Trends in Mammography Use Among Women Aged 40 to 49 Years With a History of Breast Cancer

Mammography screening reduces mortality from breast cancer (BC) through early detection; however, the optimal age for initiation of screening among US women at average risk for BC is controversial. Personal BC history is a well-documented risk factor for developing a new breast tumor, with rates ranging from 0.25% to 1% per year.¹ An annual screening mammogram is therefore incorporated into BC survivorship plans unless the patient has undergone a bilateral mastectomy^2,3

In November 2009, the US Preventive Services Task Force (USPSTF) released updated screening guidelines, recommending that initiation of mammography be deferred until age 50 years.⁴ However, this USPSTF guideline was never intended to address surveillance in women with a history of BC. We evaluated the use of mammography among young women in Michigan (stratified by history of BC), relative to the USPSTF update.

Data were extracted from Blue Cross Blue Shield of Michigan claims and enrollment files and processed through its Healthcare Effectiveness Data and Information Set engine, to include women in Michigan 40 to 49 years of age who were commercially insured in a Blue Cross Blue Shield of Michigan Preferred Provider Organization from January 1, 2008, to December 31, 2013. Eligibility was based on Healthcare Effectiveness Data and Information Set specifications, requiring at least 2 prior years of enrollment and no code-based or claim-based evidence of a bilateral mastectomy. This research involved analysis of aggregate data that were anonymized and not linked to any patient identifiers; it was therefore exempt from review by the University of Michigan Institutional Review Board.

Women were considered screened if they received 1 or more mammograms during the measurement year or 1 year prior. History of BC was defined as having at least 1 claim with International Classification of Diseases, Ninth Revision, Clinical Modification coding for BC. We used a generalized linear Poisson regression model to calculate the biennial percentage change in rates of mammography, stratified by history of BC. To determine the differences in percentage change in rates, we fitted the model with history of BC and time as interaction terms. All P values were from 2-sided tests and results were deemed statistically significant at P < .05. Analyses were performed with IBM 19 Statistics SPSS for Windows, version 22 (IBM Corp).

A total of 1.8% of the study sample (8511 of 461 125) had prior BC (Table). Use of mammography decreased from 104 199 of 166 728 patients (62.5%; 95% CI, 62.1%-62.9%) in 2008-2009 to 85 767 of 143 315 patients (59.8%; 95% CI, 59.5%-60.3%) in 2012-2013.

Decreases in the use of mammography screening were observed among women with a history of BC and women without prior BC. The estimated magnitude of the decrease was greater among women with a history of BC than among women without prior BC (5.4%; 95% CI, 2.4%-8.0% vs 2.3%; 95% CI, 1.8%-2.7%; P = .04) (Figure). The odds ratio for the use of mammography among women with vs without BC was 1.32 (95% CI, 1.29-1.34) for 2008-2009; this ratio decreased to 1.24 (95% CI, 1.21-1.26) for 2012-2013.

Our results are compatible with the intended USPSTF goal to reduce the use of mammography screening among young women. However, the reduction in the use of mammography observed among women with a history of BC conflicts with annual mammography surveillance guidelines supported by the National Comprehensive Cancer Network³ and the American Society of Clinical Oncology.² These patients require imaging for early detection of local recurrence and/or new primary BC.

Others have demonstrated that the 2009 USPSTF guideline has generated confusion and uncertainty regarding the value of mammography screening.^5,6 Controversy regarding age-based mammography recommendations for women at average risk of BC may yield the unintended consequence of diminished confidence in this screening tool among the general population as well as among women who are at high risk of BC. In our study, women 40 to 49 years of age with a history of BC who did not have a bilateral mastectomy comprised the comparison population evaluated as being at risk for this unintended consequence. We observed decreasing use of mammography in this patient population for whom screening is clearly indicated. It is possible that automated health maintenance schedules integrated into the electronic medical record program contribute to this nonadherence by not recognizing prior history of BC and generating USPSTF-based mammography schedules intended for the general population.

Our study has several limitations: lack of data regarding extended time intervals, additional age categories, socioeconomic characteristics, race/ethnicity, and alternative screening modalities (eg, breast magnetic resonance imaging). Our results cannot necessarily be generalized to women with other insurance plans or in other regions. Nonetheless, the chronology of our study implicates the USPSTF guideline update as a factor influencing mammography screening practices. This finding is hypothesis generating and warrants further evaluation in other data sets.

Accepted for Publication: May 13, 2018.

Corresponding Author: Lisa A. Newman, MD, MPH, Department of Surgery, Henry Ford Health System, 2799 W Grand Blvd, Detroit, MI 48167 (lnewman1@hfhs.org).

Published Online: August 22, 2018. doi:10.1001/jamasurg.2018.2690

Author Contributions: Drs Bensenhaver and Newman 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.

Concept and design: Bensenhaver, Albert, Hawley, Newman.

Acquisition, analysis, or interpretation of data: Bensenhaver, Perez Martinez, Albert, Petersen, Newman.

Drafting of the manuscript: Bensenhaver, Perez Martinez, Albert, Newman.

Critical revision of the manuscript for important intellectual content: Bensenhaver, Hawley, Petersen, Newman.

Statistical analysis: Perez Martinez.

Obtained funding: Newman.

Administrative, technical, or material support: Bensenhaver, Albert, Newman.

Supervision: Bensenhaver, Petersen, Newman.

Conflict of Interest Disclosures: None reported.

Funding/Support: This research was supported by the Division of Breast Oncology, Department of Surgery, at the Henry Ford Health System, Detroit, Michigan. The Henry Ford Health System provided sponsorship for the study authors to access and analyze the data included in this report.

Role of the Funder/Sponsor: The funding source 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.