Jon Zelner Social epidemiology, infectious diseases, and statistics

Social networks and diarrheal disease

Thanks to Andrew Gelman for writing about the AJPH paper on social networks and diarrheal disease I did with Joe Eisenberg, Jim Trostle, Jason Goldstick et al. a few years ago on the Washington Post’s Monkey Cage blog!

I thought this might be a good opportunity to bring up my favorite figure from the supplement of this paper (all credit for this one goes to Jason Goldstick), which didn’t make it into the main text because of space limitations:

<img src=”/images/high_low.png” align=”middle” alt=”A rising tide raises all boats!”, style=”centeredImage”/>

Difference (in standard deviations) of household degree from village average degree


What the figure shows is that the impact of relative network position (i.e., how far above or below your village average you are) can help (or hurt) your risk of diarrheal disease in villages with low average degree, i.e. that are less socially cohesive and have poorer sanitary infrastructure, but relative position is less important in highly-connected, socially cohesive villages.

I like to think of this as the “rising-tide-raises-all-boats” figure, and have always been fond of it because it adds a bit of nuance to the overall message of the paper, which is that social networks can be conduits for information that is protective against infection in addition to be being structures of disease-transmitting contact.