A method is presented for making inferences about the difference betwe
en two binomial proportions when there is partial overlap of the two s
amples. The method will simplify to the usual cases of two independent
samples or paired samples (McNemar's method) depending on whether the
re is no or complete overlap. The method is used to compare sensitivit
ies, specificities and Youden indices from a comparative study of trau
ma triage (screening) instruments.