Life Expectancy Impacts of Surviving Wartime Incarceration: Evidence from American World War I Prisoners of War

Evan Roberts, University of Minnesota

We add new evidence on the long-term effects of surviving extreme stress, by examining the survival experience of American POWs from World War I. We construct a sample of American POWs from a listing of the names of American prisoners published in 1924. We ascertain genealogical details on men who served in states with near-complete collection of post-World War I service questionnaires. These questionnaires provide sufficient identifying information to allow the men to be linked to post-war census records in the IPUMS MLP, and traced to the 1920, 1930 or 1940 censuses. Genealogical data from Ancestry or FamilySearch provides additional data on the lifespan of the prisoner, and siblings. Using the service records we draw a random control case who was close in age to the POW but not captured. Taken together, we then have data on the lifespan and post-war occupational and marital outcomes of the POW, any siblings who served in World War I from the same state, and a random control case. Our case-control analysis suggests that the impact of being a POW on men's life chances was significant when compared to siblings who served without being captured, and the unrelated control case. POWs lived 5-7 years shorter lives than controls. Data from the service questionnaires reveals that most American POWs in World War I were only imprisoned for a short time, largely being captured in the autumn of 1918. These short terms of imprisonment stand in contrast to experience of British Empire POWs of the Germans who were often imprisoned for 6-24 month. Our paper illustrates how case-control methods can be implemented in historical samples.

No extended abstract or paper available

 Presented in Session 26. Mortality Over the Life Course: Effects of Social, Family, and Historical Contexts