The role of Hispanic race/ethnicity and poverty in breast cancer survival.

The association of Hispanic race/ethnicity and poverty with general survival time and breast cancer survival time was examined for a total of 14,896 breast cancer patients (14,035 White and 861 Hispanic) included in the National Cancer Institute Surveillance Epidemiology and End Results (NCI SEER) program in New Mexico and San Francisco between 1975 and 1984. Variables examined included: age, marital status, stage at diagnosis, tumor histology, delay, treatment, period of diagnosis (1975-79 vs. 1980-84), and poverty. Univariate analysis of 14,896 patients indicated that a greater proportion of Hispanics (vs. Whites) with breast cancer were: younger than age 50, married, diagnosed at a later stage, diagnosed in New Mexico, lived in greater poverty, were diagnosed between 1980-84, and died from breast cancer. Univariate Cox Proportion Hazards analysis indicated that poverty was a significant predictor for reduced general survival time. Being diagnosed in the 1980-84 period was a predictor for improved general survival time. Poverty and Hispanic race/ethnicity were significant predictors of reduced breast cancer survival time. Multivariate Cox Proportional Hazards models indicated that Hispanic race/ethnicity was a significant risk factor for breast cancer survival time for women aged 50 and older. For White women: state, marital status, poverty, surgery, radiation/hormonal treatments, and histology were significant risk factors for breast cancer survival time. For Hispanic women: stage, surgery, hormonal treatment and period of diagnosis were significant risk factors for breast cancer survival time. For White breast cancer patients, period of diagnosis was not a significant risk factor for reduced breast cancer survival time; but for Hispanics, it was a significant risk factor. In the age and race/ethnicity-stratified models of breast cancer survival time, similar risk factors emerged for both Whites and Hispanics. For both younger and older Hispanics, being diagnosed in the early 1980's (vs. the late 1970's) was associated with reduced breast cancer survival time--vs. Whites, who experienced no significant change in breast cancer survival time in the same time period. Poverty was not a predictor for Hispanic survival time in any of the models; however, it was a predictor for younger Whites for breast cancer survival time. These results fueled discussion in three areas targeting breast cancer in underserved women: the development of racial/ethnic-specific cancer control guidelines, the development of a breast cancer integrated delivery system, and population management.