Frequency of late-stage breast cancer diagnoses despite high mammography screening rates in an HMO

Abstract

1526 Background: Kaiser Permanente Southern California (KPSC) led the nation in screening women for breast cancer (BCa) with a mammography rate of nearly 90% in 2007 according to 2008 Healthcare Effectiveness Data and Information Set (HEDIS) measures. Despite successes in improving screening rates in this health plan that serves 3+ million diverse members, the percentage of women diagnosed with late stage BCa (stage III, IV) remained stable, varying from 12.9% (N∼323) in 2003 to 10.8% (N∼270) in 2007. To identify patient and health care factors associated with late stage diagnosis and the impact of its enhanced screening implementation guidelines, KPSC undertook this study. Methods: This cross-sectional study included a cohort of 10,580 BCa patients from 2003–2007. We compared women diagnosed with late stage disease versus those with early stage disease (stages I, II). P values (2-sided) were based on the chi-square distribution. Adjusted odds ratios and 95% confidence intervals were estimated using unconditional logistic regression. Results: Factors that were positively associated with late stage diagnosis in the univariate analyses included age, lack of recent mammography screening, worse tumor features, 80+ years of age, minority race, lower geocoded household income, increased healthcare visits, and use of Pap testing (P < 0.01 for all variables). Factors significantly associated with late stage diagnosis in the multivariate model included only lack of recent mammography screening (OR = 1.35, 95% CI: 1.14–1.58) and worse tumor features including high grade (grade 3, OR = 2.58, 95% CI: 1.96–3.40), positive lymph nodes (OR = 53.49, 95% CI: 39.90–71.72), and HER-2+ tumors (OR = 1.40, 95% CI: 1.13–1.72). Conclusions: Targeting older women, those with lower utilization, and women who did not have a recent mammogram may help further lower the prevalence of late stage diagnoses. However, given the extent of the health plan's previous efforts to enhance BCa screening rates, a ceiling effect may limit additional benefit. Additional efforts to decrease the rate of advanced tumor stage at diagnosis may include improving interpretation of mammograms or earlier detection of aggressive tumors by enhanced BRCA genetic testing. No significant financial relationships to disclose.