Research

As a medical resident, I realized that many older women have screening mammograms despite few data on the benefits for this population. Therefore, as a research fellow, I used data from the National Health Interview Survey to examine rates of mammography screening among older women. I found that 39% of US women >80 years with short life expectancies and therefore little chance to benefit were screened with mammography, while 38% of women >80 years with greater than 10-year life expectancy were not screened. These findings highlighted the need to improve targeting of mammography screening to older women by life expectancy, and most guidelines now encourage clinicians to consider patient life expectancy when deciding on screening. Curious as to why so many older women with little chance to benefit were screened, I used qualitative methods to study older women’s decision making around mammography screening. I found that while some of the oldest women have strong opinions about whether or not to get screened, most women are influenced by their physicians’ recommendations. Physicians, however, report feeling ill prepared to discuss the benefits and risks of mammography or long-term prognosis with older women. 

To improve the quality of older women’s decision making around mammography screening, I developed and evaluated a comprehensive pamphlet decision aid on mammography screening for women >75 years. The decision aid is available on ePrognosis, and was deemed high quality in two independent reviews. In addition, the National Cancer Institute (NCI) designated the decision aid an evidence-based cancer control program. In a large multi-site cluster randomized clinical trial, we found that the decision aid increased older women’s knowledge of mammography’s pros and cons and led to 9% fewer women at low or average risk being screened (51% vs. 60%, p=0.006 [Schonberg et al., JAMA Intern Med., 2020]). However, the decision aid did not support clinician communication around screening, nor did it individualize outcomes of screening based on older women’s health and breast cancer risk. Therefore, we recently developed a web-based conversation aid (called decidetogether.info) to support shared decision-making between women aged 75 and older and their primary care clinicians during an encounter (Schonberg et al., J Am Geriatr Soc., 2025). 

In developing my mammography screening decision aid, I learned that older women wanted more information about their breast cancer risk to make informed decisions about mammography screening. Since the Breast Cancer Risk Assessment Tool (BCRAT, a.k.a. “Gail model”) is the most widely available and commonly used breast cancer prediction model, I examined its performance in older women and found that BCRAT overpredicted breast cancer in women >75, especially those with multiple comorbidities. Therefore,  we developed a novel model to predict 10-year risk of breast cancer, breast cancer death, and non-breast cancer death among postmenopausal women aged 55 and older using data from the Nurses' Health Study, the Black Women's Health Study, the Women's Health Initiative, and the Multiethnic Cohort. Using this model, we created a risk-based decision tool to help older women with decision making around how often to be screened, when to stop screening, and for those at higher breast cancer risk whether to consider breast cancer prevention medications. I made the risk model and the risk-based decision tool available for use
on the ePrognosis website.