Evaluating Geographic Cancer Disparities in Baltimore City, Maryland

Embargo until
Date
2016-04-04
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Research and policy continue to hold cancer as a top priority in the United States due to its public health burden. As the aging population increases, it is an issue that will likely continue to grow in scope. However, there has been less focus on the complex interaction between social conditions of environments and disparate health outcomes. This dissertation’s three aims seek to evaluate the geographic distribution of cancer incidence, cancer mortality, and cancer facilities in Baltimore City, MD as well as the subsequent neighborhood-level correlates that are associated with the observed variability. The first aim utilized a cohort of Baltimore City female residents (n= 4,966) that were diagnosed with breast, cervical, or colorectal cancer between the ages of 21 to 74 years from 2000 to 2010. The Maryland Cancer Registry provided this data along with the residential address of each cancer case at the time of diagnosis, which allowed for the calculation of incidence rates and cancer stage by neighborhood for each of the three cancer sites. From the cluster detection methods utilized, geographic variation was observed for both outcomes, and it varied by cancer site. The community characteristics explaining this variability also depended on the cancer type. Using a similar methodology, the second aim utilized the deaths (n= 1,765) that had accrued from the previous aim’s cohort of Baltimore City female residents that had developed breast, cervical, or colorectal cancer from 2000 to 2010. The residential address at diagnosis for each cohort member was geocoded to help facilitate the identification of whether certain neighborhoods had higher cancer mortality than others and whether community-level characteristics played a role in that difference. Similar as to what was observed for cancer incidence, there were clear differences as to what sections of Baltimore City had a greater burden of mortality. The high burden areas shifted depending on which of the three cancer sites was being evaluated. There was also observed variability in terms of what local-level variables were significantly associated with the geographic aggregation of mortality. The final aim took the approach of evaluating the location of facilities that provided breast and colorectal cancer services. The addresses for these resources were obtained through publicly available data. This information was then used to construct two measures (service rate and service density) that approximated physical access for each neighborhood. The findings demonstrated that most of the screening and treatment facilities were clustered in the downtown area of Baltimore City. Overall, the dissertation provided evidence that the evaluation of health outcomes and resources should entail an understanding of the social context in which they occur. The initial findings of the three aims prompt additional research to better hone in on the neighborhood drivers of disease as well as to further investigate neighboring communities with distinct risk profiles. The conclusions reached have the potential to serve as actionable items for policy and resource allocation.
Description
Keywords
cancer, disparities, geospatial, Baltimore City, epidemiology
Citation