Two research letters published in JAMA addressed lingering concerns about the appropriate use of supplemental screening for women with dense breasts.
, PhD, professor of medicine at Boston University School of Medicine and research scientist with the Veterans Affairs Boston Healthcare System, and colleagues studied the content, readability and understandability of dense breast notifications women received following screening mammography.
Kressin and colleagues found wide variability in the content of dense breast notifications across 23 states, with poor understandability and frequent discontinuity with states’ average literacy.
“Such problems may create uncertainty for women attempting to make personalized decisions about supplemental screening and may exacerbate disparities in breast cancer screening related to low health literacy,” Kressin and colleagues wrote.
Regarding supplemental screening, Elizabeth A. Rafferty, MD, who served as head of the Avon Foundation Comprehensive Breast Evaluation Center at Massachusetts General Hospital during the study conduct, and colleagues found that the addition of tomosynthesis to digital mammography increased the cancer detection rate for women with dense and nondense breasts.
Women with heterogeneously dense breasts experienced the largest combined gains with this screening platform; however, the detection rate among women with extremely dense breasts did not reach statistical significance.
Because nearly half of U.S. states require that women with dense breasts be notified of their status, much discussion has focused on the optimal means of notification.
“The text of dense breast notifications may affect women’s ability to understand their message,” Kressin and colleagues wrote.
The researchers reviewed laws requiring dense breast notifications for states with legislation effective as of Jan. 1, 2016 (n = 23). They excluded Delaware because the legislation language was not sufficiently detailed to analyze notification content.
Nineteen states (83%) mandated specific language; the remaining four states (17%) mandated required components but not specific language.
Seven states (30%) required that a generic dense breast notification be sent to every woman undergoing a screening mammogram. The remaining states only required notifications for women with dense findings.
All dense breast notifications mentioned masking bias — the situation in which dense breasts mask cancer on mammography. Seventeen states (74%) mentioned the association between dense breasts and increased cancer risks, and 15 (65%) mentioned supplemental screening options and advised women to consult their doctors.
Of the 15 states that mention supplemental screening, six (40%) inform women that such screening might be beneficial, with four mentioning specific modalities.
The researchers assessed readability by using the Flesch–Kincaid reading grade level feature in Microsoft Word and the Dale–Chall readability grade score. They assessed understandability through the Patient Education Materials Assessment Tool (PEMAT; range, 1% to 100%), and obtained the proportion of adults in each state who lacked basic prose literacy skills from available statistics.
Flesch–Kincaid readability levels varied from grades 7 to 19.4 (mean, 11.1). Most exceeded the recommended readability levels of grades 7 to 8.
The researchers observed that approximately 20% of the population read below a grade 5 level.
Overall Dale–Chall scores appeared slightly higher (grade range: 9-10 to 13-15).
All dense breast notifications scored poorly for understandability, with a PEMAT range between 11% and 33%.
Widespread discordance occurred between states’ dense breast notification readability and corresponding levels of basic literacy.
Only three states — Alabama, Hawaii and New York — had dense breast notifications at or below a grade 8 reading level. The researchers observed that some of the highest readability levels for dense breast notifications occurred in states with the lowest literacy levels.
“Many dense breast notifications appropriately encourage discussions and shared decision-making between patients and physicians,” Kressin and colleagues wrote. “The lack of evidence regarding supplemental screening may contribute to variation in dense breast notification content and to physician difficulty explaining results and conducting personalized risk assessments. These findings add to other expressed concerns regarding dense breast notification reporting laws.”
Determining appropriate screening
Breast density can reduce mammographic sensitivity and specificity, and increased breast density is associated with increased tumor size and worsened prognosis.
Currently, 24 states mandate that women with dense breasts be notified of the implications of breast density in an attempt to encourage discussions between patients and physicians regarding supplemental screening. However, the efficacy of supplemental screening methods for women with dense breasts has not been extensively studied.
Rafferty and colleagues accessed data from 452,320 breast examinations performed at U.S. hospitals to observe screening performance metrics for digital mammography alone or in combination with tomosynthesis.
The study included data from women with dense and nondense breasts. The researchers analyzed recall rates and overall and invasive cancer detection rates for screening with and without tomosynthesis.
Digital mammography alone accounted for 278,906 screenings; 173,414 screenings included digital mammography and tomosynthesis.
A total of 2,157 cancers were diagnosed (digital mammography alone, n = 1,207; digital mammography plus tomosynthesis, n = 950).
Recall rates per 1,000 screens decreased with tomosynthesis for women with nondense breasts (90 to 79; difference, –12; 95% CI, –14 to –9; P < .001) and dense breasts (127 to 109; difference, –18; 95% CI, –21 to –15; P < .001).
Positive predictive values of recall also increased with tomosynthesis for women with nondense breasts (7.1% vs. 5.1%; P < .001) and dense breasts (5.7% vs. 3.8%; P < .001).
Cancer detection rates increased in both groups, with tomosynthesis increasing the rates of invasive cancer detection per 1,000 cases from 3 to 4 among women with nondense breasts (difference, 0.9; 95% CI, 0.4-1.5; P < .001) and 2.9 to 4.2 among women with dense breasts (difference, 1.4; 95% CI, 0.9-1.9; P < .001).
Subgroup analyses showed the greatest improvements among women with scattered fibroglandular densities and heterogeneously dense breasts. Women with almost entirely fat breasts and with extremely dense breasts did not have significant differences.
The researchers acknowledged the retrospective study design and the use of population-level data as study limitations.
“For women classified as having dense breast tissue, most have heterogeneously dense breasts,” Rafferty and colleagues wrote. “[This] mandates caution in drawing conclusions regarding the performance of tomosynthesis for the small proportion of women with extremely dense breasts."– by Cameron Kelsall
Rafferty and colleagues report no relevant financial disclosures. Kressin reports receiving a research award from the Department of Veterans Affairs Health Services Research and Development Service.