Measuring the Effectiveness of Mammography

Given the time, expense, and strong clinician and patient viewpoints, it is unlikely that any large-scale randomized clinical trial on mammography will ever be done again. Therefore, important clinical questions regarding the efficacy of mammography screening will need to be answered using observational data and modeling, even when those questions would really best be answered in randomized clinical trials. Using observational data from the Breast Cancer Surveillance Consortium including 15 440 breast cancers (3340 ductal carcinomas in situ [DCIS] and 12 100 invasive), Miglioretti et al¹ try to address the question of whether annual vs biennial screenings are superior. Although there is still controversy in the United States regarding this issue, most developed countries outside of the United States recommend screening every 2 or 3 years for eligible women, including the United Kingdom,² Canada,³ and Australia.⁴ The advantages of biennial screening from a population perspective are clear: reduced health care costs and decreased risk of false-positive screening results with the attendant extra biopsies and procedures.⁵ Furthermore, the improvements and widespread implementation of adjuvant breast cancer treatments have called into question how much of the recent decrease in breast cancer mortality is due to screening and how much to treatment advances, since these changes occurred contemporaneously.⁶ Nevertheless, concerns have arisen regarding cancers that would be missed with biennial screening.