Current proposals for a monitoring and evaluation system in breast cancer screening programmes focus on mortality reduction. Here emphasis is laid on the prevention of too high a number of false-positive screening results, i.e. no subsequent demonstration of malignancy. By comparing the specificity of the screening test, the positive predictive value and the detection rate with reference values, the screening performance can be measured in a very early phase of the programme, even before the registration results on interval cancers become available. The proposed average reference values for the first screening round are 99.2%, 40% and 5.4/1000, respectively. Measures specifically for the age groups 45-49, 50-59 and 60-69 will be given, thus allowing improvements to be made if necessary.