Does digital mammography suppose an advance in early diagnosis? Trends in performance indicators 6 years after digitalization

PURPOSE: To provide a complete evaluation of the long-term impact of full-field digital mammography (FFDM) on the improvement of early diagnosis in a population-based screening program. METHODS: We included 82,961 screen-film mammograms (SFM) and 79,031 FFDM from women aged 50–69 screened biennially from 1995–2010 in Spain and followed-up to 2012. The first screening round of the program was excluded. Rates of cancer detection, interval cancer, tumoral characteristics and other quality indicators were compared between SFM and FFDM periods using the Chi-square test. Multivariate logistic regression models were fitted. RESULTS: Detection of ductal carcinoma in situ (DCIS) significantly increased with FFDM (0.05 % vs 0.09 %; p = 0.010), along with the proportion of small invasive cancers (<20 mm) (69.37 % vs 78.90 %; p = 0.040). The false-positive rate decreased with FFDM (4.79 % vs 3.38 %; p < 0.001) without differences in the cancer detection rate (0.42 % vs 0.43 %; p = 0.685) or in the interval cancer rate (0.14 % vs 0.14 %; p = 0.816). Adjusted models showed a significant increase in the detection of DCIS in the FFDM periods. CONCLUSION: Digitalization has supposed an improvement in early diagnosis because DCIS and small invasive cancers increased without a change in detection rate. Moreover, false-positive reduction without an increase in the interval cancer rate was confirmed. KEY POINTS: • Cancer detection did not increase after 6 years of digital mammography • Ductal carcinoma in situ rates remained higher throughout the digital period • The proportion of small invasive cancers was higher with digital mammography • We observed an improvement in early diagnosis with digital mammography • False-positive rates remained lower throughout the digital period without interval cancer increase