Related Experiment Videos
Breast cancer detection: maps of 2 San Francisco Bay area counties
S Selvin1, D W Merrill, C Erdmann
1Department of Biostatistics, School of Public Health, University of California, Berkeley 94720-7360, USA.
Objectives:
This study describes the incidence of late-stage and in situ breast cancer among White women, using specialized mapping techniques that reflect incidence adjusted for the population at risk, and applies these maps to characterize areas with high and low risk of breast cancer.
Methods:
Data from the Surveillance, Epidemiology, and End Results (SEER) database and the US Census Bureau were used to study the geographic distribution of breast cancer at the census-tract level in 2 San Francisco Bay Area counties for the years 1978 through 1982. Sociodemographic characteristics of areas with high and low incidence of the stage-specific disease were compared by means of a linear discriminant function.
Results:
For late-stage breast cancer, the most important variables in discriminating high-risk from low-risk areas were college education, percentage of residents over age 65, and median income. The strongest ecologic indicators of high risk for in situ breast cancer were median income and percentage unemployed.
Conclusions:
This study demonstrates the usefulness of census tracts and sociodemographic measures of income and education in describing in situ and late-stage breast cancer.