Effect on sensitivity and specificity of mammography screening with or without comparison of old mammograms

Abstract

Mammography screening with old mammograms available for comparison decreased the false-positive recall rate. The effect on sensitivity, however, was unclear.

Do the results of the process indicators in the Norwegian breast cancer screening program predict future mortality reduction from breast cancer?

“…Nevertheless, The Netherlands has reported a very low recall rate (1.3% in the prevalent and 0.7% in the incident rounds) (17), while the Carolinas in the US have achieved a recall rate of 9.0% or higher (16). Radiologists' experiences and the women's risk profile are shown to be important for the recall rate (18,19) and determination of an optimal recall rate in a population-based service-screening program is difficult because of different ways of organizing the program (20). Yankaskas et al (16) claim a recall rate between 4.9% and 5.5% as optimal, viewed in light of the calculation of sensitivity and positive predictive value.…”

Should previous mammograms be digitised in the transition to digital mammography?

“…This may result in a reduction in performance during the transition period. The value of making comparisons with the previous mammograms has long been understood: both retrospective [5] and prospective studies using selected test sets [6][7][8][9][10] provide good evidence that the use of previous mammograms improves specificity. Therefore using previous mammograms in a population screening programme will reduce the number of women recalled unnecessarily, with no adverse effect on the cancer detection rate.…”

Follow-up and final results of the oslo I study comparing screen-film mammography and full-field digital mammography with soft-copy reading

“…Previous mammograms for comparison were not offered in the reading sessions in order to avoid interpretation bias, since previous SFM but no prior FFDM would have been available. Previous mammograms do not significantly increase sensitivity (cancer detection), although specificity is improved (3,20). In any case, the consequences of no prior mammograms for comparison should be equal for both modalities.…”