[Skip to Navigation]
Sign In
Table 1.  Characteristics of 2 969 055 Screening Mammograms Included in the Study Population of 903 495 Women
Characteristics of 2 969 055 Screening Mammograms Included in the Study Population of 903 495 Women
Table 2.  Cumulative Probability of at Least 1 False-Positive Recall After 10 Years of Screening
Cumulative Probability of at Least 1 False-Positive Recall After 10 Years of Screening
Table 3.  Cumulative Probability of at Least 1 False-Positive Short-Interval Follow-up Recommendation After 10 Years of Screening
Cumulative Probability of at Least 1 False-Positive Short-Interval Follow-up Recommendation After 10 Years of Screening
Table 4.  Cumulative Probability of at Least 1 False-Positive Biopsy Recommendation After 10 Years of Screening
Cumulative Probability of at Least 1 False-Positive Biopsy Recommendation After 10 Years of Screening
1.
Brett  J, Bankhead  C, Henderson  B, Watson  E, Austoker  J.  The psychological impact of mammographic screening: a systematic review.   Psychooncology. 2005;14(11):917-938. doi:10.1002/pon.904 PubMedGoogle ScholarCrossref
2.
Brodersen  J, Siersma  VD.  Long-term psychosocial consequences of false-positive screening mammography.   Ann Fam Med. 2013;11(2):106-115. doi:10.1370/afm.1466 PubMedGoogle ScholarCrossref
3.
Chubak  J, Boudreau  DM, Fishman  PA, Elmore  JG.  Cost of breast-related care in the year following false positive screening mammograms.   Med Care. 2010;48(9):815-820. doi:10.1097/MLR.0b013e3181e57918 PubMedGoogle ScholarCrossref
4.
Mandelblatt  JS, Stout  NK, Schechter  CB,  et al.  Collaborative modeling of the benefits and harms associated with different US breast cancer screening strategies.   Ann Intern Med. 2016;164(4):215-225. doi:10.7326/M15-1536 PubMedGoogle ScholarCrossref
5.
Nelson  HD, O’Meara  ES, Kerlikowske  K, Balch  S, Miglioretti  D.  Factors associated with rates of false-positive and false-negative results from digital mammography screening: an analysis of registry data.   Ann Intern Med. 2016;164(4):226-235. doi:10.7326/M15-0971 PubMedGoogle ScholarCrossref
6.
Lehman  CD, Arao  RF, Sprague  BL,  et al.  National performance benchmarks for modern screening digital mammography: update from the Breast Cancer Surveillance Consortium.   Radiology. 2017;283(1):49-58. doi:10.1148/radiol.2016161174 PubMedGoogle ScholarCrossref
7.
Hubbard  RA, Kerlikowske  K, Flowers  CI, Yankaskas  BC, Zhu  W, Miglioretti  DL.  Cumulative probability of false-positive recall or biopsy recommendation after 10 years of screening mammography: a cohort study.   Ann Intern Med. 2011;155(8):481-492. doi:10.7326/0003-4819-155-8-201110180-00004 PubMedGoogle ScholarCrossref
8.
Lowry  KP, Trentham-Dietz  A, Schechter  CB,  et al.  Long-term outcomes and cost-effectiveness of breast cancer screening with digital breast tomosynthesis in the United States.   J Natl Cancer Inst. 2020;112(6):582-589. doi:10.1093/jnci/djz184PubMedGoogle ScholarCrossref
9.
Marinovich  ML, Hunter  KE, Macaskill  P, Houssami  N.  Breast cancer screening using tomosynthesis or mammography: a meta-analysis of cancer detection and recall.   J Natl Cancer Inst. 2018;110(9):942-949. doi:10.1093/jnci/djy121 PubMedGoogle ScholarCrossref
10.
Sumkin  JH, Ganott  MA, Chough  DM,  et al.  Recall rate reduction with tomosynthesis during baseline screening examinations: an assessment from a prospective trial.   Acad Radiol. 2015;22(12):1477-1482. doi:10.1016/j.acra.2015.08.015 PubMedGoogle ScholarCrossref
11.
Lowry  KP, Coley  RY, Miglioretti  DL,  et al.  Screening performance of digital breast tomosynthesis vs digital mammography in community practice by patient age, screening round, and breast density.   JAMA Netw Open. 2020;3(7):e2011792. doi:10.1001/jamanetworkopen.2020.11792 PubMedGoogle Scholar
12.
Trentham-Dietz  A, Kerlikowske  K, Stout  NK,  et al; Breast Cancer Surveillance Consortium and the Cancer Intervention and Surveillance Modeling Network.  Tailoring breast cancer screening intervals by breast density and risk for women aged 50 years or older: collaborative modeling of screening outcomes.   Ann Intern Med. 2016;165(10):700-712. doi:10.7326/M16-0476 PubMedGoogle ScholarCrossref
13.
van Buuren  S.  Multiple imputation of discrete and continuous data by fully conditional specification.   Stat Methods Med Res. 2007;16(3):219-242. doi:10.1177/0962280206074463 PubMedGoogle ScholarCrossref
14.
Braithwaite  D, Zhu  W, Hubbard  RA,  et al; Breast Cancer Surveillance Consortium.  Screening outcomes in older US women undergoing multiple mammograms in community practice: does interval, age, or comorbidity score affect tumor characteristics or false positive rates?   J Natl Cancer Inst. 2013;105(5):334-341. doi:10.1093/jnci/djs645 PubMedGoogle ScholarCrossref
15.
Dittus  K, Geller  B, Weaver  DL,  et al; Breast Cancer Surveillance Consortium.  Impact of mammography screening interval on breast cancer diagnosis by menopausal status and BMI.   J Gen Intern Med. 2013;28(11):1454-1462. doi:10.1007/s11606-013-2507-0 PubMedGoogle ScholarCrossref
16.
Kerlikowske  K, Zhu  W, Hubbard  RA,  et al; Breast Cancer Surveillance Consortium.  Outcomes of screening mammography by frequency, breast density, and postmenopausal hormone therapy.   JAMA Intern Med. 2013;173(9):807-816. doi:10.1001/jamainternmed.2013.307 PubMedGoogle ScholarCrossref
17.
O’Meara  ES, Zhu  W, Hubbard  RA,  et al.  Mammographic screening interval in relation to tumor characteristics and false-positive risk by race/ethnicity and age.   Cancer. 2013;119(22):3959-3967. doi:10.1002/cncr.28310 PubMedGoogle ScholarCrossref
18.
Breast Cancer Surveillance Consortium. BCSC registries. Accessed January 29, 2022. https://www.bcsc-research.org/
19.
D’Orsi  CJ, Mendelson  EB, Morris  EA,  et al.  ACR BI-RADS Atlas, Breast Imaging Reporting and Data System. American College of Radiology; 2013.
20.
White  IR, Royston  P, Wood  AM.  Multiple imputation using chained equations: Issues and guidance for practice.   Stat Med. 2011;30(4):377-399. doi:10.1002/sim.4067 PubMedGoogle ScholarCrossref
21.
Hubbard  RA, Miglioretti  DL, Smith  RA.  Modelling the cumulative risk of a false-positive screening test.   Stat Methods Med Res. 2010;19(5):429-449. doi:10.1177/0962280209359842 PubMedGoogle ScholarCrossref
22.
Hubbard  RA, Ripping  TM, Chubak  J, Broeders  MJ, Miglioretti  DL.  Statistical methods for estimating the cumulative risk of screening mammography outcomes.   Cancer Epidemiol Biomarkers Prev. 2016;25(3):513-520. doi:10.1158/1055-9965.EPI-15-0824 PubMedGoogle ScholarCrossref
23.
Hubbard  RA, Miglioretti  DL.  A semiparametric censoring bias model for estimating the cumulative risk of a false-positive screening test under dependent censoring.   Biometrics. 2013;69(1):245-253. doi:10.1111/j.1541-0420.2012.01831.x PubMedGoogle ScholarCrossref
24.
Rafferty  EA, Durand  MA, Conant  EF,  et al.  Breast cancer screening using tomosynthesis and digital mammography in dense and nondense breasts.   JAMA. 2016;315(16):1784-1786. doi:10.1001/jama.2016.1708 PubMedGoogle ScholarCrossref
25.
Pelletier  E, Daigle  JM, Defay  F, Major  D, Guertin  MH, Brisson  J.  Frequency and determinants of a short-interval follow-up recommendation after an abnormal screening mammogram.   Can Assoc Radiol J. 2016;67(4):322-329. doi:10.1016/j.carj.2015.11.001 PubMedGoogle ScholarCrossref
26.
Kirshenbaum  K, Harris  K, Harmon  J, Monge  J, Dabbous  F, Liu  Y.  BI-RADS 3 (short-interval follow-up) assessment rate at diagnostic mammography: correlation with recall rates and utilization as a performance benchmark.   Breast J. 2020;26(7):1284-1288. doi:10.1111/tbj.13838 PubMedGoogle ScholarCrossref
27.
Harris  RP, Sheridan  SL, Lewis  CL,  et al.  The harms of screening: a proposed taxonomy and application to lung cancer screening.   JAMA Intern Med. 2014;174(2):281-285. doi:10.1001/jamainternmed.2013.12745 PubMedGoogle ScholarCrossref
28.
Lowry  KP, Bell  S, Fendrick  AM, Carlos  RC.  Out-of-pocket costs of diagnostic breast imaging services after screening mammography among commercially insured women from 2010 to 2017.   JAMA Netw Open. 2021;4(8):e2121347. doi:10.1001/jamanetworkopen.2021.21347 PubMedGoogle Scholar
29.
Nelson  HD, Pappas  M, Cantor  A, Griffin  J, Daeges  M, Humphrey  L.  Harms of breast cancer screening: systematic review to update the 2009 US preventive services task force recommendation.   Ann Intern Med. 2016;164(4):256-267. doi:10.7326/M15-0970 PubMedGoogle ScholarCrossref
30.
Miglioretti  DL, Zhu  W, Kerlikowske  K,  et al; Breast Cancer Surveillance Consortium.  Breast tumor prognostic characteristics and biennial vs annual mammography, age, and menopausal status.   JAMA Oncol. 2015;1(8):1069-1077. doi:10.1001/jamaoncol.2015.3084 PubMedGoogle ScholarCrossref
31.
Kerlikowske  K, Grady  D, Rubin  SM, Sandrock  C, Ernster  VL.  Efficacy of screening mammography: a meta-analysis.   JAMA. 1995;273(2):149-154. doi:10.1001/jama.1995.03520260071035 PubMedGoogle ScholarCrossref
32.
Conant  EF, Barlow  WE, Herschorn  SD,  et al; Population-based Research Optimizing Screening Through Personalized Regimen (PROSPR) Consortium.  Association of digital breast tomosynthesis vs digital mammography with cancer detection and recall rates by age and breast density.   JAMA Oncol. 2019;5(5):635-642. doi:10.1001/jamaoncol.2018.7078 PubMedGoogle ScholarCrossref
33.
Conant  EF, Zuckerman  SP, McDonald  ES,  et al.  Five consecutive years of screening with digital breast tomosynthesis: outcomes by screening year and round.   Radiology. 2020;295(2):285-293. doi:10.1148/radiol.2020191751 PubMedGoogle ScholarCrossref
34.
Bernardi  D, Macaskill  P, Pellegrini  M,  et al.  Breast cancer screening with tomosynthesis (3D mammography) with acquired or synthetic 2D mammography compared with 2D mammography alone (STORM-2): a population-based prospective study.   Lancet Oncol. 2016;17(8):1105-1113. doi:10.1016/S1470-2045(16)30101-2 PubMedGoogle ScholarCrossref
35.
Kleinknecht  JH, Ciurea  AI, Ciortea  CA.  Pros and cons for breast cancer screening with tomosynthesis: a review of the literature.   Med Pharm Rep. 2020;93(4):335-341. doi:10.15386/mpr-1698 PubMedGoogle Scholar
36.
Miglioretti  DL, Abraham  L, Lee  CI,  et al; Breast Cancer Surveillance Consortium.  Digital breast tomosynthesis: radiologist learning curve.   Radiology. 2019;291(1):34-42. doi:10.1148/radiol.2019182305 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Oncology
    March 25, 2022

    Cumulative Probability of False-Positive Results After 10 Years of Screening With Digital Breast Tomosynthesis vs Digital Mammography

    Author Affiliations
    • 1Division of Biostatistics, Department of Public Health Sciences, University of California Davis School of Medicine, Davis
    • 2Department of Training and Scientific Research, University Medical Center, Ho Chi Minh City, Vietnam
    • 3General Internal Medicine Section, Department of Veterans Affairs, University of California, San Francisco
    • 4Department of Medicine, University of California, San Francisco
    • 5Department of Epidemiology and Biostatistics, University of California, San Francisco
    • 6Department of Biostatistics, Epidemiology & Informatics, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania
    • 7Department of Surgery, Office of Health Promotion Research, Larner College of Medicine at the University of Vermont and University of Vermont Cancer Center, Burlington, Vermont
    • 8Department of Radiology, University of Washington School of Medicine, Seattle
    • 9Department of Health Systems and Population Health, University of Washington School of Public Health, Seattle
    • 10Hutchinson Institute for Cancer Outcomes Research, Seattle, Washington
    • 11Division of General Internal Medicine, Department of Medicine, University of California, San Francisco
    • 12The Dartmouth Institute for Health Policy and Clinical Practice, Dartmouth College, Lebanon, New Hampshire
    • 13Department of Medicine, Geisel School of Medicine at Dartmouth, Hanover, Lebanon, New Hampshire
    • 14Department of Oncology, Norris Cotton Cancer Center, Lebanon, New Hampshire
    • 15Kaiser Permanente Washington Health Research Institute, Kaiser Permanente Washington, Seattle
    JAMA Netw Open. 2022;5(3):e222440. doi:10.1001/jamanetworkopen.2022.2440
    Key Points

    Question  Is there a difference between screening with digital breast tomosynthesis vs digital mammography in the probability of false-positive results after 10 years of screening?

    Findings  In this comparative effectiveness study of 903 495 individuals undergoing 2 969 055 screening examinations, the 10-year cumulative probability of receiving at least 1 false-positive recall was 6.7% lower for tomosynthesis vs digital mammography with annual screening and 2.4% lower for tomosynthesis vs digital mammography with biennial screening, a significant difference.

    Meaning  The findings of this study suggest that digital breast tomosynthesis is associated with a lower cumulative probability of false-positive results compared with digital mammography; biennial vs annual screening was associated with larger reductions in cumulative false-positive risk for both modalities.

    Abstract

    Importance  Breast cancer screening with digital breast tomosynthesis may decrease false-positive results compared with digital mammography.

    Objective  To estimate the probability of receiving at least 1 false-positive result after 10 years of screening with digital breast tomosynthesis vs digital mammography in the US.

    Design, Setting, and Participants  An observational comparative effectiveness study with data collected prospectively for screening examinations was performed between January 1, 2005, and December 31, 2018, at 126 radiology facilities in the Breast Cancer Surveillance Consortium. Analysis included 903 495 individuals aged 40 to 79 years. Data analysis was conducted from February 9 to September 7, 2021.

    Exposures  Screening modality, screening interval, age, and Breast Imaging Reporting and Data System breast density.

    Main Outcomes and Measures  Cumulative risk of at least 1 false-positive recall for further imaging, short-interval follow-up recommendation, and biopsy recommendation after 10 years of annual or biennial screening with digital breast tomosynthesis vs digital mammography, accounting for competing risks of breast cancer diagnosis and death.

    Results  In this study of 903 495 women, 2 969 055 nonbaseline screening examinations were performed with interpretation by 699 radiologists. Mean (SD) age of the women at the time of the screening examinations was 57.6 (9.9) years, and 58% of the examinations were in individuals younger than 60 years and 46% were performed in women with dense breasts. A total of 15% of examinations used tomosynthesis. For annual screening, the 10-year cumulative probability of at least 1 false-positive result was significantly lower with tomosynthesis vs digital mammography for all outcomes: 49.6% vs 56.3% (difference, –6.7; 95% CI, –7.4 to –6.1) for recall, 16.6% vs 17.8% (difference, –1.1; 95% CI, –1.7 to –0.6) for short-interval follow-up recommendation, and 11.2% vs 11.7% (difference, –0.5; 95% CI, –1.0 to –0.1) for biopsy recommendation. For biennial screening, the cumulative probability of a false-positive recall was significantly lower for tomosynthesis vs digital mammography (35.7% vs 38.1%; difference, –2.4; 95% CI, –3.4 to –1.5), but cumulative probabilities did not differ significantly by modality for short-interval follow-up recommendation (10.3% vs 10.5%; difference, –0.1; 95% CI, –0.7 to 0.5) or biopsy recommendation (6.6% vs 6.7%; difference, –0.1; 95% CI, –0.5 to 0.4). Decreases in cumulative probabilities of false-positive results with tomosynthesis vs digital mammography were largest for annual screening in women with nondense breasts (differences for recall, –6.5 to –12.8; short-interval follow-up, 0.1 to –5.2; and biopsy recommendation, –0.5 to –3.1). Regardless of modality, cumulative probabilities of false-positive results were substantially lower for biennial vs annual screening (overall recall, 35.7 to 38.1 vs 49.6 to 56.3; short-interval follow-up, 10.3 to 10.5 vs 16.6 to 17.8; and biopsy recommendation, 6.6 to 6.7 vs 11.2 to 11.7); older vs younger age groups (eg, among annual screening in women ages 70-79 vs 40-49, recall, 39.8 to 47.0 vs 60.8 to 68.0; short-interval follow-up, 13.3 to 14.2 vs 20.7 to 20.9; and biopsy recommendation, 9.1 to 9.3 vs 13.2 to 13.4); and women with entirely fatty vs extremely dense breasts (eg, among annual screening in women aged 50-59 years, recall, 29.1 to 36.3 vs 58.8 to 60.4; short-interval follow-up, 8.9 to 11.6 vs 19.5 to 19.8; and biopsy recommendation, 4.9 to 8.0 vs 15.1 to 15.3).

    Conclusions and Relevance  In this comparative effectiveness study, 10-year cumulative probabilities of false-positive results were lower on digital breast tomosynthesis vs digital mammography. Biennial screening interval, older age, and nondense breasts were associated with larger reductions in false-positive probabilities than screening modality.

    Introduction

    Early breast cancer detection via screening mammography is a key strategy to decrease breast cancer morbidity and mortality; however, mammography results in notable harms, including false-positive results, that lead to unnecessary additional imaging and biopsy procedures, financial and opportunity costs, and patient anxiety.1-5 False-positive results are common, with 12% of patients who undergo digital screening mammograms recalled for additional workup; of those recalls, only 4.4%, or 0.53% of screening mammograms overall, result in a cancer diagnosis.6 A Breast Cancer Surveillance Consortium (BCSC) study including mostly film mammography estimated that after 10 years of annual screening in women aged 40 to 59 years, including their baseline mammogram, 61% of individuals would experience at least 1 false-positive recall and 7% to 9% at least 1 false-positive biopsy recommendation.7

    Digital breast tomosynthesis is rapidly disseminating in the US, with lower recall and false-positive rates compared with screening with digital mammography.8-10 In prior BCSC research involving some of the authors of the present study, reductions in recall rates with tomosynthesis vs digital mammography were found across all Breast Imaging Reporting and Data System (BI-RADS) breast density categories except for extremely dense breasts.11 However, effective screening requires many examinations over multiple decades. Multimodel simulation studies4,12,13 and BCSC studies7,14-17 consistently reported that repeated biennial screening was associated with lower false-positive rates than annual screening. The goal of this study was to estimate the cumulative probabilities of at least 1 false-positive recall, short-interval follow-up recommendation, and biopsy recommendation during 10 years of annual vs biennial subsequent screening with digital breast tomosynthesis vs digital mammography by decade of age and BI-RADS breast density category using longitudinal data from the BCSC.

    Methods
    Study Setting, Data Sources, and Participants

    In this observational comparative effectiveness study with prospective data collection, we selected digital mammography and digital breast tomosynthesis (ie, 3-dimensional mammography) screening examinations performed between January 1, 2005, and December 31, 2018, among women aged 40 to 79 years at 126 breast imaging facilities participating in 6 BCSC registries18: San Francisco Mammography Registry, New Hampshire Mammography Network, Vermont Breast Cancer Surveillance System, Carolina Mammography Registry, Metro Chicago Breast Cancer Registry, and Sacramento Area Breast Imaging Registry. Screening examinations were defined by the clinical indication. We excluded baseline examinations, unilateral examinations, examinations with a mammogram within the last 9 months, and examinations in women with a history of breast cancer or mastectomy. We restricted our focus to subsequent examinations for comparability across age groups, as baseline examinations tend to be concentrated in younger women and have higher false-positive rates than subsequent examinations.7 The final study cohort included 2 969 055 screening examinations among 903 495 women.

    Breast Cancer Surveillance Consortium registries and the statistical coordinating center received institutional review board approval from the University of California, Davis; University of California, San Francisco; University of North Carolina at Chapel Hill; University of Illinois, Chicago; University of Vermont; Advocate Health Care; Dartmouth; and Kaiser Permanente Washington for active or passive consenting processes or a waiver of consent to enroll participants, link and pool data, and perform analysis. Procedures were Health Insurance Portability and Accountability Act compliant, and registries and the coordinating center received a Federal Certificate of Confidentiality and other protections for the identities of women, physicians, and facilities. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies, considering the adaptations recommended by the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) reporting guideline for comparative effectiveness studies.

    Measures, Definitions, and Outcomes

    At each examination, self-reported information on age and time since last mammogram was obtained from questionnaires. Radiologists (N = 699) reported assessments following American College of Radiology BI-RADS terminology and recorded BI-RADS breast density as part of the clinical interpretation as almost entirely fatty, scattered fibroglandular densities, heterogeneously dense, or extremely dense.19 Screening interval was defined using BCSC data and self-reported information to determine the date of the previous mammogram and categorized as annual (9-18 months), biennial (19-30 months), or triennial or longer (>30 months). Screening round was based on the number of prior subsequent examinations observed for each woman.

    Primary outcomes were false-positive recall, false-positive short-interval follow-up recommendation, and false-positive biopsy recommendation. Recall was defined as a BI-RADS initial assessment of 0 (needs additional imaging evaluation), 3 (probably benign finding), 4 (suspicious abnormality), or 5 (highly suggestive of cancer). Short-interval follow-up recommendation was defined as a BI-RADS final assessment of 3 after diagnostic imaging workup within 90 days of a recalled screening examination. Biopsy recommendation was defined as a BI-RADS final assessment of 4 or 5. Recalls, short-interval follow-up recommendations, and biopsy recommendations were considered false-positive if no diagnosis of invasive carcinoma or ductal carcinoma in situ occurred within 1 year of the screening examination and before the next screening examination. We imputed false-positive short-interval follow-up and biopsy recommendations for examinations with a false-positive recall but unresolved final assessment (n = 14 171 [0.5%]) based on age, breast density, screening modality, and screening interval, imputing a single value because less than 1% of data was missing.20 Diagnoses of invasive breast carcinoma and ductal carcinoma in situ were obtained by linkage to pathology databases and state or regional tumor registries. Death information was obtained by linkage with state death records.

    Statistical Analysis

    Data analysis was performed from February 9 to September 7, 2021. We summarized cohort characteristics by screening modality. We estimated unadjusted percentages of examinations with false-positive results by age group, breast density, screening interval, and modality. Cumulative probabilities of at least 1 false-positive result after 10 years of annual or biennial screening were estimated using a discrete-time survival model to account for censoring.21-23 Logistic regression was used to estimate the probability of each outcome after a single screening mammogram as a function of age (linear and quadratic), breast density, screening interval, modality, and all interactions between these variables; screening round and its interaction with modality; and censoring round, defined as the total number of screening examinations observed for a woman.21-23 We excluded screening mammograms after the first false-positive result being modeled to estimate the probability of a first false-positive result given no earlier false-positive results. Standardized 10-year cumulative probabilities of at least 1 false-positive result after 10 annual and 5 biennial screening rounds were estimated from round-specific probabilities by marginalizing over the distribution of covariate combinations and censoring round, increasing age in 1- or 2-year increments for annual or biennial examinations, increasing screening round in 1-unit increments, and holding screening interval, screening modality, density, and censoring round constant.21-23 Cumulative probabilities were adjusted for competing risks of breast cancer diagnoses or death from any cause within 1 year of an annual or 2 years of a biennial screening mammogram and estimated with a logistic regression model including age, breast density, screening interval, modality, and all interactions.23 In addition, 95% CIs were estimated using a nonparametric bootstrap with 10 000 iterations.

    Statistical analyses were performed using SAS/STAT software, version 14.2 (SAS Institute Inc), R, version 4.0.2 (R Foundation for Statistical Computing), and RStudio, version 1.3.1056 (RStudio Inc). Tests of statistical significance used a 2-sided α = .05.

    Results

    Our study included 444 704 digital breast tomosynthesis and 2 524 351 digital mammography examinations among 903 495 women; mean (SD) age at the time of screening was 57.6 (9.9) years; 58% of the examinations were in individuals younger than 60 years, and 46% were performed in women with dense breasts (Table 1). Women underwent a mean (SD) of 3.3 (2.5) examinations. A total of 2 132 274 examinations (71.8%) were annual screening mammograms, 497 829 (16.8%) were biennial screening mammograms, and 338 952 (11.4%) were triennial or longer screening mammograms. Tomosynthesis tended to be used at later screening rounds than digital mammography, with 148 728 tomosynthesis examinations (33.4%) vs 150 606 digital examinations (6.0%) performed in round 7 or later. Overall, 33 760 tomosynthesis (7.6%) and 227 485 digital mammograms (9.0%) resulted in a false-positive recall, 7865 tomosynthesis (1.8%) and 52 236 digital mammograms (2.1%) resulted in a false-positive short-interval follow-up recommendation, and 4893 tomosynthesis (1.1%) and 30 058 digital mammograms (1.2%) resulted in a false-positive biopsy recommendation (Table 1). The distribution of age group and breast density by screening interval and modality are provided in eTable 1 in the Supplement.

    False-Positive Recall

    The unadjusted percentage of mammograms with a false-positive recall on tomosynthesis vs digital mammography were 6.8% (95% CI, 6.7%-6.9%) vs 8.2% (95% CI, 8.2%-8.3%; difference, –1.4; 95% CI, –1.5 to –1.3) for annual examinations and 8.5% (95% CI, 8.3%-8.7%) vs 9.5% (95% CI, 9.4%-9.6%; difference, –1.0; 95% CI, –1.2 to –0.7) for biennial examinations and were generally lower for women with almost entirely fatty breasts and for older vs younger women (eTable 2 in the Supplement). Comparing modalities, the overall cumulative probability of at least 1 false-positive recall after 10 years of screening with tomosynthesis vs digital mammography was 49.6% (95% CI, 49.0%-50.2%) vs 56.3% (95% CI, 56.0%-56.7%) for annual screening (difference, –6.7; 95% CI, –7.4 to –6.1) and 35.7% (95% CI, 34.8%-36.6%) vs 38.1% (95% CI, 37.8%-38.5%) for biennial screening (difference, –2.4; 95% CI, –3.4 to –1.5) (Table 2). Comparing biennial vs annual screening, the overall cumulative probability was lower with tomosynthesis (difference, –13.9; 95% CI, –14.9 to –12.8) and digital mammography (difference, –18.2; 95% CI, –18.6 to –17.7). Cumulative false-positive recall probabilities generally declined with increasing age (eg, 39.8 to 47.0 vs 60.8 to 68.0 among annual screening in women ages 70-79 vs 40-49 years) and decreasing breast density. For example, among women aged 40 to 49 years, the estimated probabilities of a false-positive recall after 10 years of annual screening with tomosynthesis were 31.0% (95% CI, 25.6%-36.7%) for women with entirely fatty breasts and 67.3% (95% CI, 63.8%-70.7%) for women with extremely dense breasts. Decreases in the cumulative false-positive recall probability with tomosynthesis compared with digital mammography were largest for annual screening in women with nondense breasts (differences −6.5 to −12.8). Cumulative false-positive recall probabilities among women with extremely dense breasts tended to be higher, but not significantly different, for tomosynthesis vs digital mammography regardless of screening interval (eg, annual screening age 40-49 years: 67.3%; 95% CI, 63.8%-70.7% vs 65.0%; 95% CI, 63.6%-66.4%; difference, 2.3; 95% CI, –1.4 to 6.0; biennial screening age 40-49 years: 51.2%; 95% CI, 45.7%-56.9% vs 46.1%; 95% CI, 44.1%-48.0%; difference, 5.2; 95% CI, –0.6 to 11.1).

    False-Positive Short-Interval Follow-up Recommendations

    The unadjusted percentage of examinations with a false-positive short-interval follow-up recommendation on tomosynthesis vs digital mammography were 1.5% (95% CI, 1.4%-1.5%) vs 1.8% (95% CI, 1.8%-1.8%; difference, –0.3; 95% CI, –0.3 to –0.3) for annual examinations and 2.1% (95% CI, 2.0%-2.2%) vs 2.2% (95% CI, 2.2%-2.3%; difference, –0.2; 95% CI, –0.3 to 0.0) for biennial examinations and were generally lower for women with almost entirely fatty breasts and for older vs younger women (eTable 3 in the Supplement). Comparing modalities, the overall cumulative probability of at least 1 false-positive short-interval follow-up recommendation after 10 years of tomosynthesis vs digital mammography screening was 16.6% (95% CI, 16.1%-17.1%) vs 17.8% (95% CI, 17.4%-18.2%) for annual (difference, –1.1; 95% CI, –1.7 to –0.6) and 10.3% (95% CI, 9.8%-10.9%) vs 10.5% (95% CI, 10.2%-10.7%) for biennial screening (difference, –0.1; 95% CI, –0.7 to 0.5) (Table 3). Comparing biennial vs annual screening, the overall cumulative probability was lower with tomosynthesis (difference, –6.3; 95% CI, –7.0 to –5.6) and digital mammography (difference, –7.3; 95% CI, –7.7 to –6.9). In general, the cumulative probability of a false-positive short-interval follow-up recommendation were lower with increasing age (eg, 13.3 to 14.2 vs 20.7 to 20.9 among annual screening in women ages 70-79 vs 40-49 years) and decreasing breast density. Decreases with tomosynthesis were largest for annual screening in women with nondense breasts (differences 0.1 to −5.2).

    False-Positive Biopsy Recommendations

    The unadjusted percentages of examinations with a false-positive biopsy recommendation on tomosynthesis vs digital mammography were 0.9% (95% CI, 0.9%-1.0%) vs 1.0% (95% CI, 1.0%-1.0%; difference, –0.1; 95% CI, –0.1 to 0.0) for annual examinations and 1.3% (95% CI, 1.2%-1.4%) vs 1.3% (95% CI, 1.3%-1.4%; difference, –0.1; 95% CI, –0.2 to 0.0) for biennial examinations and were generally lower for women with almost entirely fatty breasts and for older vs younger women (eTable 4 in the Supplement). Comparing modalities, the overall cumulative probability of at least 1 false-positive biopsy recommendation after 10 years of tomosynthesis vs digital mammography screening was 11.2% (95% CI, 10.7%-11.7%) vs 11.7% (95% CI, 11.4%-12.1%) for annual screening (difference, –0.5; 95% CI, –1.0 to –0.1) and 6.6% (95% CI, 6.2%-7.1%) vs 6.7% (95% CI, 6.5%-6.9%) for biennial screening (difference, –0.1; 95% CI, –0.5 to 0.4) (Table 4). Comparing biennial vs annual screening, the overall cumulative probability was lower with tomosynthesis (difference, –4.6; 95% CI, –5.2 to –3.9) and digital mammography (difference, –5.0; 95% CI, –5.4 to –4.7). In general, cumulative false-positive biopsy recommendation probabilities were lower with increasing age (eg, 9.1 to 9.3 vs 13.2 to 13.4 among annual screening in women ages 70-79 vs 40-49 years) and decreasing breast density. Decreases in the cumulative probability of false-positive biopsy recommendation were largest for annual screening in women with nondense breasts (differences −0.5 to −3.1).

    Discussion

    We estimated the cumulative probabilities of at least 1 false-positive result after 10 years of annual or biennial screening with digital breast tomosynthesis vs digital mammography. We investigated 3 types of false-positive results occurring throughout the screening episode: recall for additional imaging, recommendation for short-interval follow-up, and recommendation for biopsy. The cumulative probability of false-positive result was lower for recall for tomosynthesis vs digital mammography with both annual and biennial screening, lower for short-interval follow-up recommendation with annual screening but not biennial screening, and similar for biopsy recommendation regardless of screening interval. All 3 types of false-positive results depended on age and breast density and were substantially lower for biennial vs annual mammography regardless of screening modality.

    Prior research suggested that digital breast tomosynthesis is associated with reductions in false-positive recalls compared with digital mammography.8-10 We found that reductions in the percentage of individuals receiving at least 1 false-positive recall were modest after 10 years of subsequent screening with tomosynthesis, with reductions of 2.4% for biennial screening and 6.7% for annual screening. Nonetheless, this percentage equates to many thousands of individuals in absolute numbers, especially for annual screening, which is the dominant practice in the US.

    The cumulative probability of receiving a false-positive recall under annual screening remained high with tomosynthesis, with almost half of women projected to experience at least 1 false-positive recall after 10 subsequent screens. Cumulative false-positive probabilities would be even higher if baseline examinations were considered, at which one-fifth of women are recalled for additional workup.11 As in earlier BCSC studies that included both film and digital mammography,7,14-17 we report substantial reductions in cumulative false-positive probabilities with biennial compared with annual screening, with 36% to 38% of women expected to experience at least 1 false-positive recall after 10 years of subsequent biennial screening. This reduction occurs because women undergo half as many screening examinations with biennial screening, although false-positive recalls are more common on biennial vs annual examinations. We also found large reductions with increasing age and decreasing breast density. Thus, screening interval, age, and breast density were associated with larger reductions in the cumulative probability of a false-positive recall after repeat screening than screening modality.

    To our knowledge, this study is the first to estimate the cumulative probability of a false-positive short-interval follow-up recommendation after 10 years of subsequent screening, with approximately 17% of women expected to experience at least 1 short-interval follow-up recommendation under annual screening compared with 10% under biennial screening. These probabilities were only slightly lower with tomosynthesis than digital mammography. We also found the cumulative probability of at least 1 false-positive biopsy recommendation was generally similar for tomosynthesis vs digital mammography, with 1 in 9 women projected to receive a benign biopsy result after 10 years of annual screening regardless of modality. Our results showed large reductions in the cumulative false-positive biopsy recommendation probability with biennial vs annual screening and large increases with increasing breast density, similar to those reported by Kerlikowske et al16 in a BCSC study of mostly film mammography.

    In general, women with almost entirely fatty breasts had the lowest false-positive probabilities regardless of false-positive type and screening modality and the largest reductions in cumulative false-positive probabilities with digital breast tomosynthesis vs digital mammography. In contrast, women with extremely dense breasts tended to have the highest probability of all 3 types of false-positive results and these probabilities were not significantly lower with tomosynthesis compared with digital mammography. This lack of difference in cumulative probabilities of false-positives by modality may be due to the lack of interspersed fat within dense fibroglandular tissue, with the contrast between the fat and tissue being a requirement for more accurate detection of suspicious features by interpreting radiologists. Our results are consistent with a prior BCSC study, which found that, on subsequent screening mammograms, women with extremely dense breasts did not benefit from improved recall or cancer detection with tomosynthesis.11 A US multi-institutional study reported improved recall rates in women with extremely dense breasts; however, the study included baseline mammograms and did not account for screening round.24

    Previous research suggests that women receiving a false-positive recommendation for additional imaging or biopsy may experience elevated anxiety and distress, even if transient.1,2,25,26 We consider false-positive short-interval follow-up as a potential harm of screening because these assessments require women to return for additional diagnostic imaging 6 months after the screening mammogram, delaying receipt of the final result and potentially resulting in additional radiation exposure and pain from mammography, psychological effects, financial strain from copays and other out-of-pocket medical costs and lost work time, and opportunity costs.27,28 However, short-interval follow-up assessments may prevent unnecessary biopsy that poses additional harms from infection, pain, and scarring.29 The relative frequency and severity of these 3 types of false-positive results should be considered in evaluating the harms of screening mammograms.

    Our study offers new information about the potential harms of repeated screening, which may be used to inform screening guidelines and decision-making between individuals and their physicians; however, it is important to weigh these and other potential harms with potential benefits of earlier diagnosis. Prior research shows the benefits of annual and biennial screening are similar for most women4,7,30,31; however, women at high risk of an advanced cancer under biennial screening, including some women with dense breasts, may reduce their risk with annual screening.4,12,14,16,30 Other research has found that any improvements in cancer detection are small for tomosynthesis vs digital mammography.11,32-34 Some biennial screening programs have reported larger improvements in cancer detection rates34; however, long-term follow-up data are needed to assess changes for nonbaseline examinations.35

    Limitations

    This study has limitations. We excluded baseline mammograms from our analysis so that we could compare results across all age groups. Including baseline mammograms would have increased the cumulative probabilities because false-positive results are more common on baseline mammograms.7 We did not estimate lifetime cumulative probabilities because doing so would require extrapolation beyond the length of observation. Even with the large BCSC cohort, sample sizes were small for tomosynthesis for women with fatty and extremely dense breasts. Tomosynthesis tended to be used at later screening rounds than digital mammography because tomosynthesis diffused into clinical practice during the study period. Given the probability of false-positive results tending to decrease with screening round, adjusting for screening round as a confounder may be important for other studies, as we did in this study. Our estimates did not take into account any potential changes by calendar year, although we expect changes in false-positive rates with the diffusion of tomosynthesis to be minimal given that prior research found that early reductions in recall rates after tomosynthesis adoption were sustained.36 False-positive rates likely vary across facilities. Our estimates reflect the population average across 126 diverse BCSC facilities.

    Conclusions

    This study noted somewhat lower cumulative probabilities of false-positive recall for digital breast tomosynthesis vs digital mammography after 10 years of annual screening, with smaller differences among women who underwent biennial screening. We did not observe consistent clinically meaningful differences in the cumulative probabilities of false-positive short-interval follow-up or biopsy recommendation by screening modality. Biennial screening interval, older age, and nondense breasts were associated with larger reductions in false-positive results than screening modality.

    Back to top
    Article Information

    Accepted for Publication: December 23, 2021.

    Published: March 25, 2022. doi:10.1001/jamanetworkopen.2022.2440

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Ho TQH et al. JAMA Network Open.

    Corresponding Author: Diana L. Miglioretti, PhD, University of California, Davis, Division of Biostatistics, Department of Public Health Sciences, One Shields Ave, Med Sci 1C, 116, Davis, CA 95616 (dmiglioretti@ucdavis.edu).

    Author Contributions: Drs Miglioretti and Bissell had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Ho and Bissell contributed equally to this work.

    Concept and design: Ho, Bissell, Kerlikowske, Hubbard, Sprague, Lee, Miglioretti.

    Acquisition, analysis, or interpretation of data: Ho, Bissell, Kerlikowske, Sprague, Lee, Tice, Tosteson, Miglioretti.

    Drafting of the manuscript: Ho, Bissell, Lee, Miglioretti.

    Critical revision of the manuscript for important intellectual content: Ho, Bissell, Kerlikowske, Hubbard, Sprague, Lee, Tice, Tosteson.

    Statistical analysis: Bissell, Hubbard, Miglioretti.

    Obtained funding: Kerlikowske, Sprague, Lee, Tosteson, Miglioretti.

    Administrative, technical, or material support: Kerlikowske.

    Supervision: Ho, Bissell, Miglioretti.

    Conflict of Interest Disclosures: Dr Kerlikowske is an unpaid consultant for Grail Inc, for the STRIVE study. Dr Lee receives personal fees from Grail Inc, for work on a data safety monitoring board, personal fees from the American College of Radiology for editorial board work, and textbook royalties from McGraw Hill Inc, Oxford University Press, and Wolters Kluwer, all outside the submitted work. No other disclosures were reported.

    Funding/Support: The research reported in this work was funded by the National Cancer Institute (NCI) (grant P01CA154292). The Breast Cancer Surveillance Consortium (BCSC) additionally supported data collection for this research with funding from the NCI (grant U54CA163303), the Agency for Healthcare Research and Quality (grant R01 HS018366-01A1), the University of Vermont Cancer Center with funds generously awarded by the Lake Champlain Cancer Research Organization (grant 032800), and the Patient-Centered Outcomes Research Institute (PCORI) (award PCS-1504-30370). The collection of Sacramento Area Breast Imaging Registry data was supported by the UC Davis Comprehensive Cancer Center, the Placer County Breast Cancer Foundation, and the UC Davis Clinical and Translational Science Center. Cancer and vital status data collection was supported by several state public health departments and cancer registries involved in this study. The collection of cancer incidence data used in this study was supported by the California Department of Public Health pursuant to California Health and Safety Code Section 103885; Centers for Disease Control and Prevention (CDC) National Program of Cancer Registries under cooperative agreement 5NU58DP006344; and the NCI Surveillance, Epidemiology, and End Results (SEER) Program under contract HHSN261201800032I awarded to the University of California, San Francisco, contract HHSN261201800015I awarded to the University of Southern California, and contract HHSN261201800009I awarded to the Public Health Institute. The ideas and opinions expressed herein are those of the author(s) and do not necessarily reflect the opinions of the State of California, Department of Public Health, NCI, and CDC or their Contractors and Subcontractors. Support was also provided by the Vermont Cancer Registry, supported in part by Cooperative Agreement NU58DP006322 from the CDC and awarded to the Vermont State Agency of Human Services; the Cancer Surveillance System of the Fred Hutchinson Cancer Research Center, which is funded by contracts N01-CN-67009, N01-PC-35142, N01-PC-2010-00029, N01-PC-2013-00012, HHSN261201300012I, N01-PC-2018-00004, and HHSN261201800004I from the SEER Program of the NCI, with additional support from the Fred Hutchinson Cancer Research Center and the State of Washington; the New Hampshire State Cancer Registry, supported in part by cooperative agreement U55/CCU-121912 and awarded to the New Hampshire Department of Health and Human Services, Division of Public Health Services, Bureau of Disease Control and Health Statistics, Health Statistics and Data Management Section; the North Carolina Central Cancer Registry, which is partially supported by the CDC under cooperative agreement DP12-120503CONT14; and research including data from the Metro Chicago Breast Cancer Registry was supported in part by the Illinois Department of Public Health, Illinois State Cancer Registry that was partially supported by the CDC under cooperative agreement DP12-120504CONT15.

    Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Disclaimer: All statements in this report, including its findings and conclusions, are solely those of the authors and do not necessarily represent the views of the NCI, the National Institutes of Health, or PCORI or its Board of Governors or Methodology Committee.

    Additional Contributions: We thank the participating women, mammography facilities, and radiologists for the data they have provided the BCSC for this study.

    Additional Information: Information about the BCSC is available at http://www.bcsc-research.org/.

    References
    1.
    Brett  J, Bankhead  C, Henderson  B, Watson  E, Austoker  J.  The psychological impact of mammographic screening: a systematic review.   Psychooncology. 2005;14(11):917-938. doi:10.1002/pon.904 PubMedGoogle ScholarCrossref
    2.
    Brodersen  J, Siersma  VD.  Long-term psychosocial consequences of false-positive screening mammography.   Ann Fam Med. 2013;11(2):106-115. doi:10.1370/afm.1466 PubMedGoogle ScholarCrossref
    3.
    Chubak  J, Boudreau  DM, Fishman  PA, Elmore  JG.  Cost of breast-related care in the year following false positive screening mammograms.   Med Care. 2010;48(9):815-820. doi:10.1097/MLR.0b013e3181e57918 PubMedGoogle ScholarCrossref
    4.
    Mandelblatt  JS, Stout  NK, Schechter  CB,  et al.  Collaborative modeling of the benefits and harms associated with different US breast cancer screening strategies.   Ann Intern Med. 2016;164(4):215-225. doi:10.7326/M15-1536 PubMedGoogle ScholarCrossref
    5.
    Nelson  HD, O’Meara  ES, Kerlikowske  K, Balch  S, Miglioretti  D.  Factors associated with rates of false-positive and false-negative results from digital mammography screening: an analysis of registry data.   Ann Intern Med. 2016;164(4):226-235. doi:10.7326/M15-0971 PubMedGoogle ScholarCrossref
    6.
    Lehman  CD, Arao  RF, Sprague  BL,  et al.  National performance benchmarks for modern screening digital mammography: update from the Breast Cancer Surveillance Consortium.   Radiology. 2017;283(1):49-58. doi:10.1148/radiol.2016161174 PubMedGoogle ScholarCrossref
    7.
    Hubbard  RA, Kerlikowske  K, Flowers  CI, Yankaskas  BC, Zhu  W, Miglioretti  DL.  Cumulative probability of false-positive recall or biopsy recommendation after 10 years of screening mammography: a cohort study.   Ann Intern Med. 2011;155(8):481-492. doi:10.7326/0003-4819-155-8-201110180-00004 PubMedGoogle ScholarCrossref
    8.
    Lowry  KP, Trentham-Dietz  A, Schechter  CB,  et al.  Long-term outcomes and cost-effectiveness of breast cancer screening with digital breast tomosynthesis in the United States.   J Natl Cancer Inst. 2020;112(6):582-589. doi:10.1093/jnci/djz184PubMedGoogle ScholarCrossref
    9.
    Marinovich  ML, Hunter  KE, Macaskill  P, Houssami  N.  Breast cancer screening using tomosynthesis or mammography: a meta-analysis of cancer detection and recall.   J Natl Cancer Inst. 2018;110(9):942-949. doi:10.1093/jnci/djy121 PubMedGoogle ScholarCrossref
    10.
    Sumkin  JH, Ganott  MA, Chough  DM,  et al.  Recall rate reduction with tomosynthesis during baseline screening examinations: an assessment from a prospective trial.   Acad Radiol. 2015;22(12):1477-1482. doi:10.1016/j.acra.2015.08.015 PubMedGoogle ScholarCrossref
    11.
    Lowry  KP, Coley  RY, Miglioretti  DL,  et al.  Screening performance of digital breast tomosynthesis vs digital mammography in community practice by patient age, screening round, and breast density.   JAMA Netw Open. 2020;3(7):e2011792. doi:10.1001/jamanetworkopen.2020.11792 PubMedGoogle Scholar
    12.
    Trentham-Dietz  A, Kerlikowske  K, Stout  NK,  et al; Breast Cancer Surveillance Consortium and the Cancer Intervention and Surveillance Modeling Network.  Tailoring breast cancer screening intervals by breast density and risk for women aged 50 years or older: collaborative modeling of screening outcomes.   Ann Intern Med. 2016;165(10):700-712. doi:10.7326/M16-0476 PubMedGoogle ScholarCrossref
    13.
    van Buuren  S.  Multiple imputation of discrete and continuous data by fully conditional specification.   Stat Methods Med Res. 2007;16(3):219-242. doi:10.1177/0962280206074463 PubMedGoogle ScholarCrossref
    14.
    Braithwaite  D, Zhu  W, Hubbard  RA,  et al; Breast Cancer Surveillance Consortium.  Screening outcomes in older US women undergoing multiple mammograms in community practice: does interval, age, or comorbidity score affect tumor characteristics or false positive rates?   J Natl Cancer Inst. 2013;105(5):334-341. doi:10.1093/jnci/djs645 PubMedGoogle ScholarCrossref
    15.
    Dittus  K, Geller  B, Weaver  DL,  et al; Breast Cancer Surveillance Consortium.  Impact of mammography screening interval on breast cancer diagnosis by menopausal status and BMI.   J Gen Intern Med. 2013;28(11):1454-1462. doi:10.1007/s11606-013-2507-0 PubMedGoogle ScholarCrossref
    16.
    Kerlikowske  K, Zhu  W, Hubbard  RA,  et al; Breast Cancer Surveillance Consortium.  Outcomes of screening mammography by frequency, breast density, and postmenopausal hormone therapy.   JAMA Intern Med. 2013;173(9):807-816. doi:10.1001/jamainternmed.2013.307 PubMedGoogle ScholarCrossref
    17.
    O’Meara  ES, Zhu  W, Hubbard  RA,  et al.  Mammographic screening interval in relation to tumor characteristics and false-positive risk by race/ethnicity and age.   Cancer. 2013;119(22):3959-3967. doi:10.1002/cncr.28310 PubMedGoogle ScholarCrossref
    18.
    Breast Cancer Surveillance Consortium. BCSC registries. Accessed January 29, 2022. https://www.bcsc-research.org/
    19.
    D’Orsi  CJ, Mendelson  EB, Morris  EA,  et al.  ACR BI-RADS Atlas, Breast Imaging Reporting and Data System. American College of Radiology; 2013.
    20.
    White  IR, Royston  P, Wood  AM.  Multiple imputation using chained equations: Issues and guidance for practice.   Stat Med. 2011;30(4):377-399. doi:10.1002/sim.4067 PubMedGoogle ScholarCrossref
    21.
    Hubbard  RA, Miglioretti  DL, Smith  RA.  Modelling the cumulative risk of a false-positive screening test.   Stat Methods Med Res. 2010;19(5):429-449. doi:10.1177/0962280209359842 PubMedGoogle ScholarCrossref
    22.
    Hubbard  RA, Ripping  TM, Chubak  J, Broeders  MJ, Miglioretti  DL.  Statistical methods for estimating the cumulative risk of screening mammography outcomes.   Cancer Epidemiol Biomarkers Prev. 2016;25(3):513-520. doi:10.1158/1055-9965.EPI-15-0824 PubMedGoogle ScholarCrossref
    23.
    Hubbard  RA, Miglioretti  DL.  A semiparametric censoring bias model for estimating the cumulative risk of a false-positive screening test under dependent censoring.   Biometrics. 2013;69(1):245-253. doi:10.1111/j.1541-0420.2012.01831.x PubMedGoogle ScholarCrossref
    24.
    Rafferty  EA, Durand  MA, Conant  EF,  et al.  Breast cancer screening using tomosynthesis and digital mammography in dense and nondense breasts.   JAMA. 2016;315(16):1784-1786. doi:10.1001/jama.2016.1708 PubMedGoogle ScholarCrossref
    25.
    Pelletier  E, Daigle  JM, Defay  F, Major  D, Guertin  MH, Brisson  J.  Frequency and determinants of a short-interval follow-up recommendation after an abnormal screening mammogram.   Can Assoc Radiol J. 2016;67(4):322-329. doi:10.1016/j.carj.2015.11.001 PubMedGoogle ScholarCrossref
    26.
    Kirshenbaum  K, Harris  K, Harmon  J, Monge  J, Dabbous  F, Liu  Y.  BI-RADS 3 (short-interval follow-up) assessment rate at diagnostic mammography: correlation with recall rates and utilization as a performance benchmark.   Breast J. 2020;26(7):1284-1288. doi:10.1111/tbj.13838 PubMedGoogle ScholarCrossref
    27.
    Harris  RP, Sheridan  SL, Lewis  CL,  et al.  The harms of screening: a proposed taxonomy and application to lung cancer screening.   JAMA Intern Med. 2014;174(2):281-285. doi:10.1001/jamainternmed.2013.12745 PubMedGoogle ScholarCrossref
    28.
    Lowry  KP, Bell  S, Fendrick  AM, Carlos  RC.  Out-of-pocket costs of diagnostic breast imaging services after screening mammography among commercially insured women from 2010 to 2017.   JAMA Netw Open. 2021;4(8):e2121347. doi:10.1001/jamanetworkopen.2021.21347 PubMedGoogle Scholar
    29.
    Nelson  HD, Pappas  M, Cantor  A, Griffin  J, Daeges  M, Humphrey  L.  Harms of breast cancer screening: systematic review to update the 2009 US preventive services task force recommendation.   Ann Intern Med. 2016;164(4):256-267. doi:10.7326/M15-0970 PubMedGoogle ScholarCrossref
    30.
    Miglioretti  DL, Zhu  W, Kerlikowske  K,  et al; Breast Cancer Surveillance Consortium.  Breast tumor prognostic characteristics and biennial vs annual mammography, age, and menopausal status.   JAMA Oncol. 2015;1(8):1069-1077. doi:10.1001/jamaoncol.2015.3084 PubMedGoogle ScholarCrossref
    31.
    Kerlikowske  K, Grady  D, Rubin  SM, Sandrock  C, Ernster  VL.  Efficacy of screening mammography: a meta-analysis.   JAMA. 1995;273(2):149-154. doi:10.1001/jama.1995.03520260071035 PubMedGoogle ScholarCrossref
    32.
    Conant  EF, Barlow  WE, Herschorn  SD,  et al; Population-based Research Optimizing Screening Through Personalized Regimen (PROSPR) Consortium.  Association of digital breast tomosynthesis vs digital mammography with cancer detection and recall rates by age and breast density.   JAMA Oncol. 2019;5(5):635-642. doi:10.1001/jamaoncol.2018.7078 PubMedGoogle ScholarCrossref
    33.
    Conant  EF, Zuckerman  SP, McDonald  ES,  et al.  Five consecutive years of screening with digital breast tomosynthesis: outcomes by screening year and round.   Radiology. 2020;295(2):285-293. doi:10.1148/radiol.2020191751 PubMedGoogle ScholarCrossref
    34.
    Bernardi  D, Macaskill  P, Pellegrini  M,  et al.  Breast cancer screening with tomosynthesis (3D mammography) with acquired or synthetic 2D mammography compared with 2D mammography alone (STORM-2): a population-based prospective study.   Lancet Oncol. 2016;17(8):1105-1113. doi:10.1016/S1470-2045(16)30101-2 PubMedGoogle ScholarCrossref
    35.
    Kleinknecht  JH, Ciurea  AI, Ciortea  CA.  Pros and cons for breast cancer screening with tomosynthesis: a review of the literature.   Med Pharm Rep. 2020;93(4):335-341. doi:10.15386/mpr-1698 PubMedGoogle Scholar
    36.
    Miglioretti  DL, Abraham  L, Lee  CI,  et al; Breast Cancer Surveillance Consortium.  Digital breast tomosynthesis: radiologist learning curve.   Radiology. 2019;291(1):34-42. doi:10.1148/radiol.2019182305 PubMedGoogle ScholarCrossref
    ×