This article is more than 5 years old. Information may no longer be current.
Type 1 diabetes prevalence linked to ‘food swamps’
A high proportion of neighborhood fast food restaurants was associated with a higher prevalence of type 1 diabetes in adults and children in New York City, according to study findings published in the Journal of the Endocrine Society.
“An adverse food environment may potentiate not only type 2 diabetes, but also type 1 diabetes,” David C. Lee, MD, assistant professor in the Ronald O. Perelman Department of Emergency Medicine at New York University, told Endocrine Today. “It may help explain why we are seeing such dramatic increases in rates of type 1 diabetes.”
Using emergency claims data from 2009 to 2013, Lee and colleagues identified New York City residents with a history of diabetes. Among adults aged at least 18 years, 11,561 had type 1 diabetes (22% with Medicare coverage) and 528,862 had type 2 diabetes (44% with Medicare coverage); among children, 3,333 had type 1 diabetes (31% black) and 1,794 had type 2 diabetes (42% black). The researchers coded patient location by home address and performed a geospatial analysis of diabetes prevalence by census tract. Neighborhood-level factors associated with higher diabetes prevalence, including demographic, socioeconomic and food environment characteristics, were determined using multivariable regression analysis. New York City Department of Health and Mental Hygiene inspection data from 2009 to 2013 were used to identify 1-mile radius areas with a high concentration of fast food stores (fast food or counter service or take out only) as “fast food swamps”; New York State Department of Agriculture and Markets inspection data from the same period were used to identify areas with many retail food outlets (bodegas and small convenience stores) as “retail food swamps.”
Among adults, prevalence of type 1 diabetes was 0.23% and type 2 diabetes was 10.5%. Type 1 diabetes prevalence among children aged 17 years and younger was 0.2%, and type 2 diabetes was 0.11% among those aged 10 to 17 years. A higher prevalence of all diabetes types, excluding pediatric type 2 diabetes, was linked to areas with a higher proportion of fast food restaurants (P < .001). Compared with areas without fast food restaurants, prevalence of adult type 1 diabetes, adult type 2 diabetes and pediatric type 1 diabetes was 1.55, 2.52 and 2.03 times higher, respectively, in fast food swamps. Retail food swamps with a 1-mile radius were not associated with diabetes prevalence. However, Lee and colleagues found that when the retail food radius was increased to 2 miles, a lower prevalence of adult type 1 diabetes (P = .004) and a higher prevalence of pediatric type 1 diabetes (P = .002) were observed.
Researchers concluded that although a higher prevalence of type 1 diabetes was associated with food environment, the same was not observed for pediatric type 2, which was concentrated in neighborhoods with a black majority population. Lee told Endocrine Today that in regard to the causes of type 1 and type 2 diabetes “lines are blurred.”
“Traditionally, we have thought of type 1 diabetes being the genetic form of diabetes and type 2 diabetes being the one caused by obesity,” Lee said. “An adverse food environment has an important influence in type 1 diabetes, and we should consider type 2 diabetes is less about the food environment and [more] about individual risk factors and family influences.”
The study findings suggest more research is required to address the needs of adults and children living in areas with geographic disparities in diabetes prevalence, Lee said.
“We've done something novel by using emergency department visits to help us find where people with these conditions live and where cases of diabetes are clustered at higher rates,” he said. “We need to not only validate the methods that we developed in our study elsewhere, but also consider using alternative data sources to identify associations of diabetes with environmental and other influences.” – by Marley Ghizzone
Disclosures: Lee reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
Perspective
Back to Top
This study is introducing a new tool in studying the prevalence of diabetes by using emergency claims data over 5-year period and linking them back to patients’ location through their address. This unique method enabled researchers, for the first time, to survey an overwhelming majority of the population in New York City. For example, the survey covered 5 million adults out of 6.5 million and 1.6 million children and adolescents out of 1.8 million. Its accuracy in identifying prevalence of type 1 and type 2 diabetes is very important and is beyond other survey methods.
Finding a surprising relationship between type 1 diabetes and fast food swamps is interesting and is really unexpected. This may open researchers' eyes to study the epigenetics of the disease and the environmental factors unique to certain locations. Specifically, gene-environment interaction should be a focus of new research of type 1 diabetes pathophysiology. Considering that more than 50% of patients of type 1 diabetes are currently overweight or obese, gestational diabetes and obesity among parents should now be included as new research targets in our quest for type 1 diabetes etiology. Also, methods of prevention and interventions that target patients in the locations of the highest prevalence should be considered.
The findings of increased prevalence of type 2 diabetes among young black individuals and the findings of higher prevalence of type 2 diabetes among adults from minority populations and low-socioeconomic standard are consistent with previous surveys. Although this new method has little bias toward patients visiting emergency departments, its 5-year period minimized the impact of this bias. This method is attractive for further research in diabetes and other chronic diseases.