Recorded presentation

Stephen Taplin, MD¹, Laura Ichikawa², Marianne Ulcickas-Yood, DSc³, M. Michele Manos, PhD, MPH⁴, Ann M. Geiger, PhD⁵, Sheila Weinman, PhD⁶, Joyce Gilbert, MPH⁷, Judy Mouchawar, MD, MSPH⁸, Wendy Leyden, MPH⁴, Robin Altaras², Kevin Beverly², Deborah Casso, MPH², Emily Westbrook², Jane Zapka, ScD⁹, and William E. Barlow, PhD². (1) Applied Research Program, National Cancer Institute, 6130 Executive Blvd., EPN 4094, Rockville, MD 20852, 301-402-1483, taplins@mail.nih.gov, (2) Center for Health Studies, Group Health Cooperative, 1730 Minor Ave. Suite 1600, Seattle, WA 98101, (3) Epidemiology, Yale, 275 Blue Trail, Hamden, CT 06518, (4) Division of Research, Kaiser Permanente, 2000 Broadway, Oakland, CA 94612, (5) Research and Evaluation Department, Kaiser Permanente Southern California, 100 South Los Robles, Pasadena, CA 91188, (6) Epidemiology, Center for Health Research, Kaiser Permanente Northwest, 3800 N. Interstate Avenue, Portland, CT 97227-1098, (7) Kaiser Permanente Hawaii, 531 Ohohia Street, Honolulu, HI 96819, (8) Research and Development, Kaiser Permanente, Colorado, 10350 E. Dakota Avenue, Denver, CO 80231, (9) Division of Preventive & Behavioral Medicine, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655

Context: Late-stage breast cancers should not occur among women with access to screening mammography. Objective: To determine where improvements in screening implementation could decrease late-stage cancer incidence. Design & setting: Case-control study using chart and automated data on late-stage (metastatic and/or tumor size > 3 cm) (cases) and early-stage breast cancers (controls) diagnosed 1995-1999 in 7 health plans serving >8.2 million people. We matched cases and controls on health plan, age, and diagnosis date from among women enrolled >3 years prior to their cancer diagnosis. Main outcome measures: One of three screening implementation breakdown categories based on healthcare 12-36 months prior to diagnosis: 1) no screening mammogram (absence of screening), 2) negative earliest screening mammogram (absence of detection), and 3) positive earliest screening mammogram (potential breakdown in follow-up). We compared the proportion of cases and controls in each group and estimated the likelihood of late-stage as a function of age, race, absence of screening, and refusal-of-breast-care. Results: The respective distributions of “absence of screening,” “absence of detection” and “potential breakdown in follow-up” differed among 1347 cases (52.1%, 39.5%, and 8.4%) and 1347 controls (33.4%, 56.9%, and 8.8%) (p = 0.03). Late-stage cases were more likely to have refused breast-care prior to the audit period (7.9% vs. 2.9%, p 0.0001). Women with late-stage cancers residing in poorer or less educated areas were more likely to be unscreened by mammography. Conclusions: Increasing screening recruitment and improving mammography detection could greatly reduce the risk of late-stage breast cancer.

Learning Objectives:

• Participants will learn the reasons why late-stage breast cancers still occur among women with access to screening mammography.
• Participant will be able to discuss where improvements in screening implementation could decrease late-stage cancer incidence.

Keywords: Breast Cancer Screening, Access to Care

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

Recorded presentation