CLOSE MENU
Following a heart-healthy Mediterranean-style diet may be difficult for people of all income levels, not just those in lower-income neighborhoods, according to a new modeling study. The challenges stem from higher prices and limited local availability of the healthy foods that make up that diet.
The study was led by researchers from PHICOR (Public Health Informatics, Computational, and Operations Research) at CUNY SPH and the University of Southern California (USC) and is part of the ongoing work of the AIMINGS Center, which is developing different artificial intelligence (AI) approaches, methods, models, and tools to help achieve better precision nutrition.
The researchers developed and used a detailed agent-based computer model to simulate the daily food choices of adults living in three Los Angeles County neighborhoods: a lower-income neighborhood, Boyle Heights; a middle-income neighborhood, Inglewood; and a higher-income neighborhood, Santa Monica. The model represented the residents in these locations, their households, and the surrounding food outlets such as grocery stores, fast-food restaurants, and corner stores. Each simulated day, these computational representations of residents ate different meals. Experiments measured how well they could follow a combined Mediterranean–DASH (MED-DASH) diet versus defaulting to the less healthy Typical American Diet (TAD).
Simulation experiments showed that the average adherence to the MED-DASH diet was only 57% in Boyle Heights, 62% in Inglewood, and 68% in Santa Monica. When the price of MED-DASH foods in all food outlets was cut by half, adherence rose to about 70% in Boyle Heights and roughly 72% in Inglewood and Santa Monica, essentially closing most of the income-related gap in the ability to follow the MED-DASH diet.
Simulation experiments also revealed that the stocking patterns in nearby convenience and corner stores had a bigger impact on adherence to the MED-DASH diet than those at restaurants and fast-food outlets.
“Clinicians and nutrition programs should routinely factor in where people live, what kinds of food outlets are nearby, and how much healthy food costs when giving dietary guidance or designing interventions,” says PHICOR Senior Analyst Jessie Heneghan, the study’s lead author. “This is an example of how many factors outside the body can affect a person’s diet, nutrition, and resulting health and should be part of precision nutrition. Precision nutrition is about better tailoring people’s diets to their unique characteristics and circumstances to improve health.”
Moving more towards precision nutrition is important because less than 2% of Americans eat a diet that is ideal for good health, says Kayla de la Haye, a co-author on the study and director of the USC Food Systems Institute in Los Angeles. She emphasized how suboptimal diets have contributed significantly to the alarming rates of heart disease, stroke, cancers, type 2 diabetes, and other non-communicable diseases seen in the U.S. However, she says, telling people to eat healthier diets is not enough.
“Eating a healthy diet is not a problem of knowing, it is a problem of doing,” says de la Haye. “In other words, we need to make it a lot easier for people to be able to afford and access healthy groceries and convenient healthy meals to go, if we want people to be able to stick to the diets that their healthcare providers recommend.”
“Since nutrition and health involve complex systems of different factors and processes interacting, AI approaches can help better represent, understand, and account for them,” explains Professor Bruce Y. Lee, executive director and founder of AIMINGS, PHICOR, and CATCH at CUNY SPH. “In this case, our models are showing that considering the cost of and potentially subsidizing healthy foods and changing the stocking patterns of stores may be important parts of ensuring that everyone can maintain healthier diets and truly achieving precision nutrition.”
