Mammography Screening Questioned

Mammography Screening Questioned

April 23, 2016



Lyon, France – A study by the International Prevention Research Institute (Lyon) and King’s College (London), published this week in PLOS ONE, confirms the likelihood of a substantial overestimation of the health benefits of mammography screening.

Breast cancer is the most common cancer in women. Its early diagnosis could clearly save lives. The key is to diagnose it before deadly metastases have spread to lymph-nodes or other organs. To this end, for decades, many countries have strongly recommended regular mammography screenings for women starting between 40 and 50 until they reach 69 to 74. However, results of population studies make it increasingly clear that breast cancer mortality did not decrease more rapidly in countries where mammography screening programmes started around 1990 than in countries where screening programmes were implemented ten to fifteen years later.

The recommendations to attend mammography screening is based on eight trials had estimated that in countries where at least 70% of women 40 to 74 would regularly attend mammography screenings, breast cancer mortality could drop by 20%. Confusingly, the results obtained by the eight trials were very diverse, from 31% mortality reduction in the Two-country trial in Sweden to no reduction in the two Canadian trials. This is more relevant now than ever as recent discussions regarding the possibly insufficient influence of mammography screening programmes on breast cancer mortality have intensified. One hypothesis for the variations is that the randomized trials ‘overestimated’ the ability of mammography screening to reduce the risk of breast cancer death.

According to one of the study’s authors, iPRI’s Professor Autier, “We evaluated whether in randomized trials on cancer screening, the observed reductions in cancer mortality published in articles was equivalent to reductions computed from reductions in the number of metastatic cancers due to screening.” The researchers found that the observed mortality reductions were much greater than reductions based on reductions of the numbers of metastatic cancers. “For instance,” says Professor Autier, “the Two-country trial published that mammography screenings could reduce the risk of breast cancer death by 31%. However, the risk reduction associated with the decrease in the number of metastatic cancer observed in this trial was only 11%.” Autier adds that ”we also applied the method on randomized trials on colorectal cancer screening and we obtained a good equivalence between observed reductions in the risk of colorectal cancer death and reductions in the risk of death calculated from decreases in metastatic cancers. This showed that in contrast to breast screening trials, the results of trials on colorectal screening were consistent with the principles of cancer screening.”

The study’s conclusion is that the poor effectiveness of mammography screening in populations can be explained by the fact that randomized trials overestimated the capacity of mammography screening to reduce the risk of breast cancer death. The inference is that mammography screening is associated with significant levels of over-diagnosis. 20% to 25% of breast cancers found in women could be over-diagnosed leading to unnecessary, potentially harmful, treatments.

Considering the potential over-diagnosis and ensuing over-treatment induced by mammography screening and its ineffective ability to reduce breast cancer mortality in general populations, there is an urgent need to search for efficient methods for the early detection of breast cancer.

# 30 #


Prof. Philippe Autier, MD, MPH, PhD

University of Strathclyde Institute of Global Public Health at iPRI

International Prevention Research Institute (iPRI)

Espace Européen d’Ecully, Bâtiment G

Allée Claude Debussy

69130 Ecully ouest Lyon



Tel:      +33 472 171 183 (direct)

+33 472 171 199 (switchboard)

+33 6 80 61 88 48 (cell phone)

Latest News

Latest Posts

Quick Contact

Please enter your name.
Please enter a message.