Modelling the likely effect of the increase of the upper age limit from 70 to 73 for breast screening in the UK National Programme

Abstract

The UK National Breast Screening Programme is planned to have the age range for invitation to screening expanded from 50-70 to 47-73. At the upper limit, this represents one additional screen taking place in the early 70s. We aimed to estimate the likely effect of this on breast cancer mortality and on overdiagnosis of breast cancer. We used estimates of breast cancer incidence and survival by detection mode (screening or symptomatic), screening lead time and mortality from competing causes to estimate the likely numbers of breast cancer deaths prevented per 1000 women screened, using both a stochastic continuous time model and a semi-deterministic discrete time model. In the continuous model, we estimated that per 1000 women screened 1.2 deaths would be prevented and in the discrete model 0.91 deaths. In the latter model, we also estimated that there would be around 6.8 years of life saved per 1000 women screened and an additional two diagnoses of breast cancer. These results suggest that the expansion of the upper age limit will be cost-effective. They remain to be confirmed by evaluation of the age extension. They provide prior estimates that may inform the evaluation of the age extension.