Burden and Morbidity of Breast Screening and Diagnostic Work Up: Results from a Community Based Approach

The work in this thesis focused on addressing the limitations of breast screening and diagnostic work up using a cohort of women from a single large healthcare organization. For all three studies that were conducted, a probabilistic linkage between the examinations dataset (included all screening and diagnostic procedures) and the Illinois State Cancer Registry (included all incident breast cancer cases between 2001 and 2011) was performed to define the outcomes of interest. The first study was designed to compare the false positive rate (FPR), false positive burden (work-up rate including additional mammography, ultrasound or Magnetic resonance imaging (MRI) imaging) and false positive morbidity (risk of receipt of a biopsy) between full field digital mammography (FFDM) and screen-film mammography (SFM). Compared to SFM, FFDM had slightly lower FPR (12.8% vs 12.3%, P=0.02), lower FP burden (225 per 1000 FFDM screens vs 233 per 1000 SFM screens, P=0.001) but slightly higher risk of receipt of biopsy (1.24% vs 1.04%, P=0.05). The second study examined whether the experience of a FP mammogram delays the return to the subsequent round of screening. The results suggest that there was a compelling evidence that the women who received a true negative mammogram had significantly higher chance of returning to their next screening mammogram compared to women with a FP mammogram HR=1.35 (95% CI: 1.34-1.36). The third study estimated the risk of a false negative (FN) finding, defined as the percentage of cancers with a benign biopsy finding, and risk of complete false negative rate, defined as the percentage of cancers with a benign or high risk benign biopsy finding, for core-needle biopsy (CNB), vacuum-assisted ultrasound CNB and non-vacuum assisted ultrasound CNB. In this study, the FN rates for image-guided CNBs were 5.4% for stereotactic-guided, 4.8% for vacuum-assisted ultrasound-guided and 3.8% for non-vacuum-assisted ultrasound-guided biopsies. The complete FN rate was 11.3% for stereotactic-guided, 7% for vacuum-assisted ultrasound-guided and 5.1% for non-vacuum-assisted ultrasound-guided biopsies. Additionally, this study found an increased FN rate and complete FN rate in minority women, younger women and women with denser breasts.