Geographic Variation in Breast Cancer Stage at Diagnosis

Institution: University of California, Los Angeles
Investigator(s): Pamela Davidson, Ph.D. -
Award Cycle: 2001 (Cycle VII) Grant #: 7PB-0128S Award: $425,497
Award Type: Request for Applications
Research Priorities
Community Impact of Breast Cancer>Health Policy and Health Services: better serving women's needs



Initial Award Abstract (2001)
Breast cancer stage at diagnosis (BCSAD) is generally reported in 3 progressive stages: 1) localized tumors are malignant and invasive but confined to the organ of origin; 2) regional neoplasms have extended beyond the organ of origin into surrounding tissues, involve regional lymph nodes, or both; and 3) distant tumors have spread to remote parts of the body from the primary site. An earlier, more localized BCSAD is highly desirable because it is directly related to survival and mortality. But previous research has shown that the percentages of breast cancer patients with a localized tumor at diagnosis ranged from 40% to 71% in California counties. The individual-level patient characteristics (e.g., age, race-ethnicity, health insurance status) associated with BCSAD have been investigated. This pioneering research will focus on the community-level factors (health policy and financing, health care delivery system, and community risk factors) associated with better or worse BCSAD that have not been investigated previously.

Data obtained from this study will be used to address two central research questions: 1) How do county level differences in health policy and financing, local health care delivery systems, and community risk factors influence BCSAD? and 2) What are the implications for developing effective community-level policies, programs, and health services to improve this cancer outcome?

The general methodology in non-technical terms: The research team will obtain county-level data from multiple data sources including: 1) Information on female breast cancer patients obtained from the California Cancer Registry; 2) A variety of other data sources to obtain information on health policy and financing, health care delivery system, and community risk factors; and 3) Information obtained from the California Health Interview Survey (CEIIS), focusing on access to health insurance and health care including cancer related questions developed by the National Cancer Institute. Data will be analyzed to compare county-level differences in BCSAD and the individual- and community-level determinants. Two members of the project Advisory Board are founders of community-based organizations providing breast cancer consumer education and advocacy programs. Advisory Board members will provide on-going consultation and advice regarding implications for supporting breast cancer patients and reducing the impact of breast cancer in California.

No studies in cancer health services research have investigated the effects of community-level determinants of cancer outcomes in a comprehensive way. However, as information systems and data analysis methods become more sophisticated, researchers are beginning to develop the capacity to obtain and analyze data from a variety of data sources, enabling a more comprehensive analysis of both patient and community characteristics influencing BCSAD. Ultimately, state and local policymakers and health services leaders should have the data to: 1) develop a better understanding of the community risk factors and resources unique to their geographic area; 2) become more aware of effective community-level interventions leading to better cancer prevention and control; and 3) have a data driven rationale for allocating resources to improve policies, programs, and health services.


Final Report (2006)
Breast cancer stage at diagnosis (BCSAD) is an important cancer outcome because it is directly related to survival and mortality. Although considerable research has been conducted on determinants of BCSAD, most cancer health services researchers have focused on individual-level determinants (age at diagnosis, race/ethnicity, socioeconomic class, and health insurance status). Less is known about the effects of community-level determinants (health policy, health care delivery system, and community risk factors) and the extent to which they contribute to geographic variation in BCSAD. The general purpose of this study was to develop and test a more comprehensive approach for investigating individual- and community-level determinants of BCSAD using multiple data sources merged with cancer registry data.

We examined the role of community risk factors and resources on BCSAD (Davidson et al., 2005). Results confirmed that community-level predictors of socioeconomic and delivery system context matter. If a woman resided in a neighborhood with greater percentages of female-headed households, persons living below the poverty level, less educated people, and more recent immigrants, then her chances of being diagnosed at an earlier stage were diminished. Conversely, if a woman resided in a neighborhood with greater percentages of older females, 65+ years (a proxy for Medicare coverage), her access to medical care and the probability of earlier BCSAD increased. Additionally, county-level insurance rates and residing in counties where greater percentages of women ever had a mammogram were associated with in situ and early stage diagnosis. Supply of primary care physicians and radiologists were also positively associated with earlier BCSAD. Utilizing services at hospitals serving lower volume of patients with breast carcinoma was associated with later BCSAD.

Additionally, we investigated breast cancer screening outcomes in California in 1990 and 2000, trends over time, and the role of community level determinants (Flores, et al., 2006, manuscript in progress). Analysis revealed that the number of in situ, localized, and regional cases increased substantially from 1990 to 2000, while the number of cases detected at the remote stage remained almost the same. Furthermore, areas with (1) lower socioeconomic status, (2) lower percentages of individuals with a high school education, and (3) higher percentages of Latinos showed no statistically significant changes in percent of cases detected at the localized stage from 1990 to 2000. Conversely, areas with (1) higher socioeconomic status, (2) higher percentages of individuals with a high school education, and (3) lower percentages of Latinos showed significant increase in the percent of cases diagnosed at the localized stage.

Our research indicates that there is a critical need for breast cancer early detection programs in high-risk communities with poorer breast cancer outcomes, such as those with lower SES, lack of health insurance coverage, and a greater Latino(a) population. Despite the existence of state administered programs, which have helped increase the percent of women with early stage at diagnosis, certain areas continue to show low early detection rates. These findings indicate that community-level research can help elucidate significant predictors of early stage at diagnosis and also be used to identify and monitor geographic areas and population subgroups showing little or no improvement in early detection.

Future research to examine variation in BCSAD in California counties should be conducted to identify those showing decline or little improvement in early detection over time. Case studies using both quantitative and qualitative interview data could be used to determine why these inequities persist. California counties could use this information to better understand the community risk factors and resources associated with earlier and later BCSAD, and to develop more effective community-level interventions. Another important research issue for investigation is the impact of the California state-administered breast cancer early detection programs (BCCCP and BCDEP). Examining CCR and Cancer Detection Section (CDS) breast cancer data within California and also using multi-state comparison data could be used to examine effects of health policy and financing on breast cancer stage of diagnosis.


Symposium Abstract (2003)
Background: Breast cancer stage at diagnosis (BCSAD) is an important cancer outcome because it is directly related to survival and mortality. Although considerable research has been conducted on determinants of BCSAD, most cancer health services researchers have focused on individual-level determinants (age at diagnosis, race-ethnicity, socioeconomic class, and health insurance status). Less is known about the effects of community-level determinants (health policy, health care delivery system, and community risk factors) and the extent to which they contribute to geographic variation in BCSAD. This pioneering research will identify significant community risk factors and resources unique to a geographic area associated with cancer outcomes.

Research Questions: Data obtained for this study will be used to test several hypotheses emerging from central research questions: 1) Which community-level variables are most highly correlated with BCSAD; 2) Is there geographic variation in BCSAD; and 3) What are the implications of the findings for developing effective community-level interventions.

Methods: The California Cancer Registry provided the data for female breast carcinoma cases diagnosed in California for the most current 5 years. Other key variables were constructed from multiple data sources including the U.S. Census, California Health Interview Survey, Area Resource File, etc. These data sources were merged and analyzed using bivariate and multivariate statistical methods.

Impact: This research will develop and implement a methodology for identifying community-level factors associated with BCSAD and for assessing geographic variation in this outcome. Ultimately, state and local policymakers and health services and community leaders will: 1) develop a better understanding of the community risk factors and resources unique to their geographic area; 2) become more aware of community-level interventions leading to better cancer prevention and control; and 3) have a data-driven rationale for allocating resources to improve cancer-related policies, programs, and services.


Symposium Abstract (2005)
Background: This study investigates the individual and community determinants of breast cancer stage at diagnosis (BCSAD) using multiple data sources merged with cancer registry data. The literature review yielded 5 studies that analyzed cancer registry data merged with community-level variables (1995-2004).

Methods: Community variables constructed for the current study reflected social and economic risk factors, physician supply, and health maintenance organization penetration. A statistical technique known as multivariate logistic regression was used to identify the major factors that can be used to predict the likelihood that a woman’s breast cancer will be at higher stages at diagnosis.

Results: Disparities remained for black and Hispanic females in California, who were least likely to be diagnosed early compared with their white counterparts. Younger (< 40 years) and middle-aged (40-64 years) females were less likely to be diagnosed at early BCSAD, compared to older females (65+ years). Utilizing services at hospitals serving lower volume of patients with breast carcinoma was associated with later BCSAD. After controlling for individual-level factors, community-level variables constructed at the census block group (CBG) and county level were tested. If a woman resided in a neighborhood with greater percentages of female-headed households, persons living below the poverty level, less educated people, and more recent immigrants, then her chances of being diagnosed at an earlier stage were diminished. If, conversely, she resided in a neighborhood with greater percentages of females 65+ years (a proxy for Medicare coverage), her access to medical care and the probability of earlier BCSAD increased. County-level insurance rates and residing in counties where greater percentages of women ever had a mammogram were associated with in situ and early stage diagnosis. Similarly, the supply of primary care physicians and radiologists was associated positively with earlier BCSAD.

Conclusions: Results confirmed community predictors of socioeconomic and delivery system context matter, although the individual level predictors showed a stronger effect. Nevertheless, analysis of community variables is promising for guiding and evaluating the effects of health policy and developing community and delivery system interventions for earlier detection and treatment of breast carcinoma.(CANCER, March 1, 2005, volume 103, number 5).


Symposium Abstract (2005)
Background: Breast cancer stage at diagnosis is directly related to survival and mortality. Since 1991, a state and federally funded program has provided free breast exams and mammograms to low-income and underinsured women in California. Although screening rates have increased over time, disparities in breast cancer stage at diagnosis continue to persist. Most research has focused on individual level determinants. Less is known about the effects of community-level determinants on breast cancer stage at diagnosis. This study identifies specific community risk factors that were correlated with breast cancer stage at diagnosis in California, in the years 1990 and 2000.

Objective: To examine trends in breast cancer stage at diagnosis in the state of California between 1990 and 2000, and identify community level determinants that may be associated with these trends over time.

Methods: We compared breast cancer stage at diagnosis (in situ, Stage I, Stage II/III, and Stage III/IV) in California for 1990 and 2000, by community socio-demographic characteristics (percent education less than high school, percent Latino(a)s, and percent less than 200% of federal poverty level). Community-level variables were constructed using the California Cancer Registry 1990-2000, the 1990 and 2000 U.S. Census.

Results: The number of in situ, localized (Stage I), and regional (Stage II and III) cases increased substantially from 1990 to 2000. The number of cases detected at the remote stage (Stage IV) remained almost the same from 1990 to 2000. Areas with 1) a low socio-economic status, 2) a lower percentage of individuals with a high school education, and 3) a high percentage of Latino (a)s showed no statistically significant changes in the percent of cases detected at the localized stage from 1990 to 2000. Conversely, areas with 1) a high socio-economic status, 2) a higher percentage of individuals with a high school education, and 3) a low percentage of Latino (a)s a showed a significant increase in the percent of cases diagnosed at the localized stage. There was no significant increase in the percent of in situ cases detected in areas with a higher percent of individuals with a high school education, areas with a low percentage of Latino (a) s, and areas with a high percentage of Latino (a)s. The percent of in situ cases increased significantly from 1990 to 2000 in all other areas.

Conclusions: Our research indicates that there is a critical need for breast cancer early detection programs in high-risk communities with poorer breast cancer outcomes, such as those with lower SES and a greater Latino (a) population. Despite the existence of state administered programs that provide breast cancer screening services, certain areas continue to have low early detection rates. These findings indicate that community-level research can help elucidate significant predictors of early stage at diagnosis. California counties could use this information to better understand the community risk factors and resources that are associated with earlier and later breast cancer stage at diagnosis, and to develop more effective community-level interventions.

Role of community risk factors and resources on breast carcinoma stage at diagnosis.
Periodical:Cancer
Index Medicus: Cancer
Authors: Davidson PL, Bastani R, Nakazono TT, Carreon DC.
Yr: 2005 Vol: 103 Nbr: 5 Abs: Pg:922-930