1) Olivia Judson this week on the possible effects of early life influences (in particular maternal obesity and its affects on the fetus) on political behavior later in life (hat tip: Daniel Rothschild).
I'm glad to see this stuff in the popular press: some of my colleagues and I are looking at the influence of birth year and early childhood conditions on people's forward looking behavior, aversion to risk and expectations later in life, and pieces like this should whet the public's appetite! I'll be summarizing preliminary results of that research, as well as other work (including the early life effects on obesity) in a later post. Stay tuned.
2) In an earlier post, I briefly discussed a paper by Douglas Almond and Bhaskar Mazumdar about the negative effects of being in the womb during Ramadan on health outcomes later in life. A working version of the paper is now available via the NBER. The authors find these stunning results for two separate samples/populations: Arabs in Michigan and Muslims in Uganda.
3) But are early life influences quantitatively important in explaining trends in population health? Are such influences important levers for public policy? This recent Slate article suggests that the answer for both questions is "no." Here are two excerpts:
"The problem, though, is that large-scale problems also wrongly get blamed on the womb—and, by extension, on the woman who houses it. Womb-centric predictions of a child's future—whether rooted in supposed genetic disparities, gestational maternal-fetal conflict, eating habits during pregnancy, or whatever else—always undersell the role of one's later environment...
"Turning to the womb to explain complex social and public-health problems ultimately means people have given up on changing the things that really matter. That's too bad. The truth is that nothing in this world worth having comes easy. And as any hard-working student who made it to college, overweight person who's changed his or her lifestyle, or adult who's worked through depression can tell you, at some point you have to stop blaming your issues on your mother's uterus."
I think the author misfires on two of his (implicit) points. First, early life conditions do not occur deterministically. Public policy has tremendous scope to intervene into the health of pregnant women and newborns, and these interventions have been shown to have positive health and socioeconomic benefits both in the long-run and across generations. Second, while we can't retrospectively change what happened to a grown-up when he/she was in the womb, those experiences may still be important for the kinds of "here and now" interventions the author suggests we should focus on. For example, in my own work, I find that the body weight and blood pressure of individuals who experienced adverse economic shocks early in life is much more sensitive to food prices and income in comparison to those who did not.
Where I think the author does have a case is that we don't know for sure the extent to which these gestational and early childhood influences can explain cross-sectional and time-series trends in health outcomes. This is a question I am really interested in tackling, though I haven't really been able to get the data I need.
4 comments:
Interesting stuff. You might want to check out a paper by Heckman and others about the formation of cognitive and non-cognitive skills:
http://www.nber.org/papers/w13810
that ramadan paper is extremely interesting but also depressing. it would probably be very difficult to get mothers to stop fasting for ramadan in order to improve the health of their children. even if the evidence says they should.
i saw a presentation by a post-doc in sociology the other week about early life health and educational attainment. the co-author, a pediatrician, interpreted the results as meaning that pediatrics is important. but from a policy perspective non-medical interventions are perhaps more interesting. and when theres variables like neurological defects that may be amenable neither to policy or to clinical care, then it doesn't really help us.
Jeremy,
Is this the Yale Soc paper you were referencing:
http://www.yale.edu/ciqle/PUBLICATIONS/Harkonen_et_al.pdf
I agree with your point that this kind of research is only useful when there is a policy implication that comes out of it. I learned the lesson the hard-ish way when some of my advisors nixed a project about long-term earthquake damage I wanted to undertake because there is no clear policy intervention to an earthquake.
However, I think there are often more policy implications than one might think. For example, the Ramadan study is a bit depressing and you wouldn't expect behavior changes. On the other hand, it might be possible to develop some kind of nutritional supplement given in the evenings to quell the negative impact of fasting.
I'll be curious to hear your thoughts on my own stuff at the HIP seminar next week. I am trying to move towards evaluating long-run effects of public health interventions, perhaps the kind of non-medical things that you might find interesting.
yeah, thats the paper. the other thing weird about that was the use of multiple imputation. how can you credibly impute things like neurological disorders?
Post a Comment