Abstract
Early detection of breast cancer is an important public health policy. Programs of regular screening examinations have been widely established in an attempt to detect the disease when the primary tumour diameter is small. In South Australia, BreastScreen SA suggests that women between the ages of 50 and 70 years be screened every 24 months. Our aim in this paper is to make assessments of various screening procedures by using statistical models with parameters estimated exclusively from South Australian data. We establish a relationship between primary tumour diameter and ultimate survival time. We estimate an advantage of 2.9 years in median survival time for those women detected with the disease by BreastScreen SA, compared with an unscreened population. We construct a computer model from which we determine the consequences of using a 12 month screening interval.