The relationship between patient and tumor characteristics, patterns of breast cancer care, and 5-year survival among elderly women with incident breast cancer. Breast Cancer Res Treat 2018 Sep;171(2):477-488
Date
06/06/2018Pubmed ID
29869776DOI
10.1007/s10549-018-4837-4Scopus ID
2-s2.0-85048054073 (requires institutional sign-in at Scopus site) 8 CitationsAbstract
PURPOSE: To examine the relationship between patient and tumor characteristics, patterns of breast cancer care, and 5-year survival among a population-based cohort of elderly women with incident breast cancer, with a special focus on identifying sources of socioeconomic (SES) disparities in outcomes.
METHODS: We identified women with newly diagnosed breast cancer in 2006-2009 from the Surveillance and Epidemiology End Result study linked with Medicare claims. A Classification and Regression Tree (CART) model was applied to 13 individual indicators of neoadjuvant and adjuvant breast cancer treatment, tumor characteristics, and patient sociodemographic variables to identify patterns with the greatest discriminant value in predicting 5-year survival. We subsequently examined the extent to which these patterns varied by the patient's SES.
RESULTS: Survival probabilities associated with the 18 unique CART-identified patterns ranged from 22 to 87%. The number of positive axillary nodes was the best single discriminator between high and lower survival outcomes. The most common discriminant factor among patterns with poor (< 25%) survival was the absence of radiation treatment, followed by the presence of comorbidities, tumor size > 2 cm, and no breast surgery. Relative to high SES women, poor women were nearly four times (12.3% vs. 3.2%, p < 0.001) as likely to be classified in the pattern associated with worse survival, and less likely (31.7% vs. 52.9%, p = 0.04) to receive the pattern associated with the greatest survival.
CONCLUSIONS: Greater adoption of effective patterns of care could improve survival of elderly women with incident breast cancer overall, and reduce SES disparities therein.
Author List
Kong AL, Nattinger AB, McGinley E, Pezzin LEAuthors
Amanda L. Kong MD, MS Professor in the Surgery department at Medical College of WisconsinEmily L. McGinley Biostatistician III in the Center for Advancing Population Science department at Medical College of Wisconsin
Ann B. Nattinger MD, MPH Associate Provost, Professor in the Medicine department at Medical College of Wisconsin
Liliana Pezzin PhD, JD Director, Professor in the Institute for Health and Humanity department at Medical College of Wisconsin
MESH terms used to index this publication - Major topics in bold
Age FactorsAged
Aged, 80 and over
Breast Neoplasms
Cause of Death
Comorbidity
Female
Humans
Incidental Findings
Medicare
Mortality
Neoplasm Grading
Neoplasm Metastasis
Neoplasm Staging
Practice Patterns, Physicians'
Prognosis
SEER Program
Socioeconomic Factors
United States
