The Nova Scotia Community Cancer Matrix: A geospatial tool to support cancer prevention

Nathalie Saint-Jacques; Patrick E Brown; Judy Purcell; Daniel G Rainham; Mikiko Terashima; Trevor J B Dummer

doi:10.1016/j.socscimed.2023.116038

The Nova Scotia Community Cancer Matrix: A geospatial tool to support cancer prevention

Soc Sci Med. 2023 Aug:330:116038. doi: 10.1016/j.socscimed.2023.116038. Epub 2023 Jun 21.

Authors

Nathalie Saint-Jacques¹, Patrick E Brown², Judy Purcell³, Daniel G Rainham⁴, Mikiko Terashima⁵, Trevor J B Dummer⁶

Affiliations

¹ NSH Cancer Care Program, Bethune Building, 1276 South Park St, Halifax, NS, Canada; Healthy Populations Institute, Dalhousie University, 1318 Robie St., Halifax, NS, Canada. Electronic address: nathalie.st-jacques@nshealth.ca.
² Department of Statistical Science, University of Toronto, Sidney Smith Hall, 100 St. George Street, Toronto, ON, Canada. Electronic address: patrick.brown@utoronto.ca.
³ NSH Cancer Care Program, Bethune Building, 1276 South Park St, Halifax, NS, Canada. Electronic address: judith.purcell@nshealth.ca.
⁴ School of Health and Human Performance, Dalhousie University, 5981 University Avenue, Halifax, NS, Canada; Healthy Populations Institute, Dalhousie University, 1318 Robie St., Halifax, NS, Canada. Electronic address: daniel.rainham@dal.ca.
⁵ School of Planning, Dalhousie University, O'Brien Hall, 5217 Morris St., Halifax, NS, Canada. Electronic address: mikiko.terashima@dal.ca.
⁶ School of Population and Public Health, University of British Columbia, 226 East Mall, Vancouver, BC, Canada. Electronic address: trevor.dummer@ubc.ca.

PMID: 37390806
DOI: 10.1016/j.socscimed.2023.116038

Abstract

Globally, cancer is a leading cause of death and morbidity and its burden is increasing worldwide. It is established that medical approaches alone will not solve this cancer crisis. Moreover, while cancer treatment can be effective, it is costly and access to treatment and health care is vastly inequitable. However, almost 50% of cancers are caused by potentially avoidable risk factors and are thus preventable. Cancer prevention represents the most cost-effective, feasible and sustainable pathway towards global cancer control. While much is known about cancer risk factors, prevention programs often lack consideration of how place impacts cancer risk over time. Maximizing cancer prevention investment requires an understanding of the geographic context for why some people develop cancer while others do not. Data on how community and individual level risk factors interact is therefore required. The Nova Scotia Community Cancer Matrix (NS-Matrix) study was established in Nova Scotia (NS), a small province in Eastern Canada with a population of 1 million. The study integrates small-area profiles of cancer incidence with cancer risk factors and socioeconomic conditions, to inform locally relevant and equitable cancer prevention strategies. The NS-Matrix Study includes over 99,000 incident cancers diagnosed in NS between 2001 and 2017, georeferenced to small-area communities. In this analysis we used Bayesian inference to identify communities with high and low risk for lung and bladder cancer: two highly preventable cancers with rates in NS exceeding the Canadian average, and for which key risk factors are high. We report significant spatial heterogeneity in lung and bladder cancer risk. The identification of spatial disparities relating to a community's socioeconomic profile and other spatially varying factors, such as environmental exposures, can inform prevention. Adopting Bayesian spatial analysis methods and utilizing high quality cancer registry data provides a model to support geographically-focused cancer prevention efforts, tailored to local community needs.

Keywords: BYM; Cancer; Environment; Prevention; Small-area disease mapping; Spatiotemporal autoregressive analyses socioeconomic status.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Bayes Theorem
Delivery of Health Care*
Humans
Nova Scotia / epidemiology
Risk Factors
Urinary Bladder Neoplasms* / epidemiology