Does the USDA's ‘One Size Fits All’ TFP Diet Fall Short?

The USDA Thrifty Food Plan (TFP) is the market basket used by the government to calculate SNAP benefits. It is optimized to provide a nutritious diet at the lowest possible cost. However, because our cultural upbringing significantly impacts our food choices, a scrupulously dietary-cultural neutral diet may not be optimized for real-world consumption. The USDA TFP is based on the total US aggregate, with no versions for specific culinary cultures. With that concern in mind, a new study in Nutrients looks to individualize the TFP for Hispanics.

Why Hispanics?

First, the important caveat is that the term Hispanic is ambiguous and covers a great deal of culinary ground across at least two continents. That said, the Hispanic population is our second largest demographic group, accounting for over 62 million in the US, or about 19% of the population. The data on the aggregate Hispanic diet comes from our old friend, the National Health and Nutrition Examination Survey (NHANES), specifically information from the 24-hour food recalls of self-reporting “Hispanics” in the What We Eat in America sub-study. A more granular breakdown indicates that half of the roughly 4200 respondents were self-described as Mexican-American. 

Another old friend, the Healthy Eating Index (HEI), used by the USDA in determining the TFP and subsequent SNAP benefits, provided information on the current nutritional value of real-world Hispanic diets. To the largest extent possible, the researchers used the same dataset and methodologies to calculate and optimize a Hispanic-TFP.

What’s For Dinner?

The Hispanic population participating in the study was generally younger, less educated, and “with significantly lower family income-to-poverty ratios.” The degree to which they and their diet were “acculturated” was not measured. The nutritional value of their diet, the HEI, was not different from the rest of the US aggregate – both are equally poor, with a value just touching 50 on a 100-point scale. 

“Based on the present analyses, the eating habits of the Hispanic group failed to conform to some stereotypical assumptions.”

• Vegetable consumption was comparable between Hispanic and non-Hispanic groups. The Hispanic group consumed fewer dark green and red/orange vegetables but much more beans, peas, and lentils.
• Fruit juice consumption was significantly higher in the Hispanic group [1] 
• Breads, tortillas, pasta, and rice were higher in the Hispanic group, but biscuits, muffins, and quick breads lower.
• Milk and yogurt consumption was similar, but the Hispanic group consumed less cheese.
• The Hispanic group consumed fewer cured meats and nuts/seeds.
• The Hispanic group consumed more mixed dishes with beans, eggs, and grains (omelets) and fewer mixed dishes with meat/seafood vegetables.
• Consumption of soups and sandwiches was higher; candy, snack bars, coffee, and tea consumption was lower in the Hispanic group.

Estimated food costs, as expected, trended with food choices; the Hispanic diet was less costly by about $6 for a family of four while maintaining a comparable HEI. 

“Selecting “healthier” diets as the departure point for optimization modeling necessarily maximizes the amounts of vegetables, whole fruit, and whole grains and minimizes the amounts of added sugars, solid fats, refined grains, and meat.”

With the data in hand, the researchers turned to the modeling. They developed two models: H-TFP 1 focused on more whole grains, fresh pork, and less poultry, and H-TFP 2, where pork was the sole meat source. Both models were cost-neutral at the $189 food budget for a family of four. Both models were better fits for Hispanic real-world eating.

• H-TFP 1 had a similar total vegetable intake but included more starchy and fewer other vegetables.
• H-TFP 1 had the same amount of fruit but selected more 100% fruit juice.
• H-TFP 1 included more whole and less refined grains.
• The H-TFP 1 model selected more cheese, no low-density dairy products, and higher amounts of pork, with less poultry and no cured meats.
• Seafood and nuts were kept at their maximum allowed amounts.
• The H-TFP 1 included more miscellaneous foods, particularly beverages like soda (both regular and diet) and nutritional drinks.

The H-TFP 2 diet, where fresh pork was preferentially selected, also provided a greater HEI at the same cost as the current USDA TFP. The researchers tried to create a vegetarian Hispanic TFP, but they, 

“failed to find a mathematical solution for several population subgroups. For the most part, the modeled food patterns did not supply adequate energy or were deficient in one or more nutrients.” [2]

Perhaps the study's most significant limitation, in addition to not identifying the myriad of Hispanic culinary cultures, was not accounting for the acculturation and changing dietary patterns that occur with time and experience. However, the study demonstrates that TFP can be optimized for traditional diets and “culturally specific ingredients.” Whether that is feasible as a real-world policy choice remains unknown, nor do we know whether such a more culturally sensitive approach would better meet nutritional needs. 

This study reveals just how tricky it is to adapt the USDA’s rigid dietary guidelines to reflect real-world culinary diversity without blowing the budget - the TFP remains an imperfect science experiment that keeps forgetting the flavor.

[1] Fruit juice is often cast as the spoiler in studies involving taxes on sugar-sweetened beverages because it is frequently a high-sugar substitute. 

[2] For males, this was often due to not being able to meet the caloric needs given food group constraints; a mathematical solution could be made by significant alterations in the diet to include eggs, dairy, pizza, sandwiches, and sweet bakery products in place of meat or if the family food budget was significantly increased. 

Source: Hispanic Thrifty Food Plan (H-TFP): Healthy, Affordable, and Culturally Relevant Nutrients DOI: 10.3390/nu16172915