False-positive screening mammography and risk of incident breast cancer: systematic review and meta-analysis

Abstract False-positive (FP) mammography, usually considered a harm of screening, may be predictive of future breast cancer (BC) risk. This systematic review aimed to quantify the risk of developing BC after a FP relative to a negative result. We searched MEDLINE, Pre-MEDLINE, EMBASE and SCOPUS (01/01/2000-01/05/2026) for studies comparing future BC risk in FP versus negative screening mammography. Studies reporting comparable outcomes and measures of association were pooled by random effects meta-analysis. Risk of bias was assessed using the Quality in Prognostic Studies (QUIPS) tool. The protocol was registered with PROSPERO (CRD42024599252). Nine eligible retrospective studies (eight cohorts, total number of screened women 4,879,866) were identified. All reported statistically significant associations between FP mammography and increased BC risk with variable measures of association, adjustment for confounding, and measurement of risk factor and outcomes. The pooled adjusted hazard ratio (aHR) for overall BC risk (interval and subsequent round screen-detected BC) from three studies was 1·77 (95% CI: 1·59–1·97). Studies accounting for potential misclassification (removal of apparent FPs that were subsequently identified as true positives) reported attenuated, but still significant study-level associations. Similar pooled aHRs were found for interval BC (1·82, 95% CI: 1·27–2·59) and subsequent round screen-detected BC (1·83, 95% CI: 1·48–2·27) (two studies each, one of which also contributed to the pooled aHR for overall BC risk). FP mammography was consistently associated with increased incident BC risk compared with a negative mammogram (not recalled at screening), which is unlikely to be attributable entirely to misclassification. Further research is needed to understand the underlying mechanisms driving this association.