Personal and neighbourhood indicators of social deprivation, and risk of prostate cancer
PI: Marie-Elise Parent
Co-Invest. : Tracie Barnett, Amélie Quesnel-Vallée (McGill), Yan Kestens (UofM), Belinda Nicolau (McGill), Andrea Benedetti (McGill), Geetanjali Datta (UofM)
Problem: Social isolation has been defined as a lack of participation in social relationships and/or lack of interaction with others and/or with society at large. Socially deprived individuals experience poorer health outcomes, including greater cancer-specific mortality and poorer prognosis for some cancer sites, including the prostate. Social deprivation has also been linked to cancer. Surprisingly, hardly any research has evaluated whether it relates to prostate cancer incidence more specifically. As social environments can facilitate or impede resource sharing, reinforce health beliefs as well as health behaviors, social environments would be expected to influence the entire prostate cancer spectrum, including its incidence and prevention.
Objective: To explore, for the first time, the association between several demonstrated or suspected indicators of social deprivation (or lack thereof), and the risk of developing prostate cancer. These will be assessed both at the personal and neighbourhood levels, and at several time points over the lifecourse.
Methodology: The proposed analyses will be conducted on a vast, already assembled Canadian dataset, which will be augmented with additional relevant data. Between 2002 and 2015, the Canadian Cancer Society and the Cancer Research Society funded the largest ever and most comprehensive population-based case-control study of prostate cancer to elucidate its causes. It comprises 1,933 cases prostate cancer cases, including 538 aggressive cancers, and 1,994 population controls.
As part of extensive in-person interviews, trained interviewers elicited details on a wide range of personal socio-economic, demographic, environmental and lifestyle factors. Some of these factors, such as being a widow or living alone, are recognized indicators of social deprivation. Many others, including the family size and structure, can be suspected to reflect the potential for social interactions and have never been studied with respect to prostate cancer risk. Detailed lifetime job descriptions (n=16,095) will inform us on workplace characteristics and their social environment, as well as on the influence of job mobility and work schedules. Lifetime residential addresses (n=13,073) have been geocoded and will be linked to neighbourhood-level indicators of social deprivation. Lifetime addresses will also be used to document the frequency of changes in residential environment and the distance between moves.
Statistical analyses will describe the association between prostate cancer on the one hand, and the various personal and neighbourhood-based indicators of social deprivation, using unconditional logistic regression. Polytomous models will seek out associations by cancer aggressiveness. For factors for which we have information over several time points over the lifetime, we will apply lifecourse analytical approaches to test critical period vs. accumulation models. The availability of detailed information on lifestyle factors at the individual level will enable us to evaluate potential confounders and mediating factors. The role material deprivation, as well as of prostate cancer screening practices, in observed associations will also be investigated.
Significance: In light of the large prostate cancer burden and complete absence of strong leads for prevention, it is essential to cover new grounds to identify factors amenable to intervention. Building on a recently assembled Canadian database, this study covers several unexplored areas and will provide entirely novel evidence on the role of social deprivation in prostate cancer incidence. Should an association be demonstrated, this would represent a major advancement towards cancer prevention. This research can indeed provide insights into prostate cancer disease processes, identify vulnerable populations, and generate results with translational impact of relevance to interventionists and policy makers.