Education and mammographic breast density.

Abstract

Purpose: There are conflicting reports on the associations of education with mammographic breast density (MBD). To address this, we investigated the associations of education with MBD and additionally determined if and to what extent this association is mediated by known confounders such as age and adiposity.

Methods: Women (n = 1155) were recruited during their annual screening mammogram at the Joanne Knight Breast Health Center at Washington University School of Medicine in St. Louis, MO. We assessed MBD using volumetric percent density (VPD; Volpara 1.5; Volpara Health^®). We performed generalized linear modeling adjusted for potential confounders to estimate the differences in VPD by education level and evaluated whether associations differ by race and menopausal status. We also performed mediation analysis using PROCESS macro version 4.3. VPD was log-transformed, and back-transformed values are reported.

Results: Women with college (n = 401) and postgraduate education (n = 396) had higher VPD (7.21% [95% CI 6.87-7.59] and 7.18% [95% CI 6.82-7.53], respectively) compared to women (n = 358) with below college education (6.62% [95% CI 6.27-7.00]; p = 0.051) in analysis adjusted for age, body mass index (BMI) and other confounders. However, the association attenuated when the analysis was adjusted for body fat% instead of BMI (college graduate: 7.26% [95% CI 6.92-7.62]; postgraduate: 7.19% [95% CI 6.85-7.55] vs. below college: 6.78% [95% CI 6.41-7.16]; p = 0.156). Body fat% and BMI mediated 71% and 64% of the association between education and VPD, respectively in premenopausal women, but not in postmenopausal women. Associations did not differ by race.

Conclusion: Education was not associated with VPD after adjusting for adiposity. Adiposity mostly mediated the association between education and VPD, particularly among premenopausal women with body fat% a slightly more robust capture of adiposity than BMI.