Fergus HW3

Fergus HW3

by Kirkpatrick Fergus -
Number of replies: 3

1.         Weaver et al propose that among rats, maternal behavior towards newborn pups influences their cortisol response to stress via epigenetic mechanisms that change the expression of glucocorticoid receptor gene for the rest of the pup’s life.  They argue that because epigenetic patterns are established at specific developmental periods, there is extreme time sensitivity to when the pup is exposed to particular maternal behaviors (licking and grooming, in this case), and maternal behavior before or after that sensitive period window is not as important.   Do you think this mechanism is relevant in humans?  If so, what behaviors are most analogous to “maternal licking and grooming”?  

This is a fascinating article and topic, as epigenetic changes can be triggered by environmental influences, conflating our assumptions about genetics, inheritance and “nature” if you will. It is also fascinating that behavior modification and changing the rat pup environment (or even infusing a histone deacytlase inhibitor) can REVERSE the epigenomic changes brought about by stress. I do think this mechanism is relevant in humans, but without studying the methylation patterns, we are left to speculation. Rat/mice study findings are known to have limited generalizability to humans, so we cannot say for sure. However, I do think that epigenetic changes at critical time periods in humans occur. This can be during fetal development (think environmental exposures for the mom/fetus), or shortly after birth like in the rat study. Assuming it is similar in humans, analogous behaviors to “maternal licking and grooming” are hugging, kissing, playing, smiling, cuddling, feeding, and patting on the back. If you broaden the perspective a bit, I would think it important to consider factors that may have a negative influence. This includes neglect (not performing behaviors described above) or more severe forms of neglect, parents arguing/yelling, violence, lack of food, smoking or smog exposure, or really any number of negative environmental triggers for a newborn baby. All of this could have an influence on epigenetic regulation of glucocorticoid receptors (either upregulation or downregulation).

2.         Gruenewald, in contrast, emphasize the cumulative effects of SES adversity on a multi-system allostatic load measure.   Do you think that the Gruenewald findings are consistent, inconsistent, or unrelated to the Weaver findings?  Explain.  

This question highlights two different (but related) theories of how epigenetic and environmental influences have biological consequences. The first theory is exemplified in the Weaver article, which is that “insults in early development periods have enduring consequences, possibly because the organism makes developmental choices optimized for a different environment (“fetal programming”).” The second is shown in the Gruenewald article, which is that “large and small insults each have small physiologic consequences which accumulate and accelerate aging” and the body can at some point no longer compensate for insults. I like to think of this as 1) a window early on in development that has lasting influence; and 2) a cumulative effect model where exposures are summative. To answer the question, I don’t think these two theories are incompatible. I think they each add a unique vantage point to understand social and environmental influences on our biological health. It is not unreasonable to think that both are true! Maybe some gene expression patterns are determined early on, and others are the result of insidious insults. By contrast to the prompt, I don’t think the article findings are “unrelated,” so I suppose I have to choose consistent even though they demonstrate different theories. I think we need to investigate both to get to the bottom of understanding the pathologic mechanism.

 

3.         Hertzmann and Boyce argue that “it is not genes or environment, nor is it genes and environment, but rather it is gene-by-environment interactions that influence developmental trajectories.”  To what extent do you think that GxE interactions can contribute to major disparities along racial/ethnic, socioeconomic, or geographic dimensions?

Hertzmann and Boyce seem to take the middle road and suggest “childhood environments work together with genetic variation and epigenetic regulation to generate socially partitioned developmental trajectories.” Further, they argue biological embedding and the force of social causation are drivers of disease and disorder through gene-environment interactions. They provide examples of this using child development as an arena of study: play; neighborhoods; SES; classroom social position; and even childcare, family support, and public health programs as environmental factors that have been studied and shown to have a role in biological embedding. To answer the question, I do think that gene by environment interactions contribute to major disparities along numerous dimensions. Clearly there is genetic variation, and if environmental influences, triggers and circumstances work synergistically to produce biologically embedded outcomes, I wouldn’t be surprised. For example, if SES and race are tightly aligned as they are in the US, it would seem that a disproportionate number of poor and minority patients would have an identical (and adverse) social influence on their genetic makeup. This agrees with Gruenewald’s theory of allostatic load because these influences can be cumulative. Further, as these authors are studying children at developmental time periods, it seems reasonable that their findings are further evidence for the importance of environmental factors during certain developmental windows. Thus HEALTH disparities will be observed along the lines of SOCIAL disparities, which is an observable truth. Although the pathophysiologic mechanism is still under investigation, I think this is an example of an epidemiologic “black box” in which understanding the molecular mechanism is not necessary to implement meaningful change. In other words, fixing social disparities will go a long way in fixing biologically embedded health disparities.


In reply to Kirkpatrick Fergus

Re: Fergus HW3

by Maria Glymour -

Thanks Kirk for the nice discussion.  See my comments to Ryan but generally although in some cases we don't need to know the molecular/biological mechanisms, it is often useful to show evidence that a social exposure is really causal; to identify a wide range of possible interventions, some of which may be more feasible than others; and to rally political commitment to addressing inequalities.  W/r/t what's causal: many social adversities cluster - eg poor kids are more likely to encounter psychosocial adversity as well as material adversity -- so it is difficult to know how to prioritize interventions.  Do we focus on material resources or psychological exposures and support?  If we had unlimited resources, perhaps we wouldn't have to make this type of decision, but that is not the situation.  And very often, intervening on one domain may have unintended adverse consequences, so it shouldn't be taken lightly, unless we are quite confident about the causal connection with poor health outcomes.  For this reason, every shred of better evidence we can acquire about causation is important.  Biological mechanisms are an important source of such evidence.

Maria



In reply to Maria Glymour

Re: Fergus HW3

by Kirkpatrick Fergus -

Thanks Dr. Glymour! I agree it is worth investigating mechanisms for the purpose of efficient use of resources, this makes total sense. I do wonder, given the context of our other epidemiology courses, whether there is current work in regards to causality that addresses themes raised in our prior epidemiology coursework. In epi 203 we discussed 6 ways statistical association could occur: chance, selection bias, measurement bias, confounding, reverse causation, and true causal relationship. Given the complexity of the socioecological model, to what extent is it the same or more difficult to determine causality in regards to biological embedding? I only ask because I am picturing myself drawing a DAG and getting a headache. How do we strike a balance between unearthing vital information to allocate resources and avoiding the harms of assuming causality when in reality one of the other five reasons explains the association? Would trial and error interventions end up being more efficient in the end from a cost/benefit standpoint? I feel like "more research is necessary" is probably the right answer, but was curious what your thoughts were. Thank you for such a quick response!

In reply to Kirkpatrick Fergus

Re: Fergus HW3

by Maria Glymour -

Really glad you are asking these questions about establishing causality regarding social determinants of health.  It is definitely tough - probably tougher than some domains of epi (e.g., clinical settings where you sometimes have good info about how the exposure is allocated and get to do RCTs relatively frequently) but not worse than others (e.g., nutritional epi - really tough to disentangle).  My view is that quasi-experimental approaches leveraging events such as policy changes, lotteries due to insufficient resources, or other apparently arbitrary factors that influence social exposures will provide the most compelling evidence.  Usually those studies have their own limitations so are best interpreted in combination with more conventional observational studies.  Rita Hamad will give the social policy lecture in the class - she is a leader in using this type of approach so hopefully will give you a flavor for it. 

Maria