The impact of mammography on breast cancer detection

Background: Mammography has different effects on the epidemiology of breast cancer, i.e., while increasing the registered incidence it can reduce mortality. This study was aimed at obtaining quantitative estimates of the impact of mammography on breast cancer detection. Subjects and methods: A case-control study was conducted in Northern Italy between 1985 and 1991 on 2596 cases of histologically confirmed breast cancer and 2005 controls in hospital for acute diseases unrelated to known or suspected risk factors for breast cancer. Results: 15.6% of the patients reported one screening mammography and 9.7% two or more. The age-adjusted breast cancer diagnosis odds ratios (OR) were 1.3 (95% confidence interval, CI 1.1 to 1.5) for one mammography and 2.0 (95% CI 1.6 to 2.6) for two or more. When allowance was made for major identified potential distorting factors, the OR decreased to 1.0 (95% CI 0.8 to 1.2) for one, and 1.3 (95% CI 1.0 to 1.8) for two or more mammographies. In the age-adjusted analysis, the association between screening mammography and breast cancer detection was apparently stronger for younger and pre-menopausal women, among more educated women and those with no history of benign breast disease or breast biopsy, and among women who had first-degree relatives with histories of breast cancer. Conclusions: Although the association between mammography and breast cancer was largely explained by selective screening, confirming at the epidemiological level that the relationship is incidental, the unadjusted ratios are of public health interest, since they give a measure of the role of mammography in breast cancer detection.