Varying levels of resources and amenities across neighborhoods may plausibly shape health-related behaviors, and thereby may affect individual disease outcomes including colorectal cancer risk. Yet, few studies have examined the specific relation between neighborhood socioeconomic status (SES) and colorectal cancer incidence, or tested for potential behavioral risk factor mediators, with no studies to date being multilevel or prospective in design. Furthermore, while theory and empirical evidence might suggest the differential uptake of health behaviors within neighborhoods according to selected individual characteristics, such potential interactions have not been previously explored for the association and possible behavioral pathways between neighborhood SES and incident colorectal cancer. The primary goal of our project is to address these knowledge gaps, by using secondary data from the Harvard-based Nurses' Health Study, a well-established, large, prospective cohort of women, while applying statistical methods in combination to enhance validity. The central hypothesis is that neighborhood socioeconomic environments influence several individual-level health behaviors that are risk factors for colorectal cancer, and thereby may affect colorectal cancer risk. These associations are also hypothesized to be modified by one's age, educational attainment, and levels of social networks and social participation. The study's Specific Aims, which apply multilevel models to path analysis and discrete-time survival analysis, and allow for time-varying covariates, are to: 1) estimate the multilevel association between SES of neighborhood of residence and individual colorectal cancer risk, controlling for other neighborhood- and individual-level characteristics; 2) evaluate the extent to which the association between neighborhood SES and individual colorectal cancer risk is mediated by behavioral risk factors; and 3) determine whether the associations between neighborhood SES and individual behavioral risk factors and colorectal cancer risk are modified by one's age, educational attainment, and/or levels of social networks and social participation. Our project is innovative in that it will yield novel empirical evidence regarding the presence of contextual effects of neighborhood SES on individual colorectal cancer risk (applying a multilevel analytic framework); will utilize path analysis to parse out the plausible simultaneous mediating roles of behavioral risk factors for colorectal cancer; and will test for potential effect modifiers of these relations. By contributing to the scientific knowledge base on the contextual determinants of and behavioral risk factor mediators for colorectal cancer incidence, this project could potentially identify more upstream/structural points of intervention to attenuate the incidence of colorectal cancer. The study's efforts may thus translate into more effective reductions in the occurrence of health disparities in colorectal cancer, and reductions in the burden of colorectal cancer in the United States. ? ? ?
| Kim, Daniel; Masyn, Katherine E; Kawachi, Ichiro et al. (2010) Neighborhood socioeconomic status and behavioral pathways to risks of colon and rectal cancer in women. Cancer 116:4187-96 |
