Exploring Rural Disparities in Breast Cancer Mortality
|Institution:||Humboldt State University Sponsored Programs Foundation|
Terry Uyeki , M.S. -
Brenda Elvine-Kreis , M.A. -
|Award Cycle:||2014 (Cycle 20)||Grant #: 20AB-1600||Award: $217,102|
|Award Type:||CRC Pilot Award|
|Community Impact of Breast Cancer>Disparities: eliminating the unequal burden of breast cancer|
Initial Award Abstract (2014)
Introduction Women in Humboldt County (HC) are dying from breast cancer (BC) at an excessive rate, which concerns many community members. While women who live in HC appear to get BC at a similar rate as women in the rest of California (CA), HC women with BC are more likely to die from it. Women diagnosed with BC at a late stage are more likely to die from it; however, research has shown that women in HC are not necessarily diagnosed any later than women living in other CA counties. This research project will explore possible reasons for the high BC death rate in HC. The medically underserved county has some unique characteristics: low population density (rural and frontier areas), high poverty, high death rates due to all causes and all cancers, and a high percent of American Indians. Moreover, low-income individuals often lack health insurance, transportation, Internet access, or a phone. We will explore geographic and demographic factors that may be associated with the high breast cancer death rate (mortality) in HC.
Questions: In Humboldt County, what factors are associated with higher BC mortality? Do women living in frontier or rural areas of California have higher BC mortality rates compared to urban areas of CA?
General methodology: We will analyze existing data about BC deaths. Specifically, we will:
1. Analyze data from a state registry with information about people diagnosed with cancer to determine what types of women in HC are dying from BC, specifically looking at race/ethnicity, age at diagnosis, stage and type of cancer, type of neighborhood a woman lives in (neighborhood poverty and education level, and rural vs. frontier), and determine travel distance from residence to treatment facilities using residence information.
2. Determine if the BC death rate in CA is different in low-population areas (frontier and rural) versus high-population areas (urban).
Innovative elements: Unique to this project will be the first-time analysis of BC related deaths in CA based on population density. There have been few research studies that have examined how population density affects BC death rates. Research has shown that rural populations tend to have more health problems compared to urban populations, and this study will help understand what may be contributing to this.
Community involvement: The high death rate from BC has long concerned both county health officials and community health advocates alike, but until now, there has not been the opportunity to adequately investigate factors that may contribute to this. The Humboldt Community Breast Health Project, a direct services breast health advocacy organization, is partnering with the California Center for Rural Policy to address community concerns about the high BC death rates in HC. The partners have co-developed the long term goals, research proposal and collaborative agreement. The community will be engaged through a Community Advisory Council who will advise on all phases of the research: refining research questions, methods for data collection, analysis, and interpretation, and how results are shared with the community. Members range from BC specialists, primary care providers, and health advocates, to BC survivors. Once the data analysis is completed, community liaisons will be consulted with about the best ways to communicate study findings with their community. They will also be trained as local resource experts on how to interpret the results of the study. Community forums will be held to disseminate and discuss the findings with community members, as well as discuss potential next steps to solving this paradox. We aim to actively engage any community member who seeks to be involved in the research.
Future Plans: As relationships between the community and researchers develop during the pilot study, the research-community partnership will have the skills and structure to pursue future research to address disparities in rural BC survivorship. Pilot study findings will have described the population of women who are dying at higher rates from breast cancer and identified the main factors associated with higher BC mortality in HC, and will have compared BC mortality rates between women living in rural and frontier areas versus urban areas in HC and throughout the state of California. This will prepare us to further investigate these factors in a full study, to better understand how to increase BC survivorship in rural communities or studies focused on developing and testing programs to decrease the high BC death rate in HC.
Progress Report 1 (2015)
The purpose of the study is to identify factors that may be associated with the high rates of breast cancer mortality in Humboldt County. For the past 20 years Humboldt County has had one of the highest breast cancer death rates (mortality rates) in California, but it is not known why. In contrast, the number of women who develop breast cancer (incidence rate) in Humboldt County has actually been slightly below the state incidence rate. Humboldt County ranks 24th in occurrence of breast cancer (out of 47 counties), yet is “Number 2” for deaths due to breast cancer (1990-2010).
Data from the California Cancer Registry (CCR) will be analyzed to describe the population of women who are dying from breast cancer and identify possible predictors of breast cancer survival in Humboldt County. This will include examining differences in survival rates among ethnic or racial minority populations, as well as other demographic and geographic sub-populations within the county. In addition, we will analyze whether women living in frontier or rural areas of the state have higher death rates from breast cancer compared to urban areas of the state. This will be the first time that breast cancer related deaths have been analyzed for California based on population density.
There was an initial 4-month delay in funding the project. The application for IRB approval of the study protocol to analyze a 1990-2010 dataset from the CCR was submitted to the Committee for the Protection of Human Subjects (CPHS) in early January. Upon CPHS approval, the application to CCR was submitted, and in March CCR recommended changes in the requested variables in the dataset. Co-investigators amended and submitted a revised protocol to CPHS and CCR in April. CPHS approved the amended protocol mid-April, and on April 29, CCR notified co-investigators of approval of the modified protocol. Researchers should receive the requested dataset in early May. In the interim, project staff have focused on development of communication strategies and materials about the study.
Collaboration with the community is an integral part of the study. A Community Advisory Council (CAC) has met quarterly to advise on the development of the research, communicating and informing the public about the research, and planning for community outreach. Three of the four CAC workgroups have met at least once to develop specific work plan goals. The Community Outreach workgroup is helping to plan the hiring and training of six Community Liaisons and eight Community Forums. The Communications workgroup is assists in developing a plan to reach and inform targeted groups about the study through newsletters, presentations, and flyers. The Data & Research workgroup has advised on protocol questions, limitations of interpretation, and writing study findings for publication. The Dissemination work group will assist in developing a plan for dissemination of research findings, both in refereed journals and the community.
Study findings will be shared through community liaisons and community forums held throughout the county in late 2015. Findings will also be made available on the study’s website (www.rbcss.org) and through social media.
By looking at factors that may contribute to Humboldt County’s unusually high breast cancer death rate, findings from this study will be used to guide a more in-depth investigation of this health disparity, as a first step to better understand how to increase breast cancer survival in rural communities.
Progress Report 2 (2015)
The study will identify factors that may be associated with the high rates of breast cancer mortality in Humboldt County, which has had one of the highest breast cancer death rates from 1990-2010 in California, but it is not known why. Data from the California Cancer Registry (CCR) is being analyzed to describe the population of women who are dying from breast cancer and identify possible predictors of breast cancer survival in Humboldt County. This includes examining differences in survival rates among ethnic or racial minority populations, as well as other demographic and geographic sub-populations within the county. In addition, we are analyzing if women living in frontier or rural areas of the state have higher death rates from breast cancer compared to urban areas of the state –This is the first time breast cancer related deaths have been analyzed for California based on population density.
After initial delays due to final approval of funding status, and addressing the CCR’s requested changes in variables case listing, requiring approval of an amended protocol by the Committee for the Protection of Human Subjects, the CCR dataset was received in May 2015. Chi-squared distribution analyses of the dataset have been carried out for female breast cancer cases diagnosed 1990-2010, and of those cases, women who have died from breast cancer. Two sets of subpopulations have been analyzed – Set 1 compared effect size between variables for cases in Humboldt County (HC) vs. cases in aggregated rural Medical Service Study Areas (MSSA’s) (excluding HC cases) vs. cases in California (excluding HC cases). These analyses allow for comparisons of the effect size between variables for Humboldt County cases compared to other populations in the state. Set 2 compared cases in aggregated rural MSSA’s vs. cases in aggregated urban MSSA’s vs. all California cases. Outcome variables are 5-year survival and death by breast cancer or other causes. Independent variables are demographic variables described above, and stage at diagnosis and tumor grade. Predictors of 5-year survival in HC have been conducted using Multilevel Survival Analysis.. This analysis examined the impact of age, ethnicity, SES, marital status, stage diagnosis, and tumor grade. In addition, Survival Analysis examined the impact of the predictors on Set 1 subpopulations described above. Age-adjusted incidence rates and incidence-related mortality rates have been calculated for the two sets of subpopulations described above. All analyses are currently undergoing quality control verification, and findings will be reported to the Community Advisory Council for their recommendations on how findings should be worded and shared with the public.
During the data analysis phase, the community partner engaged the Community Advisory Council’s (CAC) Communication Workgroup to develop a Communication Plan to generate interest in the study. Informational materials have been distributed at a variety of community events. Study findings will be shared in a number of ways: via the study’s website (www.rbcss.org), Facebook, Newsletters, Community Liaisons, and Community Forums. We have continued to add subscribers to the project mailing list which will be used to keep subscribers up-to-date on the study, to announce the upcoming Community Forums, and to disseminate the three newsletters that will be developed. The CAC’s Community Outreach Workgroup met to help determine the target audiences of the 8 Community Forums to be held through-out the county. They advised on identifying and hiring the Community Liaisons (CLs) for each of these areas, and helped create a job description and application for this position. Four of the six Community Liaison positions have been filled. Training materials for the CL’s are currently being developed. CL’s will assist with the planning and promotion of their assigned Community Forum where study results will be discussed and attendees will be encouraged to share their perspectives and insights on factors contributing to patterns of breast cancer mortality in HC. By looking at factors that may contribute to Humboldt County’s unusually high breast cancer death rate, findings from this study will be a first step to better understand how to increase breast cancer survival in our community to guide future studies of this health disparity.
Conference Abstract (2016)
Exploring Rural Disparities in Breast Cancer Mortality
Uyeki, Terry; Elvine-Kreis, Brenda
Humboldt State University Sponsored Programs Foundation; Breast and GYN Health Project
The Rural Breast Cancer Survival Study is a community-based participatory research pilot study conducted by the California Center for Rural Policy at Humboldt State University, the Breast and GYN Health Project, and community members. Due to higher breast cancer mortality rates reported for Humboldt County, study questions were:
1. What factors are associated with higher breast cancer mortality in Humboldt County?
2. Do women living in frontier or rural areas of California have higher breast cancer mortality rates compared to urban areas of California? How do these breast cancer mortality rates compare to Humboldt County mortality rates?
Data (1990-2010) was analyzed from the California Cancer Registry using 2000 U.S. Census data to calculate 20-year average age-adjusted incidence and incidence-related mortality rates for female breast cancer in California, Humboldt County, rural regions, and urban regions of the state. This study is the first to examine breast cancer related deaths based on population density. Survival analysis of registry data was conducted to explore factors which might explain disparities in breast cancer survival within Humboldt County and among rural and urban areas of California.
Results suggest that Humboldt County has higher incidence and incidence-related mortality rates of invasive breast cancer than the other regions (rural and urban) in California. Rural regions of California have slightly lower incidence and incidence-related mortality rates of invasive breast cancer than urban regions.
Several variables were related to increased risk of death from breast cancer within 5 years of diagnosis: stage, tumor grade, age and marital status at diagnosis.
• Women diagnosed at late stage showed a higher risk of death than those diagnosed at early stage – almost 6 times higher in Humboldt compared to about 5 times higher in the rest of California.
• Women diagnosed at grade 3 or unknown grade showed a higher risk of death than those diagnosed at grade 1; 2-3 times higher in Humboldt compared to over 4 times higher in the rest of California.
• Women who were unmarried at diagnosis showed a higher risk of death than those who were married – over 2 times higher in Humboldt compared to 1.5 times higher in the rest of California.
• Women who were age 65 or older at diagnosis showed a higher risk of death than those who were under age 45 – almost 3 times higher in Humboldt compared to 2 times higher in the rest of California.
Overall, women in Humboldt County with breast cancer were more likely to die within 5 years compared to women in other areas (rural and urban) of the state. Survival rates for women with breast cancer in other rural areas were more similar to survival rates in urban areas, than in Humboldt County.
Five Community Liaisons are advising the project on how to best share study findings with their communities. Eight community forums will be held throughout Humboldt County to disseminate and discuss findings and future research questions with community members towards the goal of improving survival. Findings will be shared via the study’s website and through social media.