Abstract
B51
Background: Recent data have shown that survival after a breast cancer diagnosis is poorer among First Nations women compared to other Ontario women. There are many possible determinants of this survival discrepancy, many of which have not been studied among a Native population. The main purpose of this study is to identify determinants for the poorest breast cancer survival by comparing stage, treatment and other risk factors among First Nations and non-First Nations women diagnosed with breast cancer from 1995-2004 in Ontario. Once these have been determined, actions can be taken to improve the prognosis of First Nations women with breast cancer. Objectives: 1)To compare the distribution of stage at diagnosis (stage II+ vs. stage I) for Ontario First Nations and non-First Nations women diagnosed with breast cancer between 1995 to 2004. 2) To compare other potential determinants of survival between the two populations, such as treatment, prognostic features of breast cancer, co-morbidity, and distance from an Integrated Cancer Program by stage at diagnosis. 3) Depending on the number of deaths in the First Nations women at the end of the study, we propose to compare stage specific survival between the two populations. Hypotheses: 1) First Nations women are diagnosed with breast cancer at a later stage of disease (either stage II, III or IV) than non-First Nations women in Ontario. 2) First Nations women diagnosed with breast cancer at a later stage in Ontario may differ by the treatment they receive, prognostic features of breast cancer, co-morbidity, and distance from an Integrated Cancer Program compared to the general population. 3) Stage-specific survival among women diagnosed with breast cancer is worse for First Nations women in Ontario compared to general population. Study Design: This study employs a case-case design using the cohort of First Nations people in Ontario to identify an estimated 315 women diagnosed with invasive breast cancer between 1995 and 2004. Concurrently, a random sample of 630 non-First Nations women will be selected through the population-based cancer registry at Cancer Care Ontario and matched 2 to 1 on five-year date of diagnosis, age at diagnosis (15-54 vs. 55+), and Integrated Cancer Program first attended. Data on stage at diagnosis, treatment received, risk factors and co-morbid conditions will be collected from medical charts at the provincial Integrated Cancer Programs. Analysis: To address the primary objective of the study, stage at diagnosis will be aggregated into a binary variable to obtain the distribution across the two populations. Logistic regression models will be performed to investigate the second study objective. The analyses will test the influence of the independent variables of interest by stage at diagnoses comparing the two populations. We propose a Cox-proportional hazards regression model to assess stage specific survival (for stages 2+) for the third study objective. Contribution of Study: This is a unique opportunity to study factors related to breast cancer survival in this Ontario’s First Nations women. This work is expected to inform health care decision makers about where barriers may exist with respect to cancer screening, treatment and surveillance for First Nations women. The results of this study may support improvements of cancer care for this population.
First AACR International Conference on the Science of Cancer Health Disparities-- Nov 27-30, 2007; Atlanta, GA