The Washington PostDemocracy Dies in Darkness

Food deserts don’t cause obesity. But that doesn’t mean they don’t matter.

An activist holds carrots in the air to make the point about lack of fresh food available in poor areas of Washington during a 2017 protest. (Michael S. Williamson/The Washington Post)

Do food deserts cause obesity?

That was the question I set out to answer, and I was prepared for a mishmash of evidence and a variety of expert opinion. After all, on any complex issue, that’s what you’re likely to find.

But that’s not what I found. Just about all the evidence, and every expert I read up on or checked in with, is on one side of this issue. So I’m ready to shout it from the rooftops.

No, food deserts don’t cause obesity.

The true connection between poverty and obesity isn’t what you probably think

People — experts, advocates and just plain people — used to think they do, but then a funny thing happened. Scientists studied the question, and it simply turns out that no, they don’t.

“2009 was the height of food deserts,” says Tamara Dubowitz, senior policy researcher for the RAND Corporation (a policy think tank) who has studied the issue for years. Advocacy groups — and former first lady Michelle Obama — were focused on food deserts “because access was a social justice issue. It wasn’t based on evidence because there wasn’t any evidence.” There were some studies that showed a rough correlation, but that was it.

The idea that areas that lack of access to a full-service supermarket — a.k.a. food deserts — promoted obesity “made theoretical sense,” Dubowitz says. And it was a testable thesis. So, it got tested! Scientists looked closely at the relationship grocery access has to obesity, and tracked changes to obesity and other health outcomes in low-access neighborhoods that got a new supermarket.

It turns out that grocery access doesn’t correlate cleanly with obesity, and a new grocery store is unlikely to make a dent in obesity rates. And those results came up in study after study after study.

In South Carolina, distance to the grocery store didn’t correlate with BMI. “These findings call into question the idea that poor spatial access to grocery stores is a key underlying factor affecting the obesity epidemic,” the authors conclude.

In Philadelphia, it was the same story. In Detroit, too. Ditto among veterans.

An economic model found that “exposing low-income households to the same availability and prices experienced by high-income households reduces nutritional inequality by only 9%.”

Moby caught flak for saying food stamps shouldn’t pay for junk food. But he’s right.

A paper that describes an effort to assess neighborhood changes when a supermarket moves in begins by saying, “Initiatives to build supermarkets in low-income areas with relatively poor access to large food retailers (“food deserts”) have been implemented at all levels of government, although evaluative studies have not found these projects to improve diet or weight status for shoppers.”

A review in 2017 concluded: “Improved food access through establishment of a full-service food retailer, by itself, does not show strong evidence toward enhancing health-related outcomes over short durations.”

I have seldom found a body of evidence with results so relentlessly one-sided. Anne Palmer, who directs the Food Communities and Public Health program at the Johns Hopkins School of Public Health, explained in an email that the shift away from believing in the connection between obesity and food deserts “is as a result of researchers — especially economists — proving that the link is spurious at best. That would hold true for any health outcomes, not just obesity.”

Okay then, that’s pretty definitive.

And it raises a question. Go out in the world and talk about obesity, and the issue of food deserts inevitably — and I mean literally inevitably — comes up. Food access is still widely seen as a key driver of obesity despite the fact that the last nail seems to have been driven into its coffin in academic circles. Why is that?

I blame kindness and compassion.

So often, the conversation about obesity devolves into two competing narratives: personal responsibility vs. obesogenic environment. While every public health expert I talk with says they believe that both come into play, when you don’t want to make people fighting obesity feel any worse than they already do, the obesogenic environment is a much more comfortable place to stand.

Can making vegetables cheaper improve our health?

Rebecca Puhl is deputy director of the University of Connecticut’s Rudd Center for Food Policy and Obesity. Personal behavior is “one piece of the puzzle,” she wrote me in an email. But if we focus on that piece, “the puzzle will never be solved.” But, by addressing the food environment, we also address personal behavior, by finding ways to make it easier for people to make better choices. We need to focus on “strategies and policies that create healthier defaults and support responsible behaviors for everyone,” she wrote.

It’s essential that we address the food environment, which is our collective responsibility, and there’s some evidence that we are. A recent analysis estimated that 5.6 percent of Americans — that’s 17.6 million of us — have limited supermarket access, a number that’s down 15 percent since 2010. But the food environment is complicated, and just putting a supermarket in an underserved neighborhood won’t radically change people’s diets.

Step back, and look at how much has changed in our food environment over the past several decades. We’re navigating a landscape of diabolically irresistible food specifically designed to swamp our willpower, both by being engineered for deliciousness and by being available everywhere we turn. In that context, it makes sense that simply moving a produce section into the neighborhood won’t change much.

But it still might be a good first step. The obesity needle might not move, but Dubowitz found that the new supermarket she studied, in Pittsburgh, was associated with several smaller changes. Study participants reported eating a little better — less sugar, fewer calories. People felt better about living in the neighborhood. None of this was specifically attributable to the supermarket, as people who didn’t shop at it reported the same changes as those who did, but it raises the possibility that a supermarket can help improve a neighborhood in ways that go beyond food access. It brings jobs, and could pave the way for more investment. It can make the area feel safer.

The evidence on food access doesn’t mean supermarkets don’t matter, or that we should stop trying to make sure everyone has access to one. It just means we should stop talking about them as a solution to the obesity epidemic. It’s completely reasonable to support food access for its own sake — I sure do.

Meantime, there’s a lesson here. Everyone agrees that fixing obesity will be difficult. But, as researchers tackle this, they inevitably focus on the things we can measure. Distance to the grocery store. Income and education. Gut bacteria and carbohydrate levels. Meanwhile, there’s a laundry list of factors that are much harder to quantify. Cultural factors, susceptibility to advertising, hating to cook, ungrateful family members, really liking carbohydrates — the list is long and varied.

“We’re measuring the wrong things,” Dubowitz says. “Eliminating food deserts, it’s a feel-good policy that’s low-hanging fruit.” What we really need to do, she says, is start accumulating qualitative evidence on some of those less-tangible issues. “Quantifying everything is not always possible.” It’s way too easy to attach disproportionate significance to the things we can measure simply because we can measure them.

How to get people to cook more? Get eaters to complain less.

The food desert issue got traction because lots of public-health-minded people believed everyone should have access to decent food. That was a good reason then, and it’s a good reason now. And it’s also possible that supermarkets are one in a smorgasbord of interventions that, together, might make a dent. If a supermarket moved in and kids were taught cooking and nutrition in schools and we rejiggered SNAP so that (somehow) it focused more on healthful food and restaurants reduced portion sizes and some processed foods were reformulated and people stopped selling crank diet books maybe, given time, we would see some changes.

For people’s habits and diets to change, food access is necessary. But it is emphatically not sufficient.

More from Food:

The true connection between poverty and obesity isn’t what you probably think

Moby caught flak for saying food stamps shouldn’t pay for junk food. But he’s right.

Can making vegetables cheaper improve our health?

How to get people to cook more? Get eaters to complain less.

Unearthed column archive