Racial Disparity in Breast Cancer Mortality

Institution: University of California, San Francisco
Investigator(s): Rebecca  Smith-Bindman , M.D. -
Award Cycle: 2003 (Cycle IX) Grant #: 9PB-0205 Award: $580,887
Award Type: Request for Applications
Research Priorities
Disparities>Disparities: eliminating the unequal burden of breast cancer



Initial Award Abstract (2003)
There are significant differences in breast cancer death rates among different racial groups, and these differences have increased over the last decade. African American women have lower rates of breast cancer than white women, but higher mortality rates. Additionally, African American women have not seen the same improvements in mortality over the last decade that have been achieved by white women; thus, the mortality gap has been increasing. Hispanic and Asian women tend to have lower mortality rates than white women, although these groups also have not seen the same improvements in mortality over the last decade as have been seen in white women. Although the reasons for racial disparity in breast cancer mortality are not known, several factors may play a role, including the biology of breast cancer, differences in screening mammography use, the quality of breast cancer treatment, and differences in social/cultural factors that affect interactions with health care systems.

This research will address the question what are the causes for different rates of breast cancer deaths among women from different racial and ethnic groups.

In order to disentangle the reasons for racial disparities in breast cancer mortality, it is necessary to compare the breast cancer features (such as stage of disease at diagnosis, breast cancer treatments) in a population of women where the use of screening mammography, breast cancer treatments, and other relevant variables are known. We propose to analyze data from two large, population-based data sources. These data will include approximately 95,000 women with breast cancer diagnosed between 1992-2001, including 5880 African Americans, 3240 Hispanics, and 2573 Asian women. Additionally, we propose to conduct interviews of a group of 200 women who were diagnosed with late-stage breast cancer to identify additional factors associated with late presentation to care, such as low literacy and preferences and access to different types of screening and treatment.

This research makes use of two very large, ethnically and racially diverse, comprehensive and population-based databases that will allow us to address questions regarding racial/ethnic disparities in breast cancer mortality. We will simultaneously look at several factors that might suggest that the biology of cancer, screening, or treatment might be particularly important and causal for the racial disparity in mortality, and this has not been done previously. Additionally, because racial disparities have been partially explained by later stage at presentation, our interviews with women who have fallen between the cracks of our health care systems of early detection and treatment will provide valuable information regarding a group of women who it is particularly important to reach in order to improve breast cancer mortality. Patients from California with late-stage presentation have not been interviewed previously and their experiences will give us valuable insights into why these women present late for care when their disease is less curable.


Final Report (2007)
Introduction: There is substantial variation in breast cancer outcomes by race and ethnicity. For example, African American women have significantly fewer breast cancers in comparison to white women, yet have significantly higher mortality rates. The goals of this project were to try to understand the reasons for the racial and ethnic differences in breast cancer and to disentangle the possible causes for differences in mortality including differences in screening mammography and breast cancer treatments.

Progress Towards Specific Aims: 1) This aim was to identify racial and ethnic differences in the use of screening mammography. In contrast to the widely held view that there are no longer significant differences in the use of screening, we found substantial variation persists; minority women underutilize mammography in comparison to whites and that overall the use of mammography is substantially lower than is estimated based on self report. For example, using national Medicare data, among women age 65 and older, after adjusting to a standard age distribution, approximately 52% of white women, as compared with 42% of African American, 36% of Asian and 38% of Hispanic women underwent at least one mammogram in 2000 and 20001.

These results were published Am J Prev Med 2006
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16829331&query_hl=1&itool=pubmed_docsum and

Ann Intern Med 2006
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16618948&itool=pubmed_docsum.

(2) This aim was to identify if differences in breast cancer rates are due to differences in the utilization of mammography. We were able to evaluate similarly screened women and found breast cancer rates were similar among Whites and African Americans, and lower among Asians and Hispanics after accounting for mammography. We found that once adjusted for the use of mammography, there were no significant differences in advanced cancer rates and total cancer rates between White and African American women who were screened between 1-3 years prior to diagnosis. These results were published in the Ann Intern Med 2006.

(3) Identify racial/ethnic differences in breast cancer tumor characteristics at detection, such as size and stage. There are substantial differences in breast tumors at detection by race and ethnicity, and this aim was to determine if this was due to mammography, or late presentation to care. We found that most, but not all, of the differences in tumor characteristics at diagnosis were due to later use of mammography, rather than underlying biology. These results were published in the Ann Intern Med 2006 and a second paper is nearly complete and will be submitted to Cancer.

(4) This aim was to identify differences in breast cancer treatments by race and ethnicity. We found that treatment for breast cancer varied by race and ethnicity during the 1990s, and that patients from minority groups often were less likely to receive appropriate treatments for early-stage breast cancer including breast breast-conserving surgery, and if they did undergo breast-conserving surgery, were less likely to receive recommended adjuvant radiation. Additionally, there were differences in receiving other measures of high quality care, such as measurement of estrogen/ receptor status. Further, women in rural areas, and older women were also less likely to receive optimum care.

These results were presented in Cancer 2005
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16211547&query_hl=1&itool=pubmed_docsum.

Differences in the quality of breast cancer care among vulnerable populations.
Periodical:Cancer
Index Medicus: Cancer
Authors: Haggstrom DA, Quale C, Smith-Bindman R.
Yr: 2005 Vol: 104 Nbr: 11 Abs: Pg:2347-58

Does utilization of screening mammography explain racial and ethnic differences in breast cancer?
Periodical:Annals of Internal Medicine
Index Medicus: Ann Intern Med
Authors: Smith-Bindman R, Miglioretti DL, et al, and Kerlikowske K
Yr: 2006 Vol: 144 Nbr: 8 Abs: Pg:541-553

Screening mammography in the American elderly.
Periodical:American Journal of Preventative Medicine
Index Medicus:
Authors: Kagay CR, Quale C, Smith-Bindman R.
Yr: 2006 Vol: (2) Nbr: Abs: Pg:142-9