Monday, November 12, 2007

Nätverk

Nätverk is the Swedish word for network.

This afternoon at CHESS there was a talk given by a guy from Harvard, whose work is in health care policy, sociology, and medicine. The talk was called "The Spread of Health Phenomena in Social Networks" and it was really, really interesting. He focused mainly on the work they've done looking at the spread of obesity in social networks - they recently had that paper published in the New England Journal of Medicine - although they've also looked at other things like smoking, drinking, happiness, and banana eating. They've basically used all the data from the Framingham Heart Study (and the other Framingham studies) to construct elaborate representations of the participants' social networks.

As part of the studies, which started in 1948 (at least the heart one did, which is the most famous), people were tracked every four years and information that was recorded included things about who their friends, families, and neighbors were, and because these Framingham studies were so large (and Framingham so relatively small, I suppose) there ended up being a huge amount of overlap such that these guys were able to make very detailed (although of course incomplete in many ways) maps of people's social networks. And then they're working on all sorts of incredibly elaborate statistical and mathematical methods for working with this network data to look at how things like obesity spread. They've found significant clustering, i.e. patterns within social networks that are not random, which suggests transmission and a type of social "induction" that is independent of other confounding factors. He had this incredible phrase about "peaks and valleys in sociotopological space," and part of the modelling they did was based on this degrees of separation thing that some physicist did for work on some gene. Anyhow, it was a fascinating way of looking at how there are all sorts of strong effects that people have on one another, sometimes in ways that are intuitive but sometimes in ways that are quite unexpected. We talk all the time about how obesity is (or might be) an epidemic, and this work of his in particular really speaks to the transmission and contagion aspect of what we think of as epidemics.

Someone asked a question at the end about possible implications for prevention in the case of the obesity work, for example, and I really appreciated his totally straight-forward response that he and his colleagues consider this purely social science research right now and as such there don't need to be implications all the time. This reminded me of the conversation I had with Ilona on the way back from the conference in Helsinki, which was really helpful to me. Basically, we were talking about clinical work versus research, and I was explaining how important it is to me to get to do clinical work because I don't think I would be happy just doing research for a variety of reasons. The way my mind works, I always look both towards the underlying causes and mechanisms, as well as the specific implications and uses, and with a lot of both social science and more biomedical research, often there is very little information about both ends of that spectrum. And obviously it's not because people are lazy or poor researchers, but rather it can be a feat just to demonstrate that correlations exist or that there seems to be some sort of causal effect. It's much harder to pin down actual or even possible mechanisms, and it's difficult to predict or suggest policy or practice implications. I have a tendency to get overwhelmed and frustrated with research. That was one of the things I found with my psychology work at Columbia, and I realized that it was too personally frustrating for me to be involved with work that is fascinating and brilliant but that often seems to have little to no ties to the actual world, or to policy, or real changes.

Anyway, Ilona was trained as a pediatrician but chose to go into research, which is really interesting to me. One of the things she said was that she really values her medical training because it really helped her become the kind of thinker who focuses on intervention and action. She also made some really nice points about the importance and benefits of public health type research even when it can't necessarily illuminate all the mechanisms or implications, i.e. no one would ever get anything done if you had to figure all that out before publishing papers and giving talks, and that it's important to do the best you can with incomplete information. She made an analogy to medical situations, the idea was that we try to treat cancer even though we don't totally understand exactly how those cancer cells are functioning. We do our best with what we have and what we do know.

Now it's 3:45 and it's dark out, which makes it hard to stay motivated even though the seminar was the first concrete thing I did today. At least I now have a nice person at Harvard who invited me to come visit if I get offered an interview there (I did get one at Brown, which is semi-exciting, and I haven't heard from anywhere else yet which has put me into a constant state of panic). Also, it snowed this morning, and it was pretty. And now I need to keep translating Swedish disease names, then go to the Sturebadet to do some running so that I don't fall off my marathon training schedule before I've even started.

No comments: