Spatial Clusters of County-Level Diagnosed Diabetes and Associated Risk Factors in the United States



Sundar S. Shrestha*, Karen A. Kirtland, Theodore J. Thompson, Lawrence Barker, Edward W. Gregg, Linda Geiss
U.S. Centers for Disease Control and Prevention, Division of Diabetes Translation, 4770 Buford Highway, MS K-10, Atlanta, GA 30341, USA.


Article Metrics

CrossRef Citations:
0
Total Statistics:

Full-Text HTML Views: 255
Abstract HTML Views: 203
PDF Downloads: 119
Total Views/Downloads: 577
Unique Statistics:

Full-Text HTML Views: 155
Abstract HTML Views: 125
PDF Downloads: 98
Total Views/Downloads: 378



© S. Shrestha et al.

open-access license: This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

* Address correspondence to this author at the U.S. Centers for Disease Control and Prevention, Division of Diabetes Translation, 4770 Buford Highway, MS K-10, Atlanta, GA 30341, USA.


Abstract

Introduction:

We examined whether spatial clusters of county-level diagnosed diabetes prevalence exist in the United States and whether socioeconomic and diabetes risk factors were associated with these clusters.

Materials and Methods:

We used estimated county-level age-adjusted data on diagnosed diabetes prevalence for adults in 3109 counties in the United States (2007 data). We identified four types of diabetes clusters based on spatial autocorrelations: high-prevalence counties with high-prevalence neighbors (High-High), low-prevalence counties with low-prevalence neighbors (Low-Low), low-prevalence counties with high-prevalence neighbors (Low-High), and highprevalence counties with low-prevalence neighbors (High-Low). We then estimated relative risks for clusters being associated with several socioeconomic and diabetesrisk factors.

Results:

Diabetes prevalence in 1551 counties was spatially associated (p<0.05) with prevalence in neighboring counties. The rate of obesity, physical inactivity, poverty, and the proportion of non-Hispanic blacks were associated with a county being in a High-High cluster versus being a non-cluster county (7% to 36% greater risk) or in a Low-Low cluster (13% to 67% greater risk). The percentage of non-Hispanic blacks was associated with a 7% greater risk for being in a Low-High cluster. The rate of physical inactivity and the percentage of Hispanics or non-Hispanic American Indians were associated with being in a High-Low cluster (5% to 21% greater risk).

Discussion:

Distinct spatial clusters of diabetes prevalence exist in the United States. Strong association between diabetes clusters and socioeconomic and other diabetes risk factors suggests that interventions might be tailored according to the prevalence of modifiable factors in specific counties.

Keywords: County-level, diabetes cluster, diabetes risk factors.