Abstract A63: Time course from screening mammography to biopsy to treatment among black and white, non-HMO Medicare beneficiaries in 2005-2008

Abstract

Abstract Background: While breast cancer incidence is higher among whites, breast cancer mortality is higher among blacks. Delays in diagnosis and treatment could contribute. Objective: Describe the time course of screening, diagnosis and treatment among non-Hispanic black and white non-HMO Medicare beneficiaries ages > 65 years. Data and Methods: We selected a 2005-2008 sub-sample of Medicare beneficiaries, restricted to non-Hispanic black women and white women ages 65 years and older who were: alive from at least 2005 to 2007; had out-patient (Part B) Medicare insurance; and were not enrolled in HMO's for any part of the observation period. For each continental US county or county equivalent we selected a random sample of 250 beneficiaries from each racial/ethnic group, based on analyses showing that n=250 would yield sufficient power to detect differences by race. When counties had <250 beneficiaries, all women from the deficient group(s) were selected. Race/ethnicity-specific sampling weights were obtained by dividing the number of women from a specific county who appeared in the Medicare Denominator file by the race-age-specific number of women residing in that county in excess of 250 who appeared in the sample. The final data set included 11,558 women who received a biopsy between 2005 and 2008, equivalent to a weighted population of 185,153. Claims were identified for diagnostic and screening mammography in 2005-08 as well as a diagnosis of a lump or breast mass at the time of mammography. Algorithms from Smith-Bindman et al. and Fenton et al. were used to distinguish screening from diagnostic mammograms. Positive screens were defined by finding subsequent billing for breast biopsy. Multivariable analysis was used to estimate associations between delays (defined as >21 days from screening to biopsy and >60 days from biopsy to surgery, radiation, or chemotherapy) and age in 2005, race, depression, Charlson co-morbidity scores and US Census region of residence. Results: Among women receiving biopsies, 7703 or 66.7% had 21 or more days between their most recent mammogram and biopsy. After accounting for co-variables, the delay was 20% (OR 1.20, 95% CI 1.09-1.32) more likely among black beneficiaries than whites and 15% (OR 1.15, 95% CI 1.06-1.25) more likely among beneficiaries with at least one co-morbid condition relative to those with none. The number of days waiting for biopsy were inversely associated with yearly increases in age (OR 0.97, 95% CI 0.97-0.98) and diagnosed depression (OR 0.87, 95% CI 0.77-0.97). Overall, 5236 of the women receiving biopsies were diagnosed with breast cancer (weighted population n=84,837). Among these women, 359 (6.9%) did not start treatment within sixty days and an additional 228 (4.4%) did not start treatment within 60 days or before the end study period. Black women were 2.16 times more likely (95% CI 1.72-2.71) to experience > 60 days post-diagnosis before treatment initiation than white women. Also, yearly increases in age were associated with a 3% (OR 1.03, 95% CI 1.01-1.04, p<0.0001) increase in time until treatment regardless of race. Neither co-morbidity nor depression were significantly associated with longer times between biopsy and treatment. Significant geographic differences (p<0.001) were found for both longer times between mammogram and biopsy, and biopsy and treatment initiation. Conclusions: These descriptive data are consistent with the hypotheses that black race, age, and geographic region of residence are independent predictors of longer waits for biopsy and treatment among non-HMO black and white Medicare elderly. Relationship of depression to delays between screening and biopsy warrants further study. Analytic epidemiologic study is needed to test these hypotheses and determine whether they contribute to disparities in breast cancer mortality. Citation Format: Rebecca Selove, Barbara Kilbourne, Maureen Sanderson, Maya Foster, Mary Kay Fadden, Regina Offodile, Baqar Husaini. Time course from screening mammography to biopsy to treatment among black and white, non-HMO Medicare beneficiaries in 2005-2008. [abstract]. In: Proceedings of the Seventh AACR Conference on The Science of Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; Nov 9-12, 2014; San Antonio, TX. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2015;24(10 Suppl):Abstract nr A63.