This cross-sectional study evaluated data from three cycles of the National Health and Nutrition Examination Survey (NHANES 2013–2014, 2015–2016, and 2017–2018) to assess protein and nutrient intake, quality, and cost across quartiles of plant protein intake in 14,888 adults (19 and older). As a percentage of total protein intake, the quartiles of plant protein intake were less than 28% (Q1), 28%–32% (Q2), 32%–36% (Q3), and more than 36% (Q4).

Dietary intake data were collected by trained interviewers who conducted 24-hour dietary recalls per survey set. Protein intakes from animal and plant sources were estimated using the United States Department of Agriculture Food and Nutrient Database for Dietary Studies (FNDDS) and the Food Products Equivalents Database. The FNDDS provides nutritional composition of all foods and food groups such as “protein foods”, “grains”, or “vegetables”, and the Food Products Equivalents Database provides food group composition of each food and beverage consumed. Amino acid intake was estimated according to the FNDDS and the National Nutrient Database for Standard Reference. If amino acid contents were not specified, they were estimated from matching similar foods/ingredients. Protein quality was estimated using the protein digestibility-corrected amino acid score (PDCAAS).

🔍 Digging Deeper: How is protein quality assessed?

Although proteins can serve as a source of energy, they are also the main dietary source of nitrogen and IAAs that play a crucial role in growth, function, and maintenance. Protein quality is defined by the ability of a protein source to satisfy IAA needs for a specific population. It is characterized by amino acid composition, digestibility (i.e., the proportion of ingested protein that is available for absorption) of the IAAs, and the requirements of the target population. Generally, requirements for growth and development decrease with age, and maintenance becomes the primary concern.^[4]

There are two main methods that are used to determine protein quality based on the ratio of the limiting IAA (i.e., the IAA with the lowest quantity) of a test protein compared to a digestibility-corrected reference protein. The first method is the protein digestibility-corrected amino acid score (PDCAAS) and the second method is the digestible indispensable amino acid score (DIAAS). Two main differences are that the PDCAAS uses digestibility values determined at the protein level and at the end of digestion (i.e., fecal level), where factors like nitrogen absorption and recycling and microbial metabolism and metabolites can influence results and sometimes result in overestimation. The DIAAS uses digestibility values determined at the individual amino acid level and at the end of the small intestine (i.e., ileum), where most protein is absorbed. The choice to determine digestibility at the end of digestion rather than at the end of the small intestine appears to account for most of the variance between method scores.^[4] Interestingly, neither of the methods properly account for protein complimentarity, which is when protein sources with amino acid profiles that compensate for limiting amino acids in each other are combined, such as in the case of grains and legumes or grains and nuts/seeds. The PDCAAS has an upper limit of 100% for protein quality, and the DIAAS only uses digestibility values for the limiting IAA.^[4]

The PDCAAS is the current internationally approved method because there is a large amount of data available for a large range of foods, it is easier to measure, and it has a longer history of use. However, the PDCAAS has been criticized because the digestibility between amino acids can vary widely^[5] and protein is mainly digested and absorbed in the small intestine,^[6] which can lead to overestimates of protein quality of more than 10%.^[4] In fact, the World Health Organization and Food and Agriculture Organization of the United Nations recommends the use of DIAAS as the preferred metric for protein quality assessment, but this requires amino acid digestibility data from measurements at the end of the small intestine (i.e., ileum) that are extremely invasive and have mostly only been collected in animals like pigs and rats.^[4] This means that the current digestibility values available for use with the DIAAS are only true for pigs and rats, and not necessarily humans. Alternative methods (e.g., stable isotopes) developed for this reason have yielded promising results, but they have not been rigorously validated,^[7] and thus these data gaps limit the practical use of DIAAS.^[8]

The cost of protein (grams per U.S. dollar) were estimated using the 2001–2004 Center for Nutrient Policy and Promotion’s National Food Prices Database and adjusted for inflation for the appropriate NHANES year. Dietary modeling was conducted to evaluate the effects of plant protein complimentarity when amino acids from grains were partially (25% or 50%) replaced by amino acids from higher quality plant protein foods such as legumes (nonoil seeds, pulses such as beans and lentils, and oil-seed crops, such as peanuts and soybeans).^[9] Data and analyses were adjusted for the complex sample design of NHANES and sociodemographics. In other words, the researchers used appropriate survey weights that estimate the number of U.S. residents that each survey respondent represents.

How does plant protein intake affect total protein intake, quality, and cost, and nutrient intake? Original paper