Incident cases of cancer among cohort members were
identified by computerized record linkage to the Canadian
Cancer Data Base. The cohort has been derived from a similar
cohort used in an earlier mortality study (3). Cohort members
who had no exposure data after 1969, the first year for
which cancer incidence data were available, were excluded.
Thus, the cohort is somewhat smaller, and the follow-up time
shorter, than in the National Dose Registry mortality study
(3). However, the number of incident cases should, proportionally,
exceed the number of cancer deaths, especially for
cancers with low fatality rates. For such cancers it may be
possible to assume associations with ionizing radiation that
could not be identified using mortality data alone.