How often should women be referred for further testing in breast cancer screening? In the ROCS study, Daniëlle van der Waal (LRCB), Mireille Broeders (IQ Health), and colleagues analyzed national screening data to investigate how the referral rate correlates with cancer detection. Dutch screening radiologists are already close to the optimal referral rate on average: further increases yield hardly any additional detection, while the number of false positives continues to rise. These insights help in establishing balanced target values for screening.
In the breast cancer screening program, the referral rate—the percentage of women referred for further testing after screening—is an important quality indicator. A higher referral rate can lead to more breast cancer detection, but is also associated with more false-positive referrals, additional tests, and unnecessary anxiety among women. Finding the right balance is therefore essential.
In the ROCS study, Daniëlle van der Waal (LRCB), Mireille Broeders (IQ Health), and colleagues investigated how the referral rate relates to cancer detection within the Dutch breast cancer screening program. They used national screening data and a practice-oriented analysis method to clarify how changes in the referral rate relate to cancer detection.
The results show that an increasing referral rate is initially associated with an increase in cancer detection, but that this increase becomes smaller and smaller at higher referral rates. At the same time, the number of false-positive referrals continues to rise. Dutch screening radiologists appear to operate, on average, close to the optimal referral rate, just before the detection curve flattens: at that point, the referral rate is still relatively low, while cancer detection is already relatively high, and further increases yield only limited additional detection.
These findings show that referral practices in the Netherlands align well with a balanced ratio between cancer detection and false-positive referrals. The ROCS study clarifies the point at which additional referrals within the current Dutch breast cancer screening program yield only minimal additional detection. However, the optimal referral rate depends on the specific screening setting and will need to be re-evaluated in the event of changes, such as the introduction of artificial intelligence. The method used provides a practical tool for substantiating and evaluating target referral rates within various screening programs. In this way, the study offers important guidelines for quality assurance and further optimization of breast cancer screening, both in the Netherlands and beyond.
van der Waal D, Abbey CK, Tetteroo E, Geertse TD, Smid-Geirnaerdt MJA, Sechopoulos I, Broeders MJM. Finding the optimal recall rate in breast cancer screening: results from the ROCS study. Eur Radiol. 2026 Mar 5. doi: 10.1007/s00330-026-12370-5. Epub ahead of print. PMID: 41781728. 38,929 4.7 5.3