The Feasibility of Using Social Networks to Understand the Causal Pathways of Cancer Disparities

Document Type

Conference Proceeding

Publication Date

12-2013

Publication Title

Proceedings of the Sixth American Association for Cancer Research (AACR) Conference: The Science of Cancer Health Disparities

DOI

10.1158/1538-7755.DISP13-A70

Abstract

Background: While it is recognized that multiple factors interact at different socioecological levels to influence cancer disparities, many of the proposed solutions to eliminating disparities have been narrowly focused on individual level factors. Social networks have been shown to influence health outcomes through various pathways, including shared social capital, social and cultural norms, risky behaviors, and the transmission of infectious diseases. Current research has shown how social networks influence health behavior; however there are still some critical gaps in this understanding. There is lack of data about social networks and the context in which individuals make decisions about their medical care and health behavior. The purpose of this study is to explore how certain sociocultural conditions influence the composition and nature of social networks structure; and how in turn these networks serve as mediating structures for certain psychosocial mechanisms that influence decisions about screening health behaviors.

Methods: Social network analysis of egocentric networks was conducted to visualize the social networks of 65 African American, Latino, and Caucasian men and women. Descriptive statistics (e.g. frequencies, proportions, and means) will be produced to describe sample characteristics in terms of screening behaviors performed/not performed, network sizes, and flow of information. Bivariate statistics (e.g., Chi-square, Mc Nemar, t-tests, ANOVAs) will be computed to make comparisons on these variables by racial/ethnic group membership. Multivariate statistics will be computed to evaluate the impact of social network characteristics on screening practices, health decisions, and access to health care.

Results: Our analysis allows us to describe the structural and compositional network characteristics that are related to prostate, breast, and colorectal cancer screening, and explore whether there is variation in the ways that social network characteristics influence cancer screening behaviors. Our analysis explores the structural and compositional measures of the network (e.g., size, density, strength of ties) and associations among sociocultural conditions, the composition and nature of social networks structure and how in turn these networks serve as mediating structures for certain psychosocial mechanisms that may influence decisions about cancer screening health behaviors.

Conclusion: Social network analysis has the potential to provide measureable data to help characterize how individuals interact with and access cancer screening services and other key variables that impact health risk behaviors. Before evidenced-based, multi-level interventions can be developed to improve prostate, breast, and colorectal cancer screening, more scientific knowledge is needed about how interpersonal factors shape these screening behaviors.

Share

COinS