups they looked at, though maybe a bit less
pronounced in the oldest teenagers.
(09:29):
And that pattern matches other studies from Sweden and places
like Finland too. OK, so the incidents doubled,
then flattened. Now the big one, the
heritability results. This is the core finding, right?
The potential game changer? Exactly.
This is where it gets really interesting.
First off, the overall heritability of T1D.
Across the whole study period, they estimated it to be very
high. Point 83.
(09:50):
That's 83%. But here's the kicker, the most
striking, maybe counterintuitivefinding that heritability
remained remarkably stable, thatHubbard right around .8 across
the entire observation period. Yeah.
So for example, was .80 back in 1982.83 in 2000 and still .83 in
2010. Hold on, let me make sure I've
(10:12):
got this. The number of cases nearly
doubled, right? But the proportion of risk
explained by genetics, the heritability stayed basically
the same, around 80%. That's exactly what they found.
That completely goes against theidea that changing environmental
factors were the main thing driving the increase, because if
they were, heritage should have dropped, right?
Precisely. It directly contradicts that
expectation. If new environmental factors
(10:33):
were piling on risk, the relative importance of genetics
should have gone down. But it didn't.
They also found heritability wasa bit higher in the youngest
kids. 06 years old is .85 there.But crucially, even within those
age groups it stayed stable overtime.
And remember that simulation? It showed that if the rise in
case is about 40 extra cases per10,000 kids between 82 and 2000,
(10:55):
if that was all due to environment, heritability should
have plummeted to around .59, maybe 59%.
Wow. But what they observed was
stable around .83. That stable high number really
challenges the environment. Only explanation.
OK, so if heritability stayed stable, what about those
specific environmental factors they tracked?
Did they find any link? Did their changes explain any of
(11:17):
the rise? They did find some associations,
yes. In the full cohort analysis,
maternal smoking during pregnancy seem protective.
Higher maternal age of delivery was linked to higher risk.
Also maternal bacterial infections during pregnancy,
gestational age, mode of delivery, extremely low birth
weight. Those showed associations too.
However, that sibling analysis, the really powerful one.
(11:39):
Right, controlling for family factors.
Exactly. It suggested that the link they
saw with maternal BMI during pregnancy might actually be due
to what we call familial confounding meaning, meaning it
might not be the BMI by itself, but other factors, genetic or
environmental, that are shared within the family that are
really driving that particular association.
(12:00):
OK, that's a critical difference.
So bottom line, when they lookedat the changing prevalence of
these known factors over time, like fewer people smoking,
mothers being older, how much ofthat near doubling of T1D
incidents could they actually account for?
And this is the crucial point. Very little.
When they quantified it, the changing rates of maternal
smoking and maternal age explained only a tiny fraction
(12:21):
of the increase, specifically about 3.2% from smoking changes
and only .8% from changes in maternal age.
Childhood overweight and obesitycontributed a bit more.
Still small numbers, maybe two 4% in boys, one 2% in girls,
depending on the estimate. Adding it all up.
Adding it all up, these known environmental factors, even the
(12:42):
ones whose prevalence changed significantly over those
decades, explained less than 10%of the massive rise in T1D they
observed. Less than 10%.
Wow. Yeah.
It really points towards something else driving the bulk
of that increase. So if you connect this to the
bigger picture, these findings are really quite profound.
That consistently high in stableheritability happening at the
(13:03):
same time as incidents doubled. It directly challenges that
widely held idea that the known environmental factors are the
main drivers. It strongly suggests that the
proportions of T1D risk explained by genetics versus
environment has stayed pretty much the same in Sweden over the
last 30 years. That's a big shift in thinking.
It really is, Yeah. So if the genetic contributions
(13:25):
stayed stable proportion wise and the usual environmental
suspects explain less than 10% of the rise, we have this
paradox. Incidents went way up, but our
standard explanations don't cut it.
The paper proposes this really interesting idea, this decreased
combined load hypothesis. What exactly does that mean?
How does it try to square this circle?
It's a really intriguing hypothesis.
Essentially, they suggest that the total combined load, the sum
(13:48):
of genetic risk and environmental risk needed for a
person to actually develop T1D, might have decreased over time.
So like the threshold for getting the disease got lower?
Exactly. The bar has been lowered, so to
speak, even though the overall genetic free disposition in the
population hasn't fundamentally changed in its relative
importance. OK, so how could that happen?
(14:09):
Is the idea that something in our modern environment is making
it easier for people with maybe just moderate genetic risk to
tip over over into developing T1D?
That's precisely the implication.
The hypothesis is that the penetrance of moderate risk
genes, particularly certain HLA genotypes, might have increased
over time. Pentrance being the likelihood
(14:30):
the gene actually causes the disease if you have.
It correct. So this increased penetrance is
thought to be driven by an increasingly diabetogenic
environment, whatever that involves.
This means kids who have maybe only a moderate genetic risk,
kids who perhaps wouldn't have developed T1D in the environment
of, say, the 1970s, are now developing it because today's
environment somehow makes it easier for that modern genetic
(14:52):
risk to manifest as disease. It points strongly towards gene
environment interactions being key.
That makes a certain kind of sense, but it definitely feels
like there is still big questionmarks.
What were the main limitations the researchers acknowledged?
Yeah, definitely no studies perfect.
A key limitation was the lack ofreally detailed individual level
data for some things, like childhood overweight.
(15:15):
Obesity measures weren't available for everyone
individually, and they didn't have specific genetic data like
precise HLA genotypes for the whole massive cohort.
That limited their ability to directly test those specific
gene environment interaction hypotheses.
They also couldn't distinguish certain infections like
enteroviruses, which are strongly suspected in T1D, from
(15:36):
just general infection categories in the registers.
And of course, this is all done in Sweden.
So while it's a huge and powerful study, we have to be
cautious about generalizing these exact findings, the stable
herability, the specific factor contributions to other
populations with different genetic backgrounds and
environments. OK.
So given those limitations and this really provocative finding,
(15:57):
what's the biggest unanswered question here?
What's the most important next step for research?
Well, the clearest message is that there must be other
currently unidentified environmental factors playing a
major role, especially factors that are not shared by siblings
because the analysis showed the shared environment explained
very little. So we need to look beyond the
usual suspects. The most urgent next step I
(16:19):
think is funding and conducting large scale studies that collect
really comprehensive data. We've the individual level
genetics, detailed environmentalexposures may be looking at
viruses, microbiome, diet, chemicals measured prospectively
following kids over time. Only then can we really hope to
pinpoint these missing environmental treasures and
truly understand how they're interacting with genetic risk to
(16:40):
drive this disease. We need to map out that
diabetogenic environment. This really does shift the
perspective, doesn't it? It suggests the focus shouldn't
just be on the known environmental factors, but
really on uncovering the unknowns and understanding how
genes in this changing environment interact in ways
where maybe just starting to grasp.
It feels like a new chapter for two 1D research, so a wrap up
(17:01):
our deep dive today. The key take away is that the
heritability of childhood onset type 1 diabetes in Sweden stayed
remarkably high and stable over 3 decades, even while the number
of cases nearly doubled. This strongly implies that the
environmental factors we've traditionally focused on, while
they might play some role, are not the major drivers behind
that dramatic increase. Instead, the evidence points
(17:23):
towards a complex interplay, a stable underlying genetic
predisposition interacting with still to be identified
environmental triggers, and these interactions seem to be
lowering the overall threshold needed to develop the disease.
It leaves us with a big question.
What does this mean for how we approach prevention, how we
think about genetic counseling, and crucially, where we hunt for
those elusive environmental culprits in the future of type 1
(17:45):
diabetes? This episode was based on an
Open Access article under the CCBY 4 Point O license.
You can find a direct link to the paper and the license in our
episode description. If you enjoyed this analysis,
the best way to support Base by Base is to subscribe or follow
in your favorite podcast app andleave us a five star rating.
It only takes a few seconds but makes a huge difference in
(18:06):
helping others discover the show.
Thanks for listening and join usnext time as we explore more
science base by Base.
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.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Special Summer Offer: Exclusively on Apple Podcasts, try our Dateline Premium subscription completely free for one month! With Dateline Premium, you get every episode ad-free plus exclusive bonus content.