The old saying “you are what you eat” rings true, but what may be more accurate is “you are where you eat,” according to a new study by USC researchers in collaboration with researchers at Northeastern University and Massachusetts Institute of Technology.

Why it matters: Poor diets are a leading cause of illness and death, and exposure to food environments saturated with fast-food outlets are believed to make it harder for people to choose healthy eating options.

The big picture: Millions of Americans suffer from food and nutrition insecurity, struggling to access enough healthy foods.

• For decades, efforts to combat nutrition insecurity and low-quality diets have focused on identifying “food deserts” — neighborhoods with few healthy, affordable food options — and “food swamps,” where fast-food chains abound.
• Research has shown that improvements to neighborhood food deserts and food swamps often fail to improve residents’ dietary habits, or curb rising obesity rates and other diet-related diseases.
• Since 2010, initiatives like the U.S. Department of Agriculture’s Healthy Food Financing Initiative, which supports healthy food retail in underserved neighborhoods, have been ineffective in changing eating habits.
• A 2008 policy that restricted new fast food restaurants from opening in Los Angeles’s poorest neighborhoods proved ineffective in improving people’s diets, studies found.

In her words: “We’ve been encouraging people to eat healthy for 100 years, and it often doesn’t work. Even when people want to eat healthy, there are too many things in their day-to-day lives and environments that make it hard,” said Kayla de la Haye, founding director of the Institute for Food System Equity at the Center for Economic and Social Research at USC Dornsife College of Letters, Arts and Sciences. “Now we’re gaining a more comprehensive understanding of how food environments affect diet beyond the home.”

The new study, published recently in Nature Communications, suggests a different approach to improving American diets by analyzing where people eat beyond their home neighborhoods.

• Past research has focused on food availability close to where people live, so most policies to eliminate food deserts and food swamps have targeted residential neighborhoods.
• This study provides detailed evidence of the food outlets people encounter as they move around throughout the day and how these so-called “mobile food environments” influence diets and diet-related diseases.

What they did: Data from smartphones offered insight into eating choices.

• The researchers examined 62 million visits to food outlets in 11 American cities using mobile phone data.
• The researchers were careful to keep phone data anonymous.
• The mobility data, gathered over a six-month period in 2016 and 2017, enabled the researchers to track when and where fast-food visits occurred.

Key findings: The researchers found that most fast-food visits occur relatively far from home, suggesting that interventions targeting food choices must consider the broader range of environments people navigate daily.

• The data showed that people traveled an average of 4.3 miles from home to get fast food, compared with only 2 miles to get groceries.
• There was no evidence that any particular socioeconomic group visited fast food outlets more than another.
• As people moved about their day, encountering more fast-food outlets was tied to more visits to these type of outlets.
• People visiting areas with 10% more fast-food outlets were 20% more likely to stop at one.

Good deeper: Comparing two locations each having 10 restaurants — one with five fast food and five healthy, the other with six fast food and four healthy — people are 20% more likely to choose fast food at the location with more fast-food options.

Places such as shopping malls, business parks, industrial factory areas and airports, which people frequently visit away from home, had the biggest influence on food choices.

• These places typically have an over-abundance of fast-food options, yet the study revealed that people chose healthier alternatives when available in these locations.

The researchers recommend that policymakers and others who want to improve food environments use mobility data. This kind of data can enable them to identify places with an abundance of food outlets and where they can make the most impact.

• “The data gives us a kind of roadmap to see where people are most likely to respond to an intervention — places where we know that if a healthier option were available, people visiting these locations would be more likely to choose it over fast food,” said Abigail Horn research assistant professor of industrial and systems engineering at USC Viterbi School of Engineering. “Those are the locations where we should implement interventions.”

The bottom line: Selecting optimal locations to devise food environment interventions could double, triple or even quadruple their effect on food choices compared with traditional interventions that focus on food deserts or food swamps around residential areas.