MBS: a model for risk benefit analysis of breast cancer screening

Breast cancer screening programmes employing mammography are being implemented in various European countries. Different screening protocols are used in demonstration projects and nationwide programmes. To evaluate and improve protocols, a computer model for the evaluation of breast cancer screening has been developed. The availability of such a model can be of great importance in obtaining a better insight into the influence of various parameters. The Monte Carlo computer model is based on random selection from distributions of relevant parameters including tumour onset, tumour growth rate, lifetime expectancy, tumour detection size for screening and spontaneous observation. The radiation risk is calculated for various screening protocols employing multiplicative and additive risk models combined with lifetime expectancy, number of females screened and absorbed dose per screening session. The benefit is calculated on the basis of the reduction in tumour size at detection due to screening compared with spontaneous observation and the survival as a function of tumour diameter. Data from the Swedish two-county study are used to validate the model in terms of prevalence, interval tumour rates and interval tumour diameter distributions. Except for the spontaneous tumour diameter distribution, the model can describe the Swedish two-county study. Specific information is presented on the distributions of relevant parameters.