A cohort analysis of pancreatic cancer, 1939--1969.

Pancreatic cancer mortality in the United States was examined by cohort analysis for the period of 1939 to 1969. Birth cohorts at 5-year intervals were studied. White males and non-whites of both sexes showed a shift of the cohort mortality rate curves by age toward younger groups as birth cohorts went from 1870--1874 to 1900--1904. This shift disappeared in white males and diminished in non-whites of both sexes between the cohorts of 1890--1894 and 1900--1904. The shift in non-whites was larger than in white males so that although non-whites had lower rates than whites in earlier birth cohorts, their rates exceeded those of whites in the most recent cohort. In the search for environmental cause(s) of pancreatic cancer, exposure characteristics should be compatible with these observations.