The association of socioeconomic status and late stage breast cancer in Florida: A spatial analysis using area-based socioeconomic measures

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)


Epidemiology and Public Health

First Committee Member

James D. Wilkinson - Committee Chair


Background. The overall stage specific trends confirm the fact that more and more women are being diagnosed with early stage breast cancer. The success of screening efforts is further confirmed by the decrease in the overall breast cancer mortality rates. However a higher rate of women, living in lower SES, are diagnosed with late stage breast cancer than women living in upper SES areas.Florida's late stage incident breast cancer data from 1989 to 2002 and 2000 block group data from the Census Bureau were imported into the cluster detection software SaTScan to identify areas of excess cancer in the State of Florida. The different areas were analyze to assess any differences among women with late stage breast cancer and socioeconomic status.Methods. A total of 18,683 women diagnosed with regional or distant stage breast cancer residing in 6,361 block groups were imported into SaTScan. After controlling for race and age, 767 block groups were identified as areas with higher than expected incidence of late stage breast cancer cases. Using area-based socioeconomic measures (ABSM's), the Census block group 'ratio of income to poverty' was used as the SES indicator. Based on plurality, each block group was assigned an SES designation of severe poverty, near poverty or non-poverty.Results. Overall, as compared to women living in non-poverty, women living in severe poverty are 4 times more likely to live in areas of higher than expected incidence of late stage breast cancer. SES accounted for approximately 28% of the excess of late stage breast cancer in Florida. Neither marital nor insurance status had a significant effect on these odds ratios.Conclusions. The challenges that face poor and uneducated people are very different from those faced by persons not in poverty. The fact that insurance was not a significant predictor of higher than expected incidence may indicate that access was not an issue. Rather, utilization may be an issue. Utilizing SaTScan results in combination with ABSM can increase the ability of cancer prevention and control professionals to use targeted interventions below the county level, down to the neighborhood area that are culturally/ethnically sensitive.


Health Sciences, Public Health; Health Sciences, Oncology

Link to Full Text


Link to Full Text