The aim of this study was to evaluate changes in the mode of discovery of breast cancer in the last 15 years. We compared two periods separated by a 10-year interval, during which a mass mammographic screening programme was established in our department. We made a retrospective comparison of the records of female patients with breast cancer diagnosed in our hospital over the period 1986-1989 (first period) and 1997-1999 (second period). The mass screening programme for breast cancer began in 1995. We collected 372 patients in the first period and 341 in the second. We found a significant change in the mode of the discovery of breast cancer between the two periods: 80.2% versus 51.9%, respectively, of the cases of breast cancer were discovered by breast self-examination, 10.2% versus 13.7% were discovered by a physician, and 4.8% versus 29.1% were discovered by routine mammography as part of an individual or mass screening programme. The mean size of the tumours decreased significantly (2.6 cm versus 2.3 cm: p = 0.019), and the number of tumours with initial metastases or lymph node involvement decreased, almost attaining the level of significance (p = 0.06). It is difficult to compare the survival and disease-free survival curves because of the short follow-up in the second period (median follow-up = 10 months). However, a marked difference appears to be developing (p < 0.0001): patients diagnosed by mammography are showing better survival and disease-free survival compared with the others. We observed that more widespread use of mammography screening for breast cancer led to smaller tumours being discovered during the second period, with less lymph node involvement and less initial metastasis. Breast cancer screening is one of the most intensively evaluated health care practices with eight completed randomized trials yet its net benefit has remained controversial. It has been shown that, at least for patients aged 50 to 70, properly organized mass screening for breast cancer led to a reduction in mortality rate. However, individual breast self-exam, physician and mammographic screening can interfere with assessment of mass screening programmes in terms of individual benefit. In addition, introducing a mass screening programme may induce opportunistic screening in non-invited age groups and influence health behaviour in the target and non target populations. A retrospective study was performed to evaluate the mode of discovery, the diagnostic presentation, and prognostic factors in breast cancer in a French department before and after initiation of a mass-screening programme (MSP).