Gina Kolata highlights two recent studies that call into question the concept of “food deserts,” wide swaths of the country, traditionally low-income, that are thought to lack access to healthy foods. Both looked for a relationship between obesity and access to healthy foods, measured by the number of nearby supermarkets and fast food restaurants. Neither could find it.


The research, she writes, “raise questions about the effectiveness of efforts to combat the obesity epidemic simply by improving access to healthy foods.”

These two studies are among the largest to challenge how we understand healthy eating. But they’re certainly not the first. That body of research stretches back about a decade, to right around the time that “food desert” entered the public health lexicon. And, taken together, it challenges a lot of our assumptions about the role geography plays in healthy eating — or whether it plays any role at all.

Two of the first food desert studies came out of Britain in the early 2000s, both natural experiments that studied the impact of a grocery store opening in an area with few food options.

Neither found especially impressive results. A 2002 study in Leeds found a new grocery store to increase consumption of vegetables by one-third of a cup daily. A similar 2005 study, this one in Glasgow, found no difference. Further studies have found some impact on healthy eating, albeit generally moderate. And, to date, no study has found a connection between increased access to healthy food and improved health care outcomes.

Instead, studies like these have created a new question about whether food deserts matter. Does proximity to healthy foods necessarily mean better eating habits? In other words, if you build it, will they come?

Some of the most interesting research on this front has come from Adam Drewnowski at the University of Washington. He’s been exploring the relationship between where people live, and where they buy food. Most studies on food deserts break down food access by census tracts, small geographic areas used by the Census Bureau, and look at what resources are available in the given space.

But what Drewnowski found, in a 2009 study, suggests that census tracts may not matter. In his study of Seattle, only 15 percent of the city’s residents shopped within their own census tract. Further research, published this past summer, has contested the idea that shopping at a venue with fresh produce will necessarily translate into better health.

At Albertson’s, one of the Seattle-area’s cheaper supermarkets, 37 percent of shoppers were obese. Their shoppers also tend to be lower-income, a measure that has been correlated with higher rates of obesity and adverse health outcomes. Drewnowski therefore cautions against seeing grocery stores as a safeguard against obesity.

“They have salads and they have apples,” Drewnowski told me recently. “You can’t go at them saying they only have french fries and salted foods. So, now what?”

The other challenge, Drewnowski points out, has to do with the other products that supermarkets sell. Grocery stores stock apples and carrots; they also stock chips and cookies. And even if shoppers do start supplementing their diets with fresh fruits and vegetables, it’s no guarantee they’ll ditch the calorie-dense junk food.

“The hope is that they will be buying bags of fresh apples,” says Drewnowski. “But you can also see the same reaction of, this is great, I can get my donuts much closer.”