Objectives-To assess the impact of the National Health Service breast screening programme on overall and stage-specific incidence of breast cancer in East Anglia; also, to predict the magnitude of the screening induced reduction in breast cancer mortality. Setting-Women resident in East Anglia aged 50-69, diagnosed between 1976 and 1995 Methods-Comparison of numbers and incidence of breast cancer by age, stage, and mode of detection; investigation of relative contributions of advanced (stages II, III, and IV) cancers to total incidence by detection mode; estimation of the reduction in advanced cancer incidence. Results-There has been a large increase in early stage incidence in the age group 50-64 targeted by the screening programme. By 1995, the estimated decrease in advanced cancer incidence was between 7 and 19%. In 1995, of all breast cancers arising in the age group 50-69 years, 33% were screen detected, 27% were interval cancers, 15% were in non-attenders, 9% were in lapsed attenders, 7% occurred before invitation, and 4% arose in women outside the birth year range for invitation. Of the advanced cancers diagnosed in 1995, 31% were interval cancers, 20% were screen detected, 19% were in nonattenders, 12% were in lapsed attenders, 8% occurred before invitation, and 4% presented in women outside the birth year range for invitation. Conclusions-Screening has brought about a large increase in detection of early stage cancers. This increase has not get been fully matched by a corresponding deficit in advanced cancers, possibly because the full effect of screening has not yet been achieved. Reducing the proportion of interval cancers is necessary to increase the effect of screening on mortality.
