Where the cut-off should be depends on the
implications of the test, and receiver-operator
characteristic curves are useful in making this decision.19
For example, screening for phenylketonuria in newborns
places a premium on sensitivity rather than on specificity;
the cost of missing a case is high, and effective treatment
exists. The downside is a large number of false-positive
tests, which cause anguish and further testing. By
contrast, screening for breast cancer should favour
specificity over sensitivity, since further assessment of
those tested positive entails costly and invasive biopsies