Future Directions in the Study of Early-Life Stress and Physical and Emotional Health

all right good afternoon everybody so
I’m so very excited to introduce our next speaker Greg Miller my students in
the lab have probably heard me say 30 or 40 or 50 times that if I had to do grad
school over again and and working in an area of research that really excited me
now I really really really want to learn about the topics that Greg will cover
today I think it’s fascinating the intersections between immunological and
regulatory functioning and mental help it the the work is so elegance so
exquisite and so I already had a really deep appreciation for for Greg Miller’s
work and and as a consequence of all of this of course being the research dad
that I am I always try to live vicariously through my kids and I always
tell them look my dream is to place one of you and Greg Miller’s lab he can tell
me what it’s all about I got the next best thing my close friend Katie got
postdoctoral training working with Greg so I had some idea about about what it
was like and I already really admired Greg’s work and then I got us chance to
see what kind of mentor he is yeah everyone I give you Greg Miller thanks Andy it’s hard to go forward
after an introduction like that it was nice this introduction I’ve ever gotten
and I’m not kidding thank you somebody’s lap top before I get started so I’m
thanks for having me here today thanks for all of you for coming out on this
Saturday I’m happy to make this interactive so if you have questions or
comments don’t hesitate to jump in we can talk as much as you want probably
more interesting very video gets back and forth so I’ll just go ahead and get
started I’m gonna I think shift gears a little bit from the topics you focused
on for most of today and discuss work we’ve been doing over the years on early
life stress particularly around socioeconomic status and how it relates
to physical health across the lifespan so you probably all know that in the
United States there are dramatic differences in health and longevity that
patterns strongly according to race and class this is a paper that Paula
braveman published a few years ago summarizing some CDC data and what you
can see is that if you operationally define socioeconomic status according to
a percentage of the federal poverty line in terms of household income you see
these these linear differences in life expectancy
according to SES and you see it across racial and ethnic groups so within
African Americans within white Americans and within Latino Americans they’re all
strong income gradients and life expectancy and I can show you this for a
number of different specific causes of mortality various cancers heart disease
stroke or for morbidity the patterns are generally the same there are strong
income based differences strong education based differences and strong
race and ethnicity based differences and these for this audience I think it’s
particularly important to emphasize that these disparities start very early in
life so here you’re looking data on maternal education and neonatal
mortality and infant mortality and you can see those linear gradients I showed
you in life expectancy post age 25 are actually present from the very earliest
days of life and these disparities not only did they start in early childhood
but they continued through childhood and adolescence so on my right in red you
see differences in obesity defined as being at the 95th percentile sex and age
adjusted for children and on the left you see the amount of carotid
atherosclerosis so you can take an ultrasound of the carotid artery and
measure the amount of plaque that’s present there or the thickness of the
vessel wall and it’s a good indicator of how much atherosclerosis is building up
long before you’re gonna have a heart attack or a stroke it’s the accumulation
of early plaque and you can see at ages 11 to 12 there’s this nice study of a
few thousand people showing that according to family socioeconomics that
it’s a composite of different educational income measures there’s a
gradient in plaque again it’s the same linear effect I’ve been showing you so
already by early adolescence you’re seeing differences in plaque presence
according to SES look at BMI I could show you slides on diabetes or upper
respiratory infection and they would look fairly similar and this stretches
across the life course so now we’re not going to ask about what socioeconomic
conditions relate to in terms of children’s health but we’re gonna ask
about what those socio-economic conditions in childhood relate to in
terms of people’s long-term health so this is a fascinating study published a
little over a decade ago now that focuses on a cohort of men who graduated
from Johns Hopkins Medical School in the 50s and 60s that was a time where
Hopkins only took male medical students and they were followed and they’re still
being followed as I understand it and the end point here is the presence or
absence of coronary heart disease when the men hit their mid-70s
and what it stratified by here is a very crude metric of childhood SES whether
the physicians father had a manual versus non manual occupation when the
physician was a kid and as you can see there’s a difference in the rate of
coronary heart disease that favors those who came from childhood socioeconomic
conditions that were more advantaged a dad with a nonmetal occupation and these
effects are present interestingly in a group that is high socioeconomic status
themselves by any definition right remember all of the subjects here are
Hopkins Medical School graduates if you look at occupational prestige if you
look at education if you look at income if you look at wealth by any metric
these are high SES men so what these findings suggest and there are a number
of others that that reach similar conclusions was that even with a pretty
sizable increase in economic mobility from your childhood to your adulthood
these gaps remain or some of these gaps are made an early experience seems to
have some lasting imprint that sticks around and affects your health later on
in life and isn’t completely erased by upward socio-economic mobility or better
circumstances that’s not to say early experiences destiny and there’s nothing
that can be done I don’t want to leave you with that conclusion it’s not true
there’s definitely things we can do to mitigate the impact of negative early
experiences and those things come in the form of different policies and different
interventions but there still does seem to be even in the most favorable upward
mobility conditions some evidence of early experienced wing during and then
there’s next generational evidence that’s starting to emerge the the
strongest evidence for this in terms of cause and effect definitely comes from
animal models but there’s even some suggestive evidence in humans where it’s
harder to do exactly the right study but the findings are consistent with the
animal data which suggests causality these are some data on kids with asthma
that we’ve been studying over the past five years Kati was very involved with
this work so this is 300 or so kids with asthma between the ages of about 8 and
17 that we’ve been studying in Chicago they all are receive
medical care in the same couple of clinics for asthma and we wanted to look
at whether the socioeconomic status of the children’s parents as children so we
wanted to look at these cross generational affects was related to how
the kids asthma was doing and what we see is that if you’re looking at a
teenager today at the socioeconomic status of his or her mother can tell you
something about how the kid’s asthma is doing independent of the family’s
current socioeconomic status and so what I’m showing you here is the parent and
child’s report of the extent of asthma control on my right and on my left I’m
showing you some immunologic measures where we take blood from the child with
asthma and in a test tube we expose it to things that cause the immune cells to
get activated and produced molecules that eventually are what causes their
lungs to fill with mucus and become obstructed and for them to experience
the clinical symptoms of asthma so on both a more clinical measure and on this
biological measure we see the same pattern which is independent of the
family’s current SES what happened to the kids parents in their childhood
tells you something about disease and we’ve got other data as do other groups
looking at other childhood health and perinatal outcomes which suggests the
same conclusions so I guess what I would say here is that there’s quite solid
evidence at the deme illogically substantiated by stronger studies and
animal models to suggest that early life conditions particularly around
deprivation and also as Kate talked about around threat relate to health
across the life course that’s not to say that every disease we care about
patterns perfectly according to socioeconomic status or is it the only
thing that matters is socioeconomic status there are other things that
clearly matter for health some of which are completely independent of
socioeconomic status like the genetics people get in the lottery but there are
a lot of things that do pattern by socioeconomic status like lifestyle in
some cases and access to resources and structural and environmental factors
that do combine to create these disparities in physical
health outcomes as I said the strongest effects are probably in the cardio
metabolic domain if you look at things like diabetes heart disease stroke and
also in lung disease childhood lung diseases like asthma and a dolling
diseases like chronic obstructive pulmonary disease the effects are not as
strong or consistent in things like cancer for example so when people see
data like this when I give presentations or write papers there’s a common
response which is okay these are interesting associations but it’s really
not that complicated of a phenomena it’s all about and then there are three
common it’s all about sight get so the first thing I get this is all just a
genetic story and you know the version of this goes something like well there’s
this basket of genes that predispose people to be unhealthy and kind of
stupid and make not a lot of money and those genes travel and families and what
you’re seeing here is basically just a genetic effect that you know there’s a
group of people who just aren’t so smart or aren’t so healthy and they have a lot
of kids in it Pat you’ve heard this narrative before I’m guessing and I
think that you know I’m being a little bit flip about this this explanation but
more often than you’d believe this is what you get particularly when you give
talks like this to medical audiences but if we took this explanation seriously
which I think we have to you know what would you say about the evidence and I
would say well you know how do you address it well first you can look at
studies that have been done in administrative fashion in Northern
Europe where they’ve been able to max up match up tax records with health records
and with those sorts of data you can ask questions like if a baby is adopted from
a low-income family into a high-income family how does their long term health
look are there more or less likely to have a heart attack by the time they’re
65 then their sibling who wasn’t adopted and the answer that that sort of
question when you look at administrative data is that both the genetics of the
pair and or both the socioeconomic status of
the biological parents and of the adoptive parents matters right they have
independent effects of about equal magnitude so that’s not to say there’s
no genetic effect but it’s also to say there’s a queer environmental signal
there above and beyond the genetic effect in the adoption studies now you
might argue and I think there’s room for this argument that those adoption
studies aren’t perfect they’re a good natural assort of experiment but without
random assignment you can’t be sure so if you’re gonna require random
assignment as evidence then I think you look to the animal models and I think
there’s no very persuasive evidence in mice in rats and in monkeys that if you
create early life conditions that resemble as best they can the kind of
deprivation and the kind of threat that often accompanies being a low-income kid
in urban America today that there are causal effects of those early
experiences on conditions particularly as I said in the cardio metabolic domain
how people regularly glucose their risk for diabetes heart disease long-term
health outcomes around that so this is it just about bad genes genes may have a
role and when I say genes in this context I’m talking about genes in terms
of their function of transmitting hereditary influences but the
environment still has a role above and beyond that some people think this is
all about health care right so that’s another common reaction to these
findings particularly in medical circles which is we’ve got a system in the
United States where there’s unequal access to care or even if there’s not
unequal access there’s unequal access to high-quality care um certainly
differences in care access and care utilization are part of the problem but
most of the disparities and physical health by socioeconomic status are
things that start before people even get to the doctor right and there are things
that we don’t have effective preventive treatments for so it can’t just be an
access to care thing right there are differences in health by socioeconomic
status in how people’s disease progresses once they get diagnosed those
are there but the biggest differences are in incidents in the first place even
if we had a universal access healthcare system like you might see in Canada or
in Western European countries you still wouldn’t get rid of these disparities
and that’s shown because you’ve sees disparities like this in countries with
universal access to health care so it’s not just that and then there’s the
lifestyle argument that okay we get it it’s not just genetics there’s an
environmental effect but the environmental effect is really about
low-income families behaving badly that’s sort of the argument they’re more
likely to smoke they you know potato chips for dinner you hear these sorts of
stereotypes and narratives again I would say yeah there are differences in health
behaviors that are part of the story here but you know we shouldn’t view
lifestyle and behavior in a vacuum a lot of the differences in lifestyle can be
fairly easily explained by differences in policy by differences in structure by
differences environment that really do constrain what appear to be people’s
choices and backed them into particular lifestyles that might relate to worse
disease outcomes and that even when we carefully measure physical activity diet
and cigarette smoking they explain a bit of that socio-economic gradient but they
certainly don’t explain it all so there’s more to it than that
so if it’s not just heredity and if it’s not access to care and if it’s not
lifestyle what is it how is it that early life socioeconomic conditions and
the exposures that accompany them seem to relate to physical health across the
life course in the domains that I’ve been discussing and this is a question
that our research group has been focusing on for some time and what we
try to do is use observational studies and when we can experiments and animal
models to try and understand the biological mechanisms involved in the
behavioral mechanisms to try to identify protective and vulnerability factors
that can either mitigate or amplify the associations
the relevant exposures in the relevant health outcomes and then use what we
learn from those studies to make an informed and better practice and I’m
gonna tell you a little bit about each of these lines of work as we go through
today so let’s start with the mechanistic work and let me begin by
saying that our focus has for a long time in terms of mechanisms involve
thinking about the immune system as a key interface between the social
environment and physical health and we’ve been particularly interested in
the parts of the immune system that are involved in generating and maintaining
inflammation and in particular this process called non resolving low-grade
inflammation which means that you’ve got this chronic ongoing low-level
inflammatory response that’s kind of diffusely spread across your body
usually isn’t severe enough that you notice it symptomatically but it’s that
long-term inflammation that has become increasingly known as a contributor to
lots of different chronic health problems that pattern by SES and that
caused a good deal of suffering and disability and you can see from this
slide I’ve got a pointer here I’m gonna turn the do everything off no
hi you can see up here non resulting inflammation is implicated in the
pathophysiology of things ranging from atherosclerosis and obesity and diabetes
standard cardiomyopathy I talked about and also among diseases COPD and asthma
and some autoimmune diseases it seems to be a contributor in a more or less
strong way to most of the chronic diseases that we know pattern by SES and
our thinking has you know probably can be summarized most easily in this very
oversimplified graphic which is that there’s a particular set of cells in the
immune system that underlie a lot of the chronic low-grade inflammation and then
they’re in mature form they’re called monocytes and we’re gonna
study them a lot today and as they get more mature and migrate out of the
bloodstream into tissues in your body they become macrophages or dendritic
cells and our hypothesis has been that kids who grow up in more disadvantaged
environments have monocytes and macrophages dendritic cells that assume
this phenotype that leads to more chronic low-grade inflammation over time
and that phenotype arises because these cells in essence turn out to be more
reactive to encounter if you’re an immune cell ah I get mobile that’s great
thanks Andy so if you’re a if you’re a monocyte out in blood or you’re a
macrophage or dendritic cell out in tissue what are the threats you see you
see bacteria you see viruses you also see your own
body’s cells that have been damaged through some kind of injury or trauma
and are sending out these signals saying hey there’s a mess here come clean it up
and our argument has been that early adversity Prime’s those cells to respond
to those microbial threats and those injury threats in a more aggressive
fashion and we’ll talk a little bit more about what that means
few slides and that simultaneously besides being primed to mount a more
strong response to those threats that the cells of kids growing up in
disadvantaged circumstances will be less able to shut that response down one of
the key ways you shut those inflammatory responses down is by making cortisol
cortisol generally has anti-inflammatory effects and also I’ll show you one of
the things we see is that disadvantaged kids immune cells don’t respond to
cortisol as strongly as they should and that we argue that as a consequence of
being more reactive and less able to recover disadvantaged kids have more of
this chronic ongoing inflammation and that’s what contributes across the life
course in concert with other exposures to some of the excess risk for physical
health problems so you might say okay how do you get from socioeconomic status
to this inflammatory phenotype you’re talking about socioeconomic status I can
already hear bronfenbrenner and everybody in the room who’s red brown
from better saying poverty doesn’t cause bad outcomes it’s not poverty it’s all
the other things downstream of poverty okay I’ve heard that argument before I
think the epidemiology would would speak differently about that but it is clear
the point is well-taken that it’s not just material hardship that relates to a
lot of these adverse health outcomes that there are lots of exposures at the
individual level at the family level at the community level that Co vary with
socioeconomic status particularly the United States and those more proximal
exposures are what affect the developing brain and body that point is well-taken
so we articulated in a psych bull paper about a decade ago what we thought some
of these likely pathways were and none of this will surprise you you know we
argued as have others Gary Evans has done a lot of great work in this area
about the social and physical environments of low-income American kids
being different than the social and physical environments of their more
advantaged peers and those differences play out in terms of the availability of
caregivers the stability of families the level of
and chaos and crowding and households these kids exposure to conflict and
violence in the home in the neighborhood in their schools and differences in diet
and infection and pollution and again this isn’t to say far from it that every
vow income kid in the country has all of these exposures
the thing about SES is that it captures dozens and dozens of exposures that are
higher probability if you’re a low-income kid but there’s no certainty
there’s no one-to-one relationship there are plenty of exceptions so it’s a
difficult and an easy thing to study at the same time because the circumstances
of every low SES child in America are somewhat different and you can’t make
these strong claims but what we’ve argued and I think some of the evidence
over the past decade is just is that those exposures in turn shaped the way
kids stress hormone systems get organized and shape the way their brains
develop and based on that sort of thinking we’ve argued that there are two
kind of overarching pathways that connect socioeconomic status with this
Pro inflammatory phenotype one of them is kind of a classic David Barker esque
programming or conditioning effect where early experience gets inside the cells
of the immune system and kind of is sticky in a real durable way there and
that’s a very classic as I said Barker esque hypothesis that comes out of the
animal in torture and that’s what people talk about as embedding or programming
but we also argue that there’s something secondary going on that perhaps equally
important which you all know a lot about which is that our early experiences
profoundly shaped the way our brains develop and the people will become and
how we engage our circle our social worlds and how we manage stressors and
how we cope with problems in our lives and what sorts of ways we we engage in
health behaviors and our argument has been that those pathways are equally
important because over the life course they continue to affect the immune
system the immune system is not like the way many of us have historic
wood conceives the brain where everything is set by the time you’re 25
or so the immune system has to keep evolving and keep learning so does the
brain obviously but the immune system in particular is always got to be prepared
to deal with new threats that means that there’s always going to be some
plasticity and that your behavior and your emotion and your cognition across
the life course is going to continue to be important for the way your immune
system functions so we’ve made the case that both these programming effects on
what we call accentuating effects are likely to contribute to this
pro-inflammatory phenotype although it’s a good lesson whenever people cite this
paper they talk about these effects like 90% of the papers that cited so you
never know what your readers are gonna do with what you say it’s a good lesson
in writing papers so let’s talk a little bit about what information is at the
molecular level just to get situated before we move into data so this cell
right here at the dendritic cell so this is one of the things that monocytes
transform into when they get into tissue and this dendritic cell has encountered
bacteria you can see gram-negative bacteria up there the thing with the
tail and bacteria of this species have a molecule on their surface called LPS
it’s that red dot there little polysaccharide you’ll hear more about
that in a few slides and what happens is when the dendritic cell or the monocyte
finds that LPS molecule it sets off a signaling cascade and what that
signaling cascade does I’m not going to go through all the molecular details
here what that signaling cascade does is ultimately converge on this molecule
right here called NF kappa-b nuclear factor Kappa B this is what’s
called the transcription factor and it’s the linchpin for the inflammatory
response and what i mean by that is that when this gets activated it moves into
the cell’s nucleus and it binds to the DNA and it switches on the genes that
allow the cell to make the proteins that are involved in inflammation and those
proteins involved in inflammation you may have heard of there are things
like interleukin 6 interleukin 1 tumor necrosis factor-alpha these are just
signaling molecules that immune cells send to each other and communicate
there’s bacteria in this person’s big toe get over to the big toe kill that
bacteria okay this person’s fallen and scraped their
arm there’s a wound in their arm get to the arm and start sewing up that tissue
right these molecules are just instruction signals that immune cells
send to each other and we measure them in lots of studies as a way to kind of
ease drop on what immune cells are saying to one another and so what we’re
gonna do in a lot of studies is we’re gonna take blood cells from kids and
adults and we’re gonna put them in a test tube and we’re gonna add LPS this
bacterial molecule and we’re gonna watch this process happen in a test tube and
see how much of these molecules do those cells make how much annika NF kappa-b is
activated and that’s gonna help us understand these hypotheses about early
socio-economic adversity in the pro-inflammatory phenotype now how does
this process get slowed down I talked a little bit a few slides ago about
cortisol so you all know cortisol as a stress hormone
you’ve probably studied it extensively it is a major regulator of the immune
response and in fact when you go to the pharmacy because you have bad allergies
or a bad rash or an autoimmune disease often what you’re gonna get prescribed
is a topical or an oral version of cortisol that’s been modified slightly
so prednisone dexamethasone those are all just synthetic versions of cortisol
and the reason you take them when you have a rash or asthma or an autoimmune
disease is because what cortisol does is fly across the cell membrane and bind
inside the cell to its receptor receptors call the glucocorticoid
receptor and the reason that cortisol or that synthetic version of cortisol will
treat your rash or make your asthma subside is because that complex cortisol
plus the corticoid receptor Ted nuclear factor Kappa B and prevents it
from going into the nucleus to switch on gene expression so it basically ties up
this linchpin of inflammation so we’re going to study that process – alright
let’s get to some data I said if you have questions before we move into some
data all right so this is some of our earliest work in this area we wanted to
simply ask this question in about 2007 when we started noticing this literature
on early socioeconomic status and later health of whether the data were
consistent with the scenario where the socioeconomic status that people had in
childhood could tell us something about their immune systems functions when they
were adults and in particular whether this inflammatory phenotype that we
hypothesized might be present so we did a very simple study where we recruited
young adults between the ages of 20 and 45 and we recruited half of them to have
come from families of low socioeconomic status meaning that during their first
five years of life their parents had occupations that were mainly in the
service sector and the clerical sector and we recruited a comparison group of
young adults who were matched on age and race and ethnicity and gender who came
from more privileged families where during their first five years of life
their parents were physicians were attorneys were judges were architects
but we matched on current socioeconomic status so we made sure that there are no
differences between the groups on their current occupational prestige and in
doing so tried to our best to isolate the effects of early SES rather than
later SES so we gathered this group of subjects we took blood and as I showed
you in that last cartoon in a test tube we treated people’s immune cells with
various microbial products that made the cells think that they were inside the
and seeing a microbial threat and I’m showing you here two graphs of different
microbial threats that we used in test tubes in that experience over here on
this side is poly IC which is a synthetic substance that mimics a
certain kind of double-stranded RNA virus that’s a common virus you
experience on the other side is flagellin which is a molecule that sits
on the tails of bacteria and allows them to move around and what you can see here
is that for both is viral stimulus and this bacterial stimulus the young adults
who came from more disadvantaged circumstances in early life produce more
of the inflammatory mediator right now you might ask isn’t that adaptive don’t
you actually want your immune cells pushing out more of this communication
molecule that’s gonna immobilize the threat and get you back to a healthy
state and that’s a good question and and I would say probably the answer in that
moment is yes right that you want a stronger immune response if your spouse
or your schoolmate sneezes on you and is spreading a bunch of influenza virus you
want to get rid of it so we wouldn’t necessarily argue that this is
maladaptive in any way I think what we would argue is that there’s a trade-off
and that every time your immune system sees a threat it has a bigger response
that might that trait might be beneficial in early life and it might be
beneficial in the context of infectious disease but since this chronic woe of
inflammation is the culprit in some chronic diseases of aging and
particularly heart disease that the trade-off is that you’re going to be
more at risk down the line for those sorts of diseases of aging so it’s not
just a deficit story or an exaggerated response story it’s probably more of a
trade-off to particular ecological conditions so in this study we did a
couple of other things so one is that we were very interested in trying to not
only listen in on these signaling molecules that
immune cells used to communicate with one another but to listen in on the
genome and understand which genes were getting switched on and which genes were
getting switched off as a way to tell us a little bit more about what was
happening inside the cell and understand that machinery so this is just a summary
of what we find in some more complex bioinformatic analyses the punchline
here is that even when people cells were at rest this is not after any kind of
stimulation with a microbial threat in a test-tube just at rest what we see is
that the subjects who came from families that were more disadvantaged had more
activation of this nuclear factor Kappa B pathway and remember perhaps from a
few slides ago nuclear factor Kappa B is that linchpin of the inflammatory
response that’s the transcription factor that
goes in and switches on all the genes involved in coding for other
inflammatory proteins so even at rest these cells seem to be primed they seem
to be ready to go and they have less activity of genes that involve that goo
corticoid receptor basically what this slide is showing is that the coracoid
receptor isn’t as active at keeping a lid on the inflammatory activity of
these cells in subjects who come from more disadvantaged backgrounds if you
want to use a sort of car betta for what data like this kind of suggests is that
people who come from more disadvantaged circumstances in adulthood their cells
have their feet slightly more on the gas and slightly less on the brakes of
inflammation right the cells are just primed and ready to go all right so
since all of you were developmentalist you might be asking okay so there’s at
least 25 or 30 years between the exposure to low SES and the process
you’re studying in adulthood what happens in between that’s what really
matters so I’m not a developmental psychologist or a child psychologist by
training you’ve probably already figured that out I’m a health psychologist but I
was fortunate to have KD as a postdoc and Katie along with several of your
colleagues told me that you know to really understand this and study this we
needed to move back and do developmental research and that was good advice so all
of a sudden after spending 20 years studying adults I started studying kids
and we’re now finally done with I think a good sized decent study of kids in
Chicago it’s is about I’m gonna show you data from about 300 kids who we were
recruited in eighth grade and we followed it again in tenth grade so
we’ve got now two years of longitudinal data and basically we see pretty similar
results that if you look at 12 and 13 year olds according to socioeconomic
status though kids who are living in families that are more disadvantaged
have way more activity along this NF kappa-b axis again this is the linchpin
of the inflammatory response and you see this in those monocytes that we’ve been
focusing on so we pull them specifically out of blood and kind of fish out the
cells were most interested and you see that pattern and we also in the study
went a little bit further not only did we fish out the monocytes but we fished
out the stem and progenitor cells so basically the mother cells that
differentiate ultimately into a bunch of different immune cells and we show the
same pattern in those cells and the reason that’s important is because your
monocytes and many of your immune cells die in periods of weeks to months they
turn over very quickly in the body so if you’re gonna have any kind of
programming effect along the lines that Barker talks about where experience is
sticky and it hangs around in cells it has to hang around in these stem and
progenitor cells because they’re the ones that continually repopulate tissues
so these findings give us some hints that that might be going on so we see
these NF kappa-b FX interestingly in kids we don’t see any differences along
this glucocorticoid receptor axis so the insensitivity to Gugu corticoids that we
see in adults related to early experience isn’t visible at 12 years old
and otherwise healthy kids it shows up later but that’s not inconsistent at all
with the idea that your appearances can accumulate over time and
we do see some evidence that that google corticoid insensitivity in kids with
health problems and particularly and kids with asthma but and healthy kids
it’s more just the foot on the gas difference that we see so moving back
further in development katie got me on this trajectory of studying kids and we
actually over the past three or four years have moved even further back and
we’re doing some studies in pregnancy and trying to ask when does this
inflammatory process begin does it begin in utero so you guys are all probably
familiar with the placenta right this is the interface between the mom and the
fetus and what you see here in this nice diagram of the placenta is that it has a
key part is this villus layer here right this chorionic villus layer and this
villus layer is where exchange between the mother and the fetus occurs so
maternal blood comes in from this side and it pours into this area through
these spiral arteries that develop and in that blood is nutrients and is oxygen
and nutrients and oxygen diffuse across these villus trees here that are fetal
in origin and then make their way through the umbilical cord to the fetus
and what comes out on the other side is that fetal waste comes out here into the
maternal blood and so that’s the site where you really have this interface
between the mom and the baby that’s critical to all sorts of development and
it turns out that in some pregnancies you can get pretty severe inflammation
in this villus layer and what that inflammation does is obstruct the
transfer of nutrients and oxygen and so this slide right here shows you on a
microscope the upper section a labeled a is a cross-section of a normal villus
tree and you can see all these red dots in there
and these red dots contained within the larger purple pink areas our fetal
capillaries right there basically these arms of the tree where
maternal nutrients and oxygen can diffuse in and get to the fetus what you
see so that’s a normal cross-section of the villa’s tree of a villa street in B
here we have one that’s chronically inflamed and particularly right here you
probably can’t see it you have the infiltration all these purple dots
signal infiltration of predominantly maternal immune cells and there are no
red dots here all of the fetal capillaries have been obliterated by
virtue of the inflammation from maternal cells and so and it’s showing you in
adjacent villi in this case there’s plenty of fetal capillaries you don’t
have inflammation that it’s pretty focal to this one so in our first pregnancy
studies we just wanted to ask are there differences in inflammation in this
villas later according to maternal socioeconomic
status and so we started doing some work with a placental pathologist and in a
very straightforward simple study the answer we got was yes and they’re pretty
linear and pretty profound what you’re looking at here on the y-axis is just is
there any chronic inflammatory region in the placenta at the time a woman gives
birth and we’re looking at lesions in the villus area but also in this
particular study we’re also looking at lesions on the maternal side of the
placenta and in the membranes and in all three areas there are differences by
socioeconomic status this is an aggregate measure but you can see that a
woman from a low-income background is twice as likely to have a chronic
inflammatory lesion in this study as a woman from a high-income background now
I should say that most of the lesions we see in this work are fairly low-grade
pathologist typically rate them low medium high grade we don’t see a ton of
high grade lesions nose the ones where quit
one would be very concerned about risk for fetal growth restriction and adverse
neurological outcomes in the long term these are kind of mild lesions typically
but they do pattern in the way we do expect we’ve also gone and our postdocs
and graduate students love this we actually show up in the delivery room
and once the placenta is delivered we biopsy parts of it so we actually cut it
open and we take out but all rice sized pieces of that villus layer and then we
can use molecular techniques to study the gene expression patterns and the DNA
methylation patterns in that place where the maternal and fetal communication
occurs and so we can kind of dissect that molecular lee and when we do that
this is our first study in the area and it’s you know it’s a small kind of
proof-of-concept study so I you know I want us to replicate it you know in a
larger sample before I get too confident but the pattern we see in these little
biopsies we take is that the lower-income women have more immune Avot
more evidence of immune activations and kind of low level inflammation in the
villa’s layer their placenta and the placenta also sends out a bunch of
signals that encourage fetal development so the placenta sends out signals that
cause the brain to keep growing that caused the heart to keep growing and
what we see is that in lower-income women these signals are kind of turned
down relative to higher income women again we don’t know how clinically
significant if at all they are we don’t have a big enough sample in this study
to really look at developmental outcomes and even if we did I’m not sure
differences of this size would merit any clinical concern so I don’t want to over
interpret the findings but what these preliminary data do suggest is that
there’s a sign even in utero of socio-economic differences in what’s
happening in that keep essential layer and so we’re following this up we’re
about 200 women into about a 700 woman study where we’re gonna have a
relatively large sample for this kind of biopsy genomics work to react sees
questions and see how things look and be able to relate it to
more developmental outcomes so I’ve talked all about risk up until now right
I have completely overlooked the fact that we and everybody else in the room
knows that the effects of socioeconomic status and other early adversities are
usually highly conditional on other characteristics of kids of the families
they live in of the communities they reside in and that protective of owner
ability factors are very very critical to not only knowledge but to study so
let me tell you a little bit about our work in this area so the first thing we
found and I’m gonna kind of blitz through this so I don’t go too long is
not gonna surprise any of you which is that we consistently find that for
low-income kids having supportive parenting and particularly a nurturing
caregiving environment from the mother is quite protective and almost
completely eliminates the effects of socioeconomic status that we see on most
health outcomes and we see this when we look at pro-inflammatory activity in
cells so basically having high maternal nurturance obliterates most of the
pro-inflammatory effects of Lois yes we see this in larger cohort studies these
are data from Midas where we’re looking at childhood SES as it relates to risk
for metabolic syndrome in midlife and what we see in that case is that there
is a socio-economic gradient in metabolic syndrome if you grew up in a
household of lower socioeconomic status you’re more likely to have metabolic
syndrome at midlife metabolic syndrome being this constellation of risk factors
for heart disease that include abnormal cholesterol abdominal obesity and high
blood pressure and fasting glucose impairments but that that gradient is
almost completely absent if you simultaneously recollect having
high levels of maternal nurturance and we see this pattern over and over and
over again different observational studies at different stages of the life
course again to folks who study developmental psychopathology for a
living this will not be a surprise this will be very
familiar but to folks in the physical health arena who haven’t done this work
this was surprising and reassuring about an important protective factor and one
that’s heavily modifiable along those lines a few years ago we began
collaborating with one Katy’s colleagues at Georgia named Jean Brodie who’s a
fantastic developmental psychologist that spent his career studying rural
African Americans around Athens Georgia and has pioneered family oriented
interventions that are administered some time around the transition into middle
school and have strong and durable effects on children’s mental health
academic outcomes and a variety of different human capital outcomes but
Jean historically hadn’t been interested in physical health and that’s where we
came in as collaborators and so one of the first things we did is go out into
the field and collect blood samples from a few hundred 19 year olds who when they
were 11 years old had been part of genes randomized clinical trial testing this
intervention so that intervention wasn’t designed to look at physical health it
wasn’t designed to look at inflammation the goal was to reduce drug use and to
improve mental health outcomes but he was still following them eight years
later and we said why not take a look and what we found was that the
intervention was associated with reduced inflammation eight years later just to
give you a feel for the sample because these aren’t the usual samples you see
in psychology journals so again these are rural African Americans from a few
hours radius around Athens Georgia about what am I looking at here about half the
sample has an income to poverty ratio below to the percent with an A a or a BA
degree in the sample is about 10% so most of the families have a high school
education or some college and the modal family is a single caregiver household
so it’s a demographic at risk sample and what we see is that
when you look at age 19 at kids who got the intervention versus kids who don’t
that there are market differences in inflammatory markers and this is a
composite of six different measures of inflammation that we use the D values
here on individual measures range from about point four to about point six
they’re higher than I would have expected but you can see they’re
whopping really different here eight years after the intervention now you
might say what happened how is it that an intervention at eight years old or at
11 years old changed inflammation nine years later a
bunch of analyses trying to understand what roles different sort of
psychosocial and familial changes played and the big operator here was
differences in parenting which is what the intervention intended to do the goal
the goal of the intervention was to increase nurturance involve parenting
and to decrease harsh inconsistent parenting and in families where that
happened and you saw big changes and overall parenting violating different
scores you showed the biggest reductions in inflammation eight years later so we
we don’t know because the study wasn’t designed to measure inflammation exactly
what the critical ingredients for changing that was but we are pretty
certain that it has to do with changes in the parent-child relationship that
played out over the course of a in middle adolescence and we think played a
role in helping children helping told and manage some of the challenges that
came with moving into middle and later adolescents around coping with racism
and discrimination around coping with limited opportunities in these rural
small towns and in coping with physical activity challenges in small towns that
don’t necessarily have easy places to exercise and in healthy diets those are
sort of our hypotheses and we’re going into the field hopefully in about a year
to try to replicate these findings in a trial that actually has
a priori pre-specified hypotheses around all this stuff so we should know in five
years or so whether we can replicate these but they’re interesting and
provocative suggestions that parenting and especially nurturing parenting can
buffer some of the associations we see between socio-economic disadvantage and
subsequent inflammation positive story we’re also doing a bunch of work in
pregnancy again trying to leverage social relationships in that last sort
of module of the talk the focus was on parent-child relationships and our
pregnancy work we’re focusing more broadly on helping helping women in the
middle of pregnancy build relationships that are closer either with their
partners or with other pregnant women in the community and so we’re involved in a
clinical trial focusing on group group based prenatal care and having groups of
eight to ten women go through prenatal care simultaneously in addition to the
usual medical aspects of care to pay more attention to relationship building
and stress management and coping and interactions of the medical system this
often gets called centering pregnancy that’s the formal name of the
intervention this is a riff on that and we see in some very preliminary work
that participating in this group based prenatal care changes gene expression
patterns in the placenta and an umbilical cord blood in favorable
directions these are very preliminary observational data but they helped us to
get started on a large randomized controlled trial that we’re about
halfway through we’re randomly assigning 3,000 pregnant women in a small town in
South Carolina to either get group based or individual based prenatal care
looking at birth outcomes and then again we’re in the delivery room biopsy Allah
sent us to listen in on the gene expression patterns so again we’ll know
a lot more in a few years about how robust these associations are and how
clinically valuable or not they are alright
so good news about maternal nurturance right all the data that we
it’s not just from us but other groups are showing some more patterns suggest
promising indications that it’s a protective factor and that it’s a
modifiable protective factor and that if we make better policies and have better
practices that allow for better quality family time and better quality family
interactions we might mitigate through social processes some of these
socio-economic disparities that’s that’s my read of the data the optimistic read
the other protective factor we’ve really focused on over the past few years again
out of the developmental psychopathology literature is self-control and we went
into this work thinking all right there’s plenty of evidence to suggest
from studies of academic outcomes and substance abuse outcomes and other
mental health related outcomes that low-income kids who have high levels of
self-control or high levels of executive function seem to be protected from some
of the adverse outcomes that typically pattern by socioeconomic status okay the
maternal nurturance findings transfer from psychopathology and academic
outcomes to physical health presumably the self-control one should too so again
in collaboration with Jean Brodie following up on some of the samples we
started to ask these questions in longitudinal studies about whether
self-control was protective against physical health uh outcomes for
low-income kids and the answer to this for us has turned out to be not so
protective actually and this is an op-ed that we wrote in the time some years ago
reporting the early phases of this research which suggested the troubling
and somewhat paradoxical conclusion that upward economic mobility and high levels
of persistence and striving might have unanticipated unforeseen negative health
outcomes particularly for very low income kids of color and let me tell you
a little bit about that data this is less optimistic and cheery than the
maternal nurturance findings so I’m just gonna summarize some findings here as
you may know Jean as multiple cohorts of rural
african-americans some of them are part of studies that started around the
transition to middle school some of them are folks who started during the
transition into it although this is a study called aim adults on the making
where the participants were followed from age 17 to age 20 this is different
than the intervention data that I showed you a totally different cohort and so
what we’re doing here is looking am in about 300 young late adolescent early
adults these are all rural African Americans at psychosocial outcomes
annually from age 17 to age 20 as a function of family socioeconomic status
and self control self control again measured annually by multiple reporters
by the kid and by the parent here and we’re averaging across all those and if
these slides were bigger what you would see here is that regardless of whether
you’re looking at substance use aggressive behavior depressive symptoms
or a more generalized composite of internalizing behavior what predicts
levels of the psychosocial outcomes from age 17 to 20 is not socioeconomic status
its self control right so you can see each of these four lines is basically a
cell and a 2×2 matrix that through latent cross-class analysis gives us
groups that are high or low in socioeconomic status and higher low in
self-control and basically there’s a main effect of self control SES doesn’t
really matter so again this is very consistent with previous literature
we’re not discovering anything new here we are at best replicating other
findings that self-control is beneficial for adolescents as they move into
adulthood for many outcomes but then if you look at physical health and in this
particular cohort there weren’t physical health measurements that were extensive
they had almost no blood they didn’t have blood pressure but they did have
some DNA sitting around because the early focus of the study was on genetics
and so we’re able to purpose that DNA to look at a metric of how fast people
cells are aged and you can do this using methylation
patterns in cells to gauge basically what the person’s biological age is
relative to his or her chronological age so this is an evolving metric it’s
attracted a lot of interest in recent years it’s not the be-all and end-all of
people’s health but it’s one way when you have a limited amount of sample on
health that you can start to ask questions about whether self-control
relates in the same way and what we found to our surprise was that actually
the effects of self-control or the relationship between self-control and
the cellular aging metric were not as favorable as they were for psychosocial
outcomes that there was this prominent interaction between socioeconomic status
and self-control and it was an interaction where if you were a rural
african-american adolescent from a very disadvantaged family having high levels
of persistence and self-control was related to faster cellular aging your
cells looked older than they should have worked given your chronological age
whereas if you were an adolescent from one of these same rural areas and you
had low levels of self or and you if you came from a family that was less
disadvantaged that was working class or middle class probably in more normative
terms the self-control was actually beneficial for you in terms of
epigenetic aging that your cells age more slowly and we saw this interaction
not only with this particular metric of epigenetic aging which is one that was
developed by a guy named hanim but also with a different measure that was
developed by another scholar named hanim and the two were correlated at about
point two so we see this with different measures of epigenetic aging we also
have seen this in a different sample with the same exact pattern with
allostatic load and so in two of genes cohorts we
see the same pattern where self-control portends better academic and mental
health outcomes and in terms of physical health it portends better outcomes if
you’re a kid in an advantage family a relative advantage family and worse
outcomes if you’re a kid from a more disadvantaged family than it really
depends and we really spent a couple of years scratching our heads about what
the heck this means because it wasn’t anticipated it wasn’t hypothesize we
figured that self-control would be unification uniformly beneficial and in
presenting these data to some colleagues of ours who came from different
disciplines who came from economics and public health and sociology they all
said oh of course this makes perfect sense what do you mean it makes perfect
sense not in psychology and they said haven’t you seen the data on mobility in
blacks in the United States and they showed us this graph so this is a graph
that looks at the gap in longevity between black and white Americans right
and it stratifies it by a level of education and what you see here is that
white Americans live an average of 3.1 years longer than black Americans if you
just focus on Americans who don’t have a high school degree right and if you look
at high school graduates the gap is about 3.7 years
favoring whites over blacks right where’s the biggest gap between whites
and blacks college degree this isn’t what you’d expect if education was
necessarily the problem that needed to be solved here right the mortality gap
between blacks and whites with college degrees is about 4.25 years right so the
racial disparities in longevity and in health our most extreme again suggesting in this kind of
roundabout way in the same way our data is that more education and potentially
more affluence and maybe more striving to get that education in a country in an
environment that historically has been very unsupportive is associated with
some sort of physical health and that’s what we mean when we’ve talked in this
editorial about hidden costs or mobility or in some cases some people talk about
this as skin deep resilience that this hard-driving persistence that allows
some Americans to live out this mythology that you pull yourself up on
the boots by the bootstraps and you work hard to you can get anywhere in America
might be associated with outward indicators of success high educational
attainment potentially good mental health lots of income but for some
Americans that path towards upward mobility might exact a health cost and
so I showed you a couple of different slides from South Georgia again you know
in publishing these studies and and thinking about these findings we had no
idea first of all how robust they were we had no idea if they were something
unique about rural Georgians and rural African Americans or whether they were
more generalizable to other disadvantaged groups in the United
States so we recently did some analyses of the add health data set with Kathy
Harris who’s the PI there and Lauren gay Tosh who’s a sociologist now at
Vanderbilt to ask this question in a nationally representative three
decade-long longitudinal study and get what we thought would be a more
definitive answer and so here in at health what we’re looking at first of
all the mental health outcome this is just CESD scores people in early
adulthood looking at it according to childhood disadvantage so this is a
index of disadvantage across family neighborhood and school that’s done
during adolescents by whether or not the subject ever gets a college degree this
is a four year degree and if you look at mental health outcomes
this is depressive symptoms you get a very clear answer which is going to
college is good for your mental health right
subjects who got a college degree have lower depressive symptoms between age 24
and 32 and if you redo the analyses only focusing on people who have already
finished college you see the same thing suggesting that this isn’t necessarily a
social selection effect point being mobility is good for your health it
doesn’t matter in this sample if you are white if you’re black if you’re a Latino
or what for the most part your childhood socio-economic circumstances were if you
went to college or mental health is better not so with your physical health
so here instead of depressive symptoms we’re looking at metabolic syndrome that
constellation of cardiovascular risk factors that includes abnormal
cholesterol high blood pressure abdominal fat what you see here is that
for white Americans going to college predicts reduce risks of metabolic
syndrome and it does so for the most part regardless of what your childhood
circumstances are but that’s not true for black Americans and it’s not true
for Latino Americans for these groups going to college is better for your
long-term metabolic health if you come from a family that was middle class to
advantaged in the first place but if you come from a family that was
disadvantaged there’s this apparent health cost so again this is a very
different data set it’s much larger we’ve got three decades of prospective
data so I think we can come to some stronger conclusions here than we could
in South Georgia none of them is a happy conclusion
they’re fairly sobering but what they do suggest is that there potentially are
some costs to striving but maybe even more importantly their cost to arriving
at high educational success and workplace success that maybe have a lot
to do with and I can’t get through this right now our suspicion is that these
are really effects about climate that for low-income minority students who
strive and succeed at the highest levels they find themselves an academic and
workplace that if not aren’t particularly
welcoming are in some cases outright hostile and it’s that climate that
really erode some of the physical health we have a paper that is consistent with
that finding that I’ll just show you here that another postdoc Cynthia Levine
just published in PNAS suggesting that if you look at climates around diversity
it can tell you a lot about how the health of kids in middle school going we
think the school climate issues are really really important in explaining
those skin deep resiliency mobility all right I have gone too long
thank you very much

Leave a Reply