Research article
Geographic Distribution of Diagnosed Diabetes in the U.S.: A Diabetes Belt

https://doi.org/10.1016/j.amepre.2010.12.019Get rights and content

Background

The American “stroke belt” has contributed to the study of stroke. However, U.S. geographic patterns of diabetes have not been as specifically characterized.

Purpose

This study identifies a geographically coherent region of the U.S. where the prevalence of diagnosed diabetes is especially high, called the “diabetes belt.”

Methods

In 2010, data from the 2007 and 2008 Behavioral Risk Factor Surveillance System were combined with county-level diagnosed diabetes prevalence estimates. Counties in close proximity with an estimated prevalence of diagnosed diabetes ≥11.0% were considered to define the diabetes belt. Prevalence of risk factors in the diabetes belt was compared to that in the rest of the U.S. The fraction of the excess risk associated with living in the diabetes belt associated with selected risk factors, both modifiable (sedentary lifestyle, obesity) and nonmodifiable (age, gender, race/ethnicity, education), was calculated.

Results

A diabetes belt consisting of 644 counties in 15 mostly southern states was identified. People in the diabetes belt were more likely to be non-Hispanic African-American, lead a sedentary lifestyle, and be obese than in the rest of the U.S. Thirty percent of the excess risk was associated with modifiable risk factors, and 37% with nonmodifiable factors.

Conclusions

Nearly one third of the difference in diabetes prevalence between the diabetes belt and the rest of the U.S. is associated with sedentary lifestyle and obesity. Culturally appropriate interventions aimed at decreasing obesity and sedentary lifestyle in counties within the diabetes belt should be considered.

Introduction

The American “stroke belt,” defined in terms of a contiguous group of states with high age-adjusted stroke mortality rates, was first identified in the mid-1960s.1 The states that define the stroke belt are Alabama, Arkansas, Georgia, Indiana, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, and Virginia. The stroke belt's identification contributed to further study of the etiology of stroke and to helping target interventions to high-risk states.2, 3

Diabetes is similar to stroke in that it is strongly affected by behavioral, cultural, and environmental factors clustered and overlaid on genetic susceptibility. However, U.S. geographic patterns of diabetes have not been as specifically characterized in the manner that stroke has been. Identification of areas of high prevalence of diabetes could have an impact on understanding of diabetes.

Recently, the CDC produced estimates of the prevalence of diagnosed diabetes for every U.S. county or county equivalent.4 Figure 1 displays a map of the estimated prevalence of diagnosed diabetes for 2007, with counties/county equivalents shaded to represent the percentage of the population with diagnosed diabetes, dividing the counties into quintiles. This map suggested to the authors the existence of a “diabetes belt,” to be precisely defined later, probably not previously characterized because of the lack of county-level diabetes surveillance data.

This paper (1) proposes a definition of the diabetes belt; (2) examines how residents of that belt compare with residents of the rest of the country in demographics and prevalence of selected risk factors for diabetes and in the association of demographics and risk factors associated with diabetes; and (3) calculates the fraction of the excess risk of diabetes associated with residence in the diabetes belt that is attributable to selected risk factors.

Section snippets

Data Source

Data from the 2007 and 2008 Behavioral Risk Factor Surveillance System (BRFSS) were used. The BRFSS is a state-based system of health surveys that annually assesses key behavioral risk factors and chronic conditions among non-institutionalized U.S. adults aged ≥18 years. Nationally, the median state response rates for BRFSS were 50.6% (2007) and 49.8% (2008). The BRFSS collects county-level data but, because of sample-size limitations, most analyses are done at a state or national level. Using

Results

Details of the spatial smoothing are not reported, for brevity. However, the results of this smoothing appear in Figure 2. Figure 2 displays a map of the 644 counties that define the diabetes belt. This belt includes portions of the states of Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, Texas, Virginia, and West Virginia, as well as the entire state of Mississippi. The prevalence of diabetes in the diabetes belt was

Discussion

To the authors' knowledge, this paper is the first to identify a diabetes belt. Both diabetes and stroke belts are primarily located in the southeastern U.S. However, differences exist. Much of West Virginia is in the diabetes belt, but not in the stroke belt. Indiana is part of the stroke belt but contains no diabetes belt counties.

Public health is often carried out at a county level. Identifying a diabetes belt via counties allows policymakers to identify regions where need is greatest,

References (12)

There are more references available in the full text version of this article.

Cited by (228)

  • Precision Nutrition in Diabetes

    2023, Precision Nutrition: the Science and Promise of Personalized Nutrition and Health
View all citing articles on Scopus
View full text