Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Author ORCID Identifier



Open Access Dissertation

Document Type


Degree Name

Doctor of Philosophy (PhD)

Degree Program

Public Health

Year Degree Awarded


First Advisor

Katherine W. Reeves

Second Advisor

Susan R. Sturgeon

Third Advisor

Nicholas G. Reich

Fourth Advisor

Lynnette Leidy Sievert

Subject Categories



One of the strongest predictors of breast cancer risk is mammographic density; however, incomplete understanding of the mechanisms relating density to risk has limited its use as a marker for breast cancer susceptibility. Hormone fluctuations during the menopausal transition may influence declines in mammographic density and may also trigger the onset of menopausal vasomotor symptoms (VMS), which have been associated with lower breast cancer risk. The effects of hormone changes on density, VMS, and breast cancer risk are complicated by external factors such as changing body mass and hormone therapy use during the menopausal transition.

We evaluated the association between change in BMI and change in mammographic density using volumetric measurement methods. We found that an annual increase in BMI was associated with a decrease in absolute dense volume and percent dense volume. Longitudinal studies of density and breast cancer, or those using density to reflect breast cancer risk, should consider controlling for BMI gain/loss to understand the independent relationship between density and risk. We further investigated the association of VMS and percent mammographic density. We observed no overall association, but found some evidence of an inverse relationship among perimenopausal women and those using hormone therapies. This suggests that an association between VMS and breast cancer risk is not strongly mediated by changes in breast density. Finally, we evaluated VMS and incident breast cancer risk. VMS were associated with a 38% reduction in risk. Adjustment for endogenous hormone levels did not alter our results, suggesting that endogenous hormones play a lesser role in the association between VMS and breast cancer risk than previously hypothesized.

These studies further our understanding of breast cancer etiology. If confirmed, the association between VMS and breast cancer risk could propose VMS as an easily measured factor that could enhance risk prediction. Our findings that this association is not strongly mediated through breast density nor endogenous hormone levels raise provocative questions regarding the mechanisms that link VMS to breast cancer risk. Extending our knowledge of breast cancer etiology through new measurement methods and risk factors may lead to improved risk prediction and opportunities for disease prevention.


Included in

Epidemiology Commons