By Kyley McGeeney and Elizabeth Mendes
In the United States, obesity in “food deserts” is above average. However, it is not solely — or even primarily — access to grocery stores that appears to be the issue — higher obesity rates are more likely to be linked to lower incomes. In other words, a lack of access to food in and of itself doesn’t matter when it comes to obesity. It only matters if Americans are also low-income. Further, income always matters, regardless of whether an individual has access to grocery stores or not.
“Food deserts” are typically defined as either an area that has limited access to grocery stores or as an area that is low income and lacks access to grocery stores. Regardless of which definition is used, what is clear is that the lack of access to grocery stores alone is not related to higher obesity rates — rather, it is more a lack of income.
Americans living in areas that are both low-income and that have limited access to grocery stores are only marginally more likely to be obese than those living in areas that are low-income, but that don’t lack access to grocery stores — about three in 10 in each. Further, Americans living in areas that have limited access to grocery stores — but are not low-income — are less likely to be obese (25 percent). They also are about equally as likely to be obese as those living in an area that is neither low-income nor low-access.
This analysis combines food desert data from the U.S. Department of Agriculture’s Food Access Research Atlas with obesity data from the Gallup-Healthways Well-Being Index, in a first-of-its-kind study.
The Food Access Research Atlas labels geographic areas — known as census tracts —as low income, low access, both, or neither based on income data from the 2006-2010 American Community Survey and access data from two different lists of the locations of supermarkets, supercenters, and large grocery stores in 2010. Gallup and Healthways collect obesity data down to the ZIP code level and used more than 300,000 interviews with Americans aged 18 and older from 2010, assigning Americans to the most appropriate census tract.
To further assess the connection between food deserts and obesity — or lack thereof — Gallup conducted an in-depth analysis of the factors most related to obesity. Gallup looked at a wide variety of factors: low income and low access to food (and the interaction between those two), gender, race/ethnicity, age, education, marital status, and region. The findings show that, while low income alone is indeed a significant predictor of obesity, low-access alone is not. Low-income and low-access combined do have a significant relationship to obesity, though. The bottom line is that having a lower income is the dominant factor in an American’s increased likelihood to be obese.
Two recent studies have called into question the popular theory of the connection between food deserts and obesity, but both studies only looked at children.
A 2012 study from the Public Policy Institute of California found that children in disadvantaged neighborhoods actually had greater access to supermarkets and grocery stores than those in better-off communities. The author, Helen Lee, concluded that “differential exposure to food outlets does not independently explain weight gain over time” for the group of school-aged children she studied.
Another 2012 study, by the RAND Corporation, similarly found no relationship between the food available where children lived and what they weighed. RAND used data on more than 13,000 California children aged 5 to 17 years from the 2005 and 2007 California Health Interview Survey.
These studies and Gallup’s new analysis reveal that a focus on increasing access to places to buy healthy food alone may not be an answer to help improve how people in “food deserts” eat, nor a way to positively impact their health and weight. The data highlight that addressing food affordability — and possibly even increasing knowledge about what constitutes healthy foods — is more likely the answer to decreasing obesity in America’s food deserts.