The proposed project is a longitudinal, observational study incorporating both area and individual level variables to assess contextual effects of the neighborhood on colorectal cancer incidence and stage of diagnosis. Colorectal cancer (CRC) is the second highest cause of cancer related deaths in the US. Minority populations are disproportionately affected and are known to have adverse outcomes. Although screening guidelines exist, a large proportion of cases are still being diagnosed at late stages. Since CRC has a long latency period, exposure to a deleterious social environment along with other risk factors portend poor prognosis. For this study, a conceptual model depicting the varying levels of exposure to risk factors, in multiple domains, with differing pathways in access to resources, resulting in a certain level of community health was adapted for use. In addition, secondary analysis of Cancer Registry and Census data circumvents the prohibitive costs of primary data collection, particularly for small research projects.
The specific aims of this multilevel and ecological study are to examine relationships between contextual variables and the cases of CRC at block group and census tract level, early/late stage diagnosis at multiple levels simultaneously and also to examine the geographic variation in spatial and temporal trends in CRC incidence and late stage diagnosis for the years 1998-2003 in Houston, using Poisson models in SaTScan. Census derived readily available social environment variables such as occupational class, education level, poverty, overcrowding, linguistic isolation etc. will be the independent variables. CRC cases and early/late stage diagnosis will be analyzed as the dependent variables. Addressing disparities in cancer etiology is a high priority area for National Institute of Health and National Cancer Institute and secondary analysis of existing datasets is an important requirement of the R03 mechanism and the program announcement. As the local health department in the fourth largest city in the country, we are responsible for identifying geographic areas of need and lack of access to resources and developing appropriate intervention strategies to address them. We anticipate that our results will be easily translatable to other large cities, similar to ours, with diverse populations. This study will help us better understand cancer disparities and tease out the socio-environmental factors that play a role in adverse outcomes using ecological, multilevel and geospatial analytic techniques.
Project Narrative Public Health Relevance: Our study will result in greater understanding of the distribution, etiology, spatio-temporal patterns of colorectal cancer (CRC) in Houston, the fourth largest city in the country. It will also identify areas of the city with greater than expected rates. This will help ascertain important social, environmental and behavioral risk factors that play a role in late stage diagnosis of this disease. This study will not only help inform science, but also community health professionals in designing programs to address some of the modifiable factors, in settings similar to ours. In addition, the policy implications at various levels of the government such as the City and State level and translation of research into practice by enhancing health promotion, health literacy and CRC screening in high risk areas are some of the potential benefits that will be achieved.
