Participants in the Terman Life-Cycle Study completed open-ended questionna
ires in 1936 and 1940, and these responses were blindly scored for explanat
ory style by content analysis. Catastrophizing (attributing bad events to g
lobal causes) predicted mortality as of 1991, especially among males, and p
redicted accidental or violent deaths especially well. These results are th
e first to show that a dimension of explanatory style is a risk factor for
mortality in a large sample of initially healthy individuals, and they impl
y that one of the mechanisms linking explanatory style and death involves l
ifestyle.