69 Impact of full-field digital mammography versus film-screen mammography: systematic review

Objectives Most breast screening programs worldwide have replaced screen-film mammography (SFM) with full-field digital mammography (FFDM) in expectation of technical, clinical and economic advantages. However, we are only just now able to begin to measure the effects of this practice shift in population screening on health outcomes among asymptomatic women eligible for population screening. This systematic review aims to assess the impact of digital mammography on breast cancer detection rates at screening and on interval cancer rates, as indicators of additional net benefit through early detection, or additional net harm from overdiagnosis. Method We searched Medline, Premedline, PubMed, Embase, NHSEED, DARE and Cochrane databases and identified 2139 potentially eligible papers. 31 papers were included after exclusions for relevance, duplication and other exclusion criteria. Primary outcomes are detection rates and interval cancer rates. Secondary outcomes include recall rates, false positive rates, and positive predictive values. Results are stratified by first and subsequent screening rounds. Results Preliminary results for primary outcomes are available at the present time and reveal a small increase in screen detected cancers across all studies. However, in 7 studies with data on interval cancer rates, we observed no statistically significant increase in detection rate, nor a reduction in interval cancer rates. Final data for these primary outcomes, and for secondary outcomes, are being prepared and will be presented at the conference. Conclusions Overall there has been a small increase in screen-detected cancers with the transition from film to digital mammography screening. However we observed no reduction in interval cancers, and the effect, if any, remains unclear. This observed pattern of results is consistent with a possible, small increment in cancer detection which may result in future benefit for screened women, but is also consistent with a net increase in overdiagnosis. These data reinforce the need to carefully evaluate effects of future changes in technology such as 3D mammography to ensure incremental changes to screening programs do not lead to a poorer ratio of benefit to harm from screening.