Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Jan;29(1):167-183.
doi: 10.1007/s10552-017-0980-1. Epub 2017 Dec 8.

Characterizing the Neighborhood Obesogenic Environment in the Multiethnic Cohort: A Multi-Level Infrastructure for Cancer Health Disparities Research

Affiliations
Free PMC article

Characterizing the Neighborhood Obesogenic Environment in the Multiethnic Cohort: A Multi-Level Infrastructure for Cancer Health Disparities Research

Shannon M Conroy et al. Cancer Causes Control. .
Free PMC article

Abstract

Purpose: We characterized the neighborhood obesogenic environment in the Multiethnic Cohort (MEC) by examining the associations of obesity with attributes of the social and built environment, establishing a multi-level infrastructure for future cancer research.

Methods: For 102,906 African American, Japanese American, Latino, and white MEC participants residing predominately in Los Angeles County, baseline residential addresses (1993-1996) were linked to census and geospatial data, capturing neighborhood socioeconomic status (nSES), population density, commuting, food outlets, amenities, walkability, and traffic density. We examined neighborhood attributes and obesity (body mass index ≥ 30 kg/m2) associations using multinomial logistic regression, adjusting for individual-level (e.g., demographics, physical activity, and diet) and neighborhood-level factors.

Results: NSES was associated with obesity among African Americans, Latinos, and whites (p-trend ≤ 0.02), with twofold higher odds (adjusted odds ratios, 95% confidence intervals) for living in the lowest versus highest quintile among African American women (2.07, 1.62-2.65), white men (2.11, 1.29-3.44), and white women (2.50, 1.73-3.61). Lower density of businesses among African American and white women and lower traffic density among white men were also associated with obesity (p-trends ≤ 0.02).

Conclusions: Our study highlights differential impacts of neighborhood factors across racial/ethnic groups and establishes the foundation for multi-level studies of the neighborhood context and obesity-related cancers.

Keywords: Neighborhood environment; Obesity; Race/ethnicity; Socioeconomic status.

Conflict of interest statement

Conflicts of Interest: None

Figures

Figure 1
Figure 1
Odds ratios and 95% confidence intervals for being obese [body mass index (BMI) ≥ 30] compared to normal weight (BMI < 25) among men (circles) and women (squares) in the Multiethnic Cohort residing in California at baseline (1993–1996). Models adjusted for age, marital status, BMI at age 21, smoking and cigarette pack years, alcohol intake, education, moderate and vigorous activity, diet intake (red meat, processed red meat, vegetables and fruits, dairy products, total calories), neighborhood attributes (all variables listed, population density, restaurant environment index, retail food environment index, traffic density), and clustering effect of block group. Neighborhood socioeconomic status (SES) and commute patterns are U.S. Census block group-level measures, with quintiles (Q) based on distribution for block groups in Los Angeles County. Businesses/parks are within walking distance of residence (1.6 km pedestrian network), with categories based on study participant distribution. Symbol size proportional to effect size. Odds ratios on the natural logarithmic scale.

Similar articles

See all similar articles

Cited by 5 articles

Feedback