ObesityTrends in Tract-Level Prevalence of Obesity in Philadelphia by Race-Ethnicity, Space, and TimeQuick, Harrisona; Terloyeva, Dinaa; Wu, Yaxina; Moore, Karib; Diez Roux, Ana V.a,bAuthor Information From the aDepartment of Epidemiology and Biostatistics, Drexel University, Philadelphia, PA bUrban Health Collaborative, Drexel University, Philadelphia, PA. Submitted February 16, 2019; accepted September 27, 2019. This research was supported by the Drexel University Urban Health Collaborative pilot award program. The authors report no conflicts of interest. Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com). The data used in the study were collected by and obtained from Public Health Management Corporation Correspondence: Harrison Quick, Department of Epidemiology and Biostatistics, Drexel University, 3215 Market Street, Philadelphia, PA 19104. E-mail: email@example.com. Epidemiology: January 2020 - Volume 31 - Issue 1 - p 15-21 doi: 10.1097/EDE.0000000000001118 Buy SDC Metrics Abstract The growing recognition of often substantial neighborhood variation in health within cities has motivated greater demand for reliable data on small-scale variations in health outcomes. The goal of this article is to explore temporal changes in geographic disparities in obesity prevalence in the City of Philadelphia by race and sex over the period 2000–2015. Our data consist of self-reported survey responses of non-Hispanic whites, non-Hispanic blacks, and Hispanics from the Southeastern Pennsylvania Household Health Survey. To analyze these data—and to obtain more reliable estimates of the prevalence of obesity—we apply a Bayesian model that simultaneously accounts for spatial-, temporal-, and between-race/ethnicity dependence structures. This approach yields estimates of the obesity prevalence by age, race/ethnicity, sex, and poverty status for each census tract at all time-points in our study period. While the data suggest that the prevalence of obesity has increased at the city-level for men and women of all three race/ethnicities, the magnitude and geographic distribution of these increases differ substantially by race/ethnicity and sex. The method can be flexibly used to describe and visualize spatial heterogeneities in levels, trends, and in disparities. This is useful for targeting, surveillance, and etiologic research. Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.