Why it takes more than a grocery store to eliminate a ‘food desert’

“Our goal is ambitious. It’s to eliminate food deserts in America completely in seven years.”

Michelle Obama spoke these words almost four years ago to students at Philadelphia’s Fairhill Elementary School, as part of her Let’s Move! campaign to end childhood obesity in the United States. But the topic of her speech went beyond the issues of child health and nutrition to focus on a related — and just as critical — issue: 23.5 million Americans live in areas without supermarkets or other places where they can access fresh, nutritious foods. To change the situation in these areas — known as “food deserts” — Mrs. Obama called for action.

“This is happening all across the country. We’re setting people up for failure if we don’t fix this.”

Fast forward to 2014, though, and the problem of food deserts — and their effect on diet and health — still persists. The U.S. Department of Agriculture has mapped thousands of locations across the country where residents continue to live in low-income, low-access areas. Those who live in these areas are often subject to poor diets as a result, and are at a greater risk of becoming obese or developing chronic diseases.

So why hasn’t there been more drastic change? It’s not for lack of initiative. For example, Pennsylvania has launched a program whereby 88 new or expanded food retail outlets have been created, giving healthy food access to around 500,000 children and adults. And in fact, when the House passed the long-awaited farm bill on Wednesday, it included a provision for the HealthyFood Financing Initiative, which would allocate $125 million for expanding food resources in underserved communities across the nation.

The problem may not lie solely with food accessibility; it could also be due to people’s shopping and eating habits. Steven Cummins, a professor of population health at the London School of Hygiene and Tropical Medicine, suggests that merely adding a new grocery store to a neighborhood won’t be enough to motivate individuals to shop there for healthier foods.

“If you think about Kevin Costner in the Field of Dreams — ‘If you build it, they will come’ — I guess that’s the kind of logic model that underpins these interventions. But that doesn’t do everything it’s supposed to do,” says Cummins. “It can improve perceptions of food access, but it doesn’t necessarily translate into a behavior change.”

Could that be why more food deserts haven’t truly been eliminated? In his new study, published in the journal Health Affairs, Cummins examines how adding a supermarket to a food desert can change — or not change — the shopping habits and diet of those who live there. Surveying hundreds of residents in the grocery store’s neighborhood, Cummins and his team sought to determine the impact that the store had on individuals’ fruit and vegetable intake, BMI, and perceptions of food access.

Cummins sat down with the NewsHour last week to talk more about the study and its possible implications.

Steven Cummins is a professor of population health at the London School of Hygiene and Tropical Medicine.

PBS NEWSHOUR: Steven, thanks very much for speaking with us today. First off, give us some background. What’s the current policy surrounding food access in underserved communities?

STEVEN CUMMINS: In the last 15 to 20 years, particularly in the U.S., there’s been a lot of debate among researchers and policymakers about what to do about things called “food deserts” in underserved or low-income communities throughout the U.S. There’s lots of good research out there that tells us that a lack of access to food resources in a neighborhood is associated with a range of diet-related chronic diseases, such as diabetes, obesity, and cardiovascular disease. So it’s very plausible and entirely reasonable that people come up with interventions to promote food access in deprived communities.

The issue is, though, that despite the loads of policy over the past few years both from local initiatives, such as in Pennsylvania and New York City, as well as national policies — like those promoted in the Let’s Move! campaign for childhood obesity — we don’t know whether these kinds of interventions are effective. What I mean by interventions is increasing food access by encouraging supermarkets to locate to underserved communities, usually through packages of tax incentives. So that’s the context. And our study is essentially an evaluation of what happens when you put a new food supermarket in one of these underserved communities.

PBS NEWSHOUR On that note, explain a little more about your study — what were you aiming for, and what did you find? :

CUMMINS: We focused on a supermarket in Philadelphia, which was funded under something called the Pennsylvania Fresh Food Financing Initiative — a statewide scheme whereby around 88 new stores have opened up in underserved communities throughout Pennsylvania. They’ve had financial support through grants and other types of financial incentives to relocate to these areas. With that in mind, we undertook an evaluation to see whether actually locating these kinds of stores in these kinds of places makes a difference to diets. In our study, the “headline” finding is that we don’t really find any direct impacts on diets, as measured by fruit and vegetable consumption as well as obesity. But we do find improvements in people’s perceptions of their access in their food environment. They think that things have gotten better in their neighborhood, but haven’t necessarily turned their awareness into a change of behavior. So that’s the main finding. But the second finding, which I actually think is more interesting, is that when given the opportunity, very few people try and switch to using newer or better provisions within their local community.

PBS NEWSHOUR: And why do you think so few people switched, or changed their diet habits, when the new grocery store opened?

CUMMINS: We didn’t actually collect information about this in our study, but we have done some other work related to this in the UK. And it’s kind of very interesting, actually. When you go and interview people about how they shop, you tend to find that they have routine places to shop that they’re familiar and comfortable with. We had a woman in one of our UK studies who actually lived next door to a brand new supermarket, but still traveled to an area over three miles away because that’s where she was born, that’s where she grew up — it was a place she had a connection to. So there are all sorts of other reasons apart from the rational ones around cost when people make these decisions.

I guess it’s also to do with the fact that these kinds of stores might not always sell cheaper food. People who are on low incomes tend to be very savvy shoppers, and they often shop around, using multiple stores to get the best deals on a range of items. So you tend to find that people who are on low incomes move around the neighborhood a lot, even if there is a time cost.

Another insight to be had is that when you shop in a modern supermarket, you’re assaulted — your senses are assaulted. There’s a greater range of choice that can tempt some families to overspend or purchase foods that don’t comprise elements of a healthy meal. So if you’re trying to avoid all those other temptations or incentives to not spend money in the way that you would like, then you often avoid those kinds of opportunities to purchase food.

PBS NEWSHOUR: In that case, do you think researchers and the media may have jumped the gun when reporting on food deserts?

CUMMINS: Well no, not really, actually. I think the evidence is well established. It tells us that firstly, food deserts do exist in many urban and rural areas of the U.S. They are a reality for many people who live in disadvantaged circumstances, either in low-income communities or other kinds of communities that might have poor access to neighborhood resources. Also, the evidence does tell us very strongly that those who live in these kinds of neighborhoods do tend to have poorer diets and are at an increased risk of chronic conditions. So that’s not in dispute. The issue here is that actually, very little research evaluates the effectiveness of interventions. We know that an association exists, but we don’t know much about what happens when you try to change the environment. As often is the case in politics, policies are made from the best available evidence and with the best intentions. So in this case, policy has been made from the wider body of evidence that supports an association — since when it comes to looking at the effectiveness of intervention, the evidence base is incomplete.

And clearly, I think it’s worth mentioning that people do need to be able to access a reasonable range of healthy food at a reasonable cost. My message is that just building supermarkets to increase access will only take you half of the way in improving diet. If you think about Kevin Costner in the Field of Dreams – “If you build it, they will come” — I guess that’s the kind of logic model that underpins these interventions. But that doesn’t do everything it’s supposed to do. It can improve perceptions of food access, but it doesn’t necessarily translate into a behavior change.

PBS NEWSHOUR: How strong is your message, though, considering that your study only looked at one grocery store in one community for just six months?

CUMMINS: When we were funded to do this work, it was actually intended to be a pilot study with the idea of generating some interesting findings that could then be used to essentially run a definitive study further down the line. That didn’t happen for various reasons – one of which is that the recession happened, and the supermarket opened three years later than it was supposed to. So the opportunity passed, I guess. But you’re right – this is a pilot study, and it’s a small one at that. We also know from looking at business literature that stores often take a fair bit of time to bed into the local community. In some cases there might be an uptick in usage, which then tails off; in other places, it may be that it takes a very long time for people to become used to the store and see it as a routine place to go. It may be that our study, which only lasted six months, is not enough time for people to really realize that it’s there and use it for their main food shop. So if we took a look at this one, two, or three years down the line, we might see a greater amount of people who have adopted this store as their main store.

I wouldn’t want to hang my hat on the findings, but we need larger studies in more diverse communities that follow people for longer. But I guess the strength is that even though the work that does exist in this area has been quite small scale, it’s remarkably consistent in its findings.

PBS NEWSHOUR: So what will be more effective in helping improve people’s diet and health in these regions? And what should future policies focus on?

CUMMINS: At a basic level, people need to have good access to food. But good access does not necessarily mean you have to have access to a supermarket, such as a Walmart or Kroger or somewhere similar. It can also include other kinds of food outlets, such as farmers’ markets, community food initiatives, and actually, even convenience stores. I think there should be better support for the convenience store sector. Poor neighborhoods in the U.S. and other countries tend to have a very large and diverse convenience store sector, and many people do their shopping there. So trying to get good food in those kinds of places might be an additional way of thinking about improving the food environment in those neighborhoods.

In addition to improving physical access to food in disadvantaged neighborhoods, you also need to think about policies that help bridge this gap between perception and action. These might include things such as economic initiatives — like taxes or subsidies for healthy foods — but could also include harnessing in-store marketing to promote the purchase of healthy foods as opposed to unhealthy foods.

There’s also evidence that suggests you need good health education programs that teach the skills needed for buying and cooking healthy foods. We’ve done some work here in the UK where we’ve followed consumers around the store as they actually buy food. Even though all of them were from low-income communities, they all had different strategies about how they decide what to put in their shopping baskets. Some people are incredibly good — one woman used to put her total budget on her mobile phone’s calculator, and every time she purchased something she would take that amount of money off the total. When it reached zero, that was it — she was very disciplined. But others have more chaotic approaches — shopping by offers like 2-for-1 deals. And at the end of the day, once they spent their total amount of money, they would look in their basket and realize they had a bunch of items that they couldn’t construct a meal from. So even though people from disadvantaged neighborhoods are tagged in the same way, there’s a real diversity in how people behave when it comes to buying and cooking food. We have to think very carefully about giving people the skills to make better decisions when they’re in stores, as well as providing access to the stores in the first place.

PBS NEWSHOUR: What do you want people to take away most from your study?

CUMMINS: I want to stress that supermarket interventions — even though I don’t think they’re necessarily effective in the way people think they’ll be effective — are very important, and I am actually quite supportive of them. Access at the basic level is something you need to have before you have anything else. But the key message I want to get across is that they’re just not successful on their own. We need them plus a range of other things that might make a difference to improve people’s diet. You’ve got to think about the culture around shopping and eating food, too.

PBS NEWSHOUR: Steven Cummins, thanks so much for joining us.

CUMMINS: Thank you.

The Robert Wood Johnson Foundation provides grants to help alleviate the problem of food deserts, and they are a funder of our health coverage.