The National Breast Screening Service in the UK depends upon triple assessm
ent, with or without core biopsy, to determine the need for diagnostic biop
sy or treatment for screen-detected lesions. To determine the efficacy of c
ytology in this process a nationwide survey of sensitivity and specificity
of the technique was instituted by the National Coordinating Committees for
both Radiology and Pathology. Although some units were performing to the n
ational recommended standards, a significant number of units were shown to
have problems with a high false negative rate and high inadequate rates fro
m lesions which were eventually diagnosed as cancer. This suggests problems
in localizing lesions adequately at aspiration. Generally, indicators of p
athology diagnostic performance (e.g. positive predictive values) were good
. The figures suggest that some units are not benefiting from an optimal cy
tological service. Consequently training programmes targeted at aspiration
technique may be beneficial. (C) 1999 Harcourt Publishers Ltd.