The evolution of breast cancer screening in the Medicare population: Clinical and economic implications.

Abstract

1595 Background: Breast screening has evolved as newer approaches to mammography, ultrasound, and MRI have diffused into clinical practice. The use of these technologies and their impact on screening-related costs and outcomes remain undefined, particularly among older women. Methods: Using the Surveillance Epidemiology and End Results – Medicare linked database, we identified women aged 66 and older without a diagnosis of breast cancer. We constructed two cohorts (2001 vs. 2006) and followed each for two years. We assessed changes in imaging technology, screening-related costs (defined as costs for screening and subsequent imaging and testing, adjusted to 2009 USD), and stage at diagnosis between the two cohorts. Results: There were 136,845 women in the 2001-2002 (earlier) cohort and 137,733 in the 2006-2007 (later) cohort. The mean age was 76.9 and 77.2 respectively, (p<.001). The proportion of women receiving any screening mammogram was 42.5% in the earlier cohort and 43.4% in the later cohort, (p<.001). The use of digital mammography for screening increased from 2.2% to 15.0%, (p<.001). The use of any computer aided detection (CAD) increased from 3.2% to 29.3% (p<.001). MRI use increased from 0.03% to 0.2%, and ultrasound use from 4.0% to 4.5% (p <.001 for both). Average screening-related cost increased 31%, from $101 to $132 (p<.001). There was no significant difference in early stage at diagnosis over time (58.1% of women were in situ/stage I in early period vs. 57.2% in later period, p=.65). Conclusions: The use of digital mammography and CAD increased substantially between 2001 and 2007, contributing to a 31% increase in screening-related costs for women in the Medicare program. The increased cost of screening and downstream testing must be evaluated in context of an absence of benefit in terms of stage at diagnosis.