Women younger than 35 years of age were excluded
because of the small number of expected cases.
Women older than 69 years were not included in the
analysis. Since information on employment history
was available only from 1974, older subjects would
potentially end up with too short employment histories
for a meaningful analysis. Controls were extracted
from the regional list of citizens registered with the
National Health Service as of December 31, 2005,
who worked in the plant for at least one year between
1974 and 2005. Thus, the cases and the controls
belonged to the same base population.