Titanic-like gradients

In the world of social determinants there is one teaching tool that bests all others.  Deaths on the Titanic were highly skewed towards men and those in the lower classes.  This is an excellent point from which to begin to present information on the manner in which health is patterned by social processes, and disproportionately affects some groups within society.

In recent days there has been a certain amount of publicity surrounding a new paper (gated version, ungated version) by Mikael Elinder and Oscar Erixson at Uppsala University that looks in detail at survival in 18 maritime disasters between 1852 and 2011.  The authors’ goal was to look at gender differences in survival and how this might be linked to social norms.  Their headline result is that women have poorer survival rates than men (by about 20%), although things have been improving over time.  Sadly, as John Timmer at ArsTechnica notes

before we conclude that chivalry was dead and [is now] seeing a revival, we’ll caution you that it may be just that more women now learn how to swim.

My interest, however, is on the social side of things, and here two items jump out at me.

1. First, the authors make the following assertion in the final sentence of their abstract:

Taken together, our findings show that human behavior in life-and-death situations is best captured by the expression “every man for himself.”

However, both Timmer’s commentary and the article itself seem to gainsay this by noting that in cases where the Captain issues the order “Women and children first”, the female survival penalty is roughly halved.  For me, this points to the importance of social control in addition to social norms.   The authors even note elsewhere in their piece:

it seems as if it is the policy of the captain, rather than the moral sentiments of men, that determines whether women are given preferential treatment in shipwrecks.

2. Second, only a minority of ships contain information on class of passenger and the authors look at this only as a footnote.  But, it does appear that being in 1st class was protective relative to being in 2nd or 3rd class overall (there were 7 ships with class info: 3 significantly positive results, 1 significantly negative).  So it does seem that the SDoH classes can keep using this example for now at least.

H/T to Marginal Revolution for drawing me to the paper.