Abstract B79: Racial and socioeconomic disparities in time to treatment for breast cancer in the District of Columbia

Abstract

Abstract The objective of this research was to determine if there were racial and socioeconomic disparities in time to treatment among women diagnosed with breast cancer in the District of Columbia from 1998-2006. The D.C. Cancer Registry provided a data file containing female breast cancer cases identified from 1998-2006 along with age at diagnosis, census tract, race, ethnicity, diagnosis date, stage, date of first treatment, treatment type, grade, and insurance status. Cancer registry data were linked to 2000 Census data by census tract to create a variable for socioeconomic status based on percentage of the population below the poverty level in each census tract. Analysis of covariance was utilized to determine if the time from diagnosis to first treatment was predicted by race, Hispanic ethnicity, insurance or socioeconomic status while controlling for stage, grade, and age at diagnosis. Time to treatment was also dichotomized into <2 months and >=2months and a multiple logistic regression model using the above variables was utilized to generate odds ratios and 95% confidence intervals to determine if any of the above variables predicted a treatment delay. Finally analysis was stratified by calendar year periods to determine if disparities were reduced or persisted over time. After controlling for the above variables black women were significantly more likely than white women to have a time from diagnosis of breast cancer to treatment of >=2 months (OR=2.19; 95% CI: 1.57, 3.06). Women who were diagnosed in 2001–;2003 (OR=1.72; 95% CI: 1.20,2.40) and 2004-2006 (OR=2.31; 95% CI: 1.62,3.35) were also significantly more likely to have a time to first treatment of 2 or more months compared to women diagnosed in 1998-2000. The adjusted mean times to treatment were significantly different for black versus white women (26.1 vs. 14.1 days). The mean time to treatment differed significantly by time period as well, at 14.8,19.4, and 26.1 days for 1998-2000,2001–;2003, and 2004-2006 respectively. A subgroup analysis of women with estrogen and progesterone receptor status data yielded similar results to the overall analysis. Lastly, when stratified by calendar period, black women were significantly more likely to experience longer time to treatment if diagnosed in 2001–2003 and 2004-2006 but not in 1998-2000. Overall, black women with breast cancer in the District of Columbia are more likely to experience delays in treatment regardless of insurance type, socioeconomic status, and cancer characteristics such as stage and grade. Racial disparities in time to treatment have not abated over time and may have increased. Further research should be undertaken to determine the reasons that could underlie these delays and determine if disparities are actually worsening over time. Citation Information: Cancer Epidemiol Biomarkers Prev 2010;19(10 Suppl):B79.