Less frequent utilization of post-diagnosis mammograms explains the shorter survival of Latina breast cancer patients

Research Objective: Latina women diagnosed with breast cancer are 20% more likely to die in five years when compared to White women of the same age and with the same stage at diagnosis. The literature has noted this disparity for decades but no research has specifically addressed possible reasons for this observation. Furthermore, reports of mammogram utilization in the US Latina population are conflicting; cross-sectional studies using the National Health Interview survey report Latinas use less mammograms and the Behavioral Risk Factor Surveillance System Studies report no difference in mammogram use by ethnicity. This study will use a stronger longitudinal cohort study design and examining time-to-utilization of health care variables identify utilization practices that have the greatest potential to eliminate the long observed survival disadvantage in Latina breast cancer patients. Methods: Women age 66 and older diagnosed with stage I and II breast cancer between 1992 and 2000 were followed until 2005 using the Surveillance Epidemiology and End Results- Medicare registry (n=47,593). We assessed the impact of time-to-mammograms and time-to-clinician visits by a validated definition of ethnicity on survival time. We controlled for “a priori” variables including age, stage at diagnosis, comorbidity, marital status, median income a language variable and initial cancer treatment. We used epidemiological and statistical methods to assess the impact of utilization practices on the Latina survival disadvantage. Principal Findings: Although receipt of initial quality cancer care is equally important for White and Latina women (hazard ratio and 95% confidence interval (HR95%CI)=0.85 (0.80,0.91), receipt of yearly mammograms was the most significant factor contributing to the Latina survival disadvantage (HR95%CI)=0.14(0.1,0.2) . Epidemiological and statistical analyses showed utilization of timely mammograms independently and entirely explain the mortality disparity between Latina and White women diagnosed with early breast cancer and this result was stable with and without other confounding and interaction variables. Conclusions: These results indicate a driving factor for the breast cancer specific survival disadvantage in Latina women is due to less frequent utilization of post-diagnosis mammograms resulting in undetected cancer recurrence or new cancers. Difficulties in obtaining quality care at the patient level might be addressed through culturally sensitive navigation programs to improve familiarity with Medicare and inclusion of family members. Obstacles for patient's referral of follow-up care at the institutional level might be addressed by institutional quality improvement efforts in multidisciplinary coordination.