Abstract
Background: Although social and built environment attributes are associated with obesity, limited research has characterized their underlying relationships and differential effects across race/ethnicity. Our study developed composite measures of the neighborhood obesogenic environment for California and investigated their associations with obesity and breast, colorectal, and prostate cancer risk within the Multiethnic Cohort Study (MEC).
Methods: Principal component factor analysis (PCFA) was conducted to characterize block group-level obesogenic environments in California. Nine neighborhood attributes derived from geospatial data were selected based on their potential to contribute to obesity: population density, traffic density, commuting by car/motorcycle, numbers of businesses, parks, recreational facilities, restaurants, retail food outlets, and street connectivity. Multinomial logistic regression was used to examine associations between neighborhood obesogenic factors and obesity at baseline by sex and race/ethnicity (African American (AA), Japanese, Latino and White) in the California component of the MEC (n=102,906; obese=24,701). Multivariable Cox proportional hazards regression was used to examine associations between these neighborhood obesogenic factors at baseline and risk of breast (n=48,247; cases=2,341), colorectal (n=81,187, cases=1,973), and prostate (n=41,565; cases=4,650) cancers in the California MEC, overall and stratified by race/ethnicity.
Results: PCFA identified four neighborhood obesogenic factors: 1) “urban environment” (high population density, high street connectivity, low commuting by car/motorcycle, high traffic density); 2) “mixed land use” (high number of recreational facilities and businesses); 3) “unhealthy food outlets” (high ratios of unhealthy to healthy for restaurants, and for retail food outlets); 4) “parks” (any parks). For males, less urban environments among AAs and Whites; lower mixed land use among Latinos; and any parks among AAs were associated with increased odds of obesity. For females, more unhealthy food outlets among Whites was associated with increased odds of obesity. For breast cancer, more unhealthy food outlets for all racial/ethnic groups combined and lower mixed land use in Latinos were associated with increased risk of disease. For colorectal cancer, some unhealthy food outlets for all groups combined, higher mixed land use in Japanese males, no parks in AA males, and lower mixed land use in White females were associated with increased risk of disease. For prostate cancer, higher mixed land use in Whites was associated with increased risk of disease and more unhealthy food outlets in AAs was associated with increased risk of aggressive disease.
Conclusion: The associations between the four neighborhood obesogenic factors with obesity and cancer risk varied across sex and race/ethnicity, demonstrating the complex relationships involved in cancer disparities.
Citation Format: Salma Shariff-Marco, Li Tao, Juan Yang, Alison Canchola, Christina Clarke, Yurii Shvetsov, Shannon Conroy, Song-Yi Park, Andrew Hertz, Cheryl Albright, Laurence Kolonel, Kristine Monroe, Loic Le Marchand, Lynne Wilkens, Scarlett Lin Gomez, Iona Cheng. Development of neighborhood “obesogenic factors” and its applicability to obesity and cancer studies in diverse populations: The Multiethnic Cohort Study. [abstract]. In: Proceedings of the Ninth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2016 Sep 25-28; Fort Lauderdale, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2017;26(2 Suppl):Abstract nr B65.