Suggested Research Questions For Consideration By the CBCRP
1) Nasseri reports in his manuscript that when looking at the breast cancer mortality trends in California from 1970 to 1999, nearly all of the 26% decline in breast cancer mortality over this period was due to reducing mortality among non-Hispanic White women. He further reports that while there was some reduction in mortality among Latinas, there was almost no reduction among African American women and that the there was a 78% increase among Asian/Pacific Islander women. The Hoegh and Davis manuscript reports that the data for Asian/Pacific Islander women are due to increased reporting because these women are newer immigrants and so have been slower to avail themselves of mammograms; they further suggest that this population will reach steady stabilization rate.
While this is good news for non-Hispanic White women, it is shocking that so little reduction in breast cancer mortality has occurred among women of color. Unless there is a wide difference in breast cancer histology between White women and women of color, these data suggest that while early, annual mammograms, detection and diagnostic programs, and state-of-the-art treatment may be available to all California women, they are apparently not equally accessible to women of color. The term available means that the services exist; it does not speak to whether the end users of available services consider them accessible and so choose to use them. More research needs to be done to better understand why available services are not accessible to all populations of California women.
This disparity in breast cancer mortality needs immediate and focused attention by the California Breast Cancer Research Program. Especially because of its diverse population, California has the opportunity to do innovative research among its diverse populations of women, to tease out specifically what variables, in addition to race and ethnicity are responsible for this seemingly intractable occurrence.
2) Yost says in her manuscript that in California, there is an association between socioeconomic status (SES) and breast cancer (not causal), and that the reason for the association is not known. She hypothesizes that it “may be due to differences in diet, lifestyle, delayed or no childbearing or delayed or no breastfeeding.” In order to begin to address this association, it is important to first determine whether it exists.
Currently, California uses census block group data to approximate an individual's SES level by creating an index combining measures of occupation, income, education and cost of living.13 California was one of the states that led the way in developing what was then considered an effective formula for SES; it may not longer be a state-of-the-art measure. Over the last 10 years a multidisciplinary, intellectual discourse on this complex subject has continued in refereed journals; it is time for California to participate in and learn from the current dialogue. SES is extremely important variable to measure with greater accuracy, in order to better understand its impact on breast cancer disparities, especially as it is confounded by race (or color) and ethnicity. The current formula used by California has components that make assumptions that are no longer considered valid. For example, the average cost of a home within a census track is based upon the inaccurate assumption that the average cost of a home is evenly distributed across the census track, as opposed to it's being bimodal, for example. It is quite possible that there is a large up-scale housing stock in a particular census track with a small neighborhood of poor housing stock, in which very low income people live. In this case, averaging such data might result in the erroneous deduction that this particular census track contains only middle class people. A measure of average income may suffer from the same false assumption, and could be further confounded by another erroneous assumption, that the average income within a census track has the same impact across all racial groups within that census track. However, because wealth is not measured and/or culture as it impacts on expenditure priorities is not measured, it is very likely that we are not measuring what we think we are, which could have a huge impact on understanding the role of SES on breast cancer mortality disparity.
There is a growing literature in peer-reviewed journals on the measurement of SES. It is very important that California review and better refine the way that it measures SES. In the meantime, if the current methodology is used in the short run, it should enhanced by interviewing a sample of women from each census track, to validate then accuracy of the proxy formula that is used.
Without a more accurate measure of SES, we don't know whether the question is why do African American Women have such a high breast cancer mortality rate, or whether the correct question is why do poor women (of any race/color) have such a high breast cancer mortality rate. Until California has a better measure of SES, it is impossible to know the role SES plays in breast cancer mortality disparities, or how to address it.
3) The Morris manuscript says that stage at diagnosis represents the best prognostic factor for breast cancer (tumor characteristics are also important). According to the Hoegh and Davis manuscript, non-Hispanic White and Asian Californian women are diagnosed with breast cancer at an earlier stage than African American and Latina Californians, who receive annual mammograms at a later age and later stage of disease. Is the diagnosis of breast cancer at a later stage of disease among Latina and Black Californians as cited by Hoegh and Davis14, due to: 1) having breast cancer at a younger age; 2) having a first mammogram at an older age than recommended, or less frequently than annually, or 3) a more aggressive breast cancer that was not present at the previous annual mammogram? Depending on the answer, two other questions might also be considered: 1) should African American women be encouraged to initiate annual mammograms before age 40? And/or 2) should they be encouraged to have them more frequently than annually? These are important questions to answer if we are to address the increased burden of breast cancer mortality among women at highest risk.
The Mills and Jain manuscript says that rates of early breast disease in California are highest among the urban and among the affluent, with lower rates of early disease reported among the rural and among those in impoverished areas. This is surprising given the monograph that was recently published by the Journal of the National Cancer Institute, which examines the association between cancer and socioeconomic status. Relying on SEER data, the Report says that women in poor areas are more likely to have breast cancer, that the breast cancer is likely to be diagnosed at a later stage and that the women are more likely to die from their disease. If breast cancer rates nationally are highest among the poor, California's lower early cancer detection rates among rural women and women in impoverished areas are likely to have an adverse impact on the breast cancer mortality rates in these populations.
It is important for the CBCRP to continue to do research on timely, annual mammography among these women at higher risk: 1) African American women, 2) Latinas and 3) those women, across all races/ethnicities with lower education, 4) lower SES, or 5) who reside in rural areas, to obtain annual mammograms and to start the practice at an earlier age, when disease is more likely to be found at an earlier stage.
4) Many researchers report that all or nearly all of the disparity in breast cancer mortality is the result of reduced access that African American and low SES women have to early detection and treatment. However, several years ago, research was published on differences in treatment recommendations by provider physicians for African American women and men with cardiovascular disease. The Morris manuscript says that surgical treatment (BCS vs. mastectomy) for breast cancer in California varied with stage (as one might expect), race/ethnicity and SES. Non-Hispanic White women were more likely to receive radiation post BCS than women of color. When working to reduce breast cancer mortality in California, it is important to ask whether populations of color, especially African American women and low-income women of any race, are receiving the appropriate standard-of-care treatment.
The question is, do all California women receive same treatment given stage of disease? Standard-of-care should be driven by current scientific research. In order for the California Breast Cancer Treatment Program to do research on this question for California women with breast cancer, it would be important to develop a partnership with a national professional organization of oncologists to determine the stage-of-disease-based standard of care during a particular period of time and then look retrospectively at patient records to determine whether there is evidence as to whether 1) that standard of care was recommended and 2) whether the patient received that standard of care. The hypotheses to consider would be race/ethnicity, socioeconomic status, method of payment for health services and/or geography increases the likelihood of physicians recommending substandard treatment for breast cancer.
5) With the identification of the human genome, it has become clear that there are more differences within a racial group than there are across racial groups and that race is not a biological construct, but a social one. However, because of persistent racism in the United States and its impact on access to education, the ability to earn income, and access to services that require payment, race continues to be a reasonable measure of equal access to health information, screening, detection, diagnosis, and treatment. Until these social justice issues are resolved, race continues to play an important role in how we evaluate the outcomes of health promotion, disease prevention and treatment programs. Thanks to the recent failure of Proposition 54, California will continue to collect breast cancer data by race and ethnicity as a means to determine whether there are racial disparities in breast cancer incidence and mortality. California breast cancer data continue to show racial and ethnic disparities persist.
However, as the U.S. population becomes more racially mixed, it is important to identify more accurate measures of the impact of barriers to access, on health status, illness behaviors, and treatment outcomes, especially as we work to better address breast cancer disparities. Given this, there is some value in the breast cancer research community, and especially the California breast cancer research community to begin considering that race may be a proxy for something else that may be easier to collect and a more accurate measure. For example, is race a proxy for color; for socioeconomic status; for some combination of color and class? Given California's commitment to reducing racial and ethnic disparities and its ethnically and diverse populations, the CBCRP is ideally situated to begin this important research conversation.
