Predicting the risk of a false-positive test for women following a mammography screening programme

The objectives of this study was to provide a simple estimate of the cumulative risk of a false-positive test for women participating in mammography screening. To test the method, we used data from two well-established, organized mammography screening programmes offering biennial screening to women aged 50-69 years in Copenhagen and Fyn, Denmark.

We defined the outcome from a screen as being either a false-positive test or not a false-positive test. We then tested whether the outcomes from subsequent screens were independent, and afterwards estimated the risk over 10 screens of a false-positive test, i.e. the risk of getting at least one false-positive test for a woman participating in all 10 screens typically offered in Europe.

The outcomes of subsequent screens were found to be independent. After completion of screening rounds 3-5, the risk of a false-positive test over 10 screens was predicted to be 15.8-21.5% for a woman participating in the programme in Copenhagen, and 8.1-9.6% for a woman participating in the programme in Fyn.

Our study showed that a relatively robust prediction of the risk of a false-positive test over 10 screens can be calculated in a simple way relatively early after the start of a mammography screening programme.