ABSTRACT
This article presents an overview of the five randomized trials in Sweden in which the mortality of those invited to screening by mammography alone was compared with those not invited. The overview was based on 282 777 women followed for 5-13 years in randomized trials in Malmö, Kopparberg, Ostergotland, Stockholm and Gothenburg. Unlike meta-analyses, the overview was not restricted to published reports, but the individual cohorts were linked to regional cancer registries to identify cases of breast cancer and with the Swedish Cause of Death Register to ascertain fatal cases. These were reviewed independently by an ‘end-point committee’ which included an oncologist, surgeon, radiologist and pathologist who reached a consensus on determining whether breast cancer was the ‘underlying cause of death’ or ‘breast cancer present at death’. These two evaluations gave similar results, there being a consistent risk reduction in all studies which, in women of 50-69 years of age, was a significant 29%.
