Spontaneous remissions in breast cancer underline the need of more evidence: screening should not detect more cancer but earlier cancer

Abstract

Aim: Analysis of the hypothesis that 22% of breast cancer cases detected by screening mammography would disappear spontaneously. Methods: Critical appraisal of the relevant scientific literature using established methods. Results: The hypothesis—although it is unusual—seems to be very well supported by data from several independent research groups and justifies three lessons. Conclusions: First, it is necessary to complete the last step of the concept of traditional evidence-based medicine (EBM), i.e., to confirm that the outcome that is predicted by scientific evidence will also be observed in daily clinical practice. Second, the focus of medical teaching should be directed to final clinical outcomes and patient benefit rather than surrogate parameters, such as completed procedures or test results. Third, the availability of data to answer unsolved questions may become a more important indicator for high-quality health-care systems than the performance of services of unknown incremental value.