Food systems affect population health, natural resource use, and socioeconomic issues and are in turn vulnerable to environmental changes. This vulnerability – coupled with growing demands to feed a global population of nearly 10 billion by 2050 – has accelerated recommendations to achieve “food systems transformation” (Fanzo et al., 2020). These recommendations take many forms, including national guidance through food-based dietary guidelines (FBDG) or reports from non-governmental organizations.

A prominent theme is a shift to plant-based or -exclusive diets, with emphasis on plant-based proteins, largely for environmental reasons. Cell cultured proteins, insects, and mycoproteins are also being considered as animal protein alternatives, though uncertainties remain regarding their viability on a mass scale. While some recommendations acknowledge the valuable role of both ASF and plant-based foods (PBF) as part of a healthy diet, others argue for PBF to replace ASF (Herforth et al., 2019; Comerford et al., 2021). Among one-third of national FBDG with “protein food” messages, PBF are explicitly presented as substitutes for ASF or are implied to be alternatives through inclusion in the same general message as ASF (Herforth et al., 2019). This perspective over-indexes on the environment and overlooks other domains of sustainable food systems, including nutrition, health, and sociocultural factors (Drewnowski, 2017). This paper focuses on what overlooking other domains of sustainability, particularly nutrition and health, could mean for protein quality.

Protein contains essential amino acids (EAA) that are needed for physiological functions across all life stages (Institute of Medicine, 2003). Foods and dietary patterns differ in protein and amino acid (AA) content and thus in their protein quality (Millward et al., 2008). Protein quality is defined as the ability of a dietary protein to meet the body's metabolic demand for AA and nitrogen. It is based on AA composition, digestibility of the dietary protein, and bioavailability of the AA from that dietary protein (Boye et al., 2012; FAO, 2013). Protein quality is therefore critical when assessing nutrient adequacy of the food supply and dietary intake at the individual- and population-level (Cifelli et al., 2016).

There is significant variation in protein quality across ASF and PBF, which is important to consider when recommending shifts in dietary patterns (Gwin et al., 2021). ASF like dairy, eggs, and meat are highly digestible (>90%); depending on the processing method and/or presence of antinutrients, PBF like maize, oat, bean, and pea typically have lower digestibility (45–80%) (van Vliet et al., 2015). There may also be differences in how the protein is metabolized and utilized by the body. For example, AA from soy and wheat are more readily converted to urea than those from milk, which results in a lower potential of these PBF to stimulate muscle protein synthesis (van Vliet et al., 2015). Dietary patterns that include a diverse mixture of ASF and PBF (including common staple foods and neglected or underutilized crops) often have high protein quality as their AA profiles complement one another. It is possible to consume complete protein through a combination of different types of PBF with complementary AA compositions; however, doing so is challenging among population groups that have higher protein requirements and/or are not accustomed to consuming a diverse variety of PBF.

Although evidence demonstrates the importance of considering protein quality and quantity when designing dietary recommendations, the topic has often been overlooked or oversimplified (Millward et al., 2008; Burd et al., 2019; Comerford et al., 2021). For example, driven by concerns for the environmental impact of current dietary practices, several countries have adapted FBDG that promote increased consumption of PBF and decreased consumption of ASF, either directly or indirectly, while not acknowledging a consideration for protein quality (Brink et al., 2019; Meltzer et al., 2019).

Typical recommended PBF to address protein intake include legumes, nuts, and seeds. Achieving transformation to sustainable healthy diets as defined by the EAT-Lancet Commission would require >100% increase in the global consumption of foods like legumes and nuts (Willett et al., 2019). Yet, intake of these foods is low – average per capita consumption of legumes is 21 g/day globally and 9.3 g/day in the U.S., which is below the recommendation in the Dietary Guidelines for Americans (DGA) (Dry Bean Council US, 2021; Semba et al., 2021a). Shifting dietary patterns toward higher legume, nut, and seed consumption requires significant changes in behavior, knowledge, and food preparation skills. A question therefore remains on how feasible such shifts would be given current dietary practices (Semba et al., 2021b).

Some research indicates recommendations to swap PBF for ASF can negatively impact intakes of protein and select micronutrients, particularly when modeled based on current consumption patterns. A study modeling different dietary scenarios using NHANES data found that increased intake of PBF resulted in an increased percentage of children (2–18 years) and adults (≥19 years) not meeting the Estimated Average Requirement (EAR) for protein, vitamins A and D, and calcium, which are nutrients of concern in the U.S. (Cifelli et al., 2016). Another modeling study found that doubling the intake of PBF led to a decrease in protein intake by about 22% among women and men aged ≥51 years. Additionally, protein intake among women ≥71 years decreased below the RDA and the percentage not meeting the EAR increased to 33% (Houchins et al., 2017). This demonstrates the potential detrimental effect on dietary outcomes if population groups increase PBF intake while decreasing ASF intake.

ASF are recognized for contributing to the overall quality of diet but have also received negative attention for their environmental impact. Early assessments of the effect of macrobiotic diets (i.e., diets based on whole-grain cereals, pulses, and vegetables) on infant and child growth and development in the Netherlands demonstrated the importance of including ASF in the diet. Results from these studies observed markedly lower intakes of energy and protein among children receiving macrobiotic diets compared to those receiving omnivorous diets, which was linked with linear growth faltering, fat and muscle wasting, and delayed development (Dagnelie and van Staveren, 1994).

ASF are rich sources of essential fatty acids and multiple micronutrients that are commonly lacking in the diets in LMIC, including vitamin A, vitamin B12, vitamin D, iron, zinc, and calcium (Neumann et al., 2001). They are particularly important for infants, young children, adolescents, and pregnant and lactating women who are undergoing physiological changes and have higher nutrient requirements (Nordhagen et al., 2020). Micronutrients in ASF have high bioavailability and enhance the absorption of nutrients from PBF with high phytate and fiber content that may inhibit the absorption of minerals (Gibson et al., 2003).

Animal products differ in their nutrient composition. Using dairy as an example, milk and milk products contain 13 essential nutrients, including high-quality protein, vitamin A, vitamin B12, vitamin D, riboflavin, folate, and calcium (Allen and Dror, 2011). Studies have consistently shown a positive association between dairy intake and linear growth in children aged 12–60 months. Further, the elimination of cow's milk from the diet has been found to be associated with a reduction in height and an increased risk of bone fractures among children (Goulding et al., 2004; Clark et al., 2020).

In the U.S., dairy is under-consumed relative to recommendations in the DGA (Krebs-Smith et al., 2010). Few people reach the recommended intakes of several key nutrients without consuming the recommended amounts of dairy foods (Weaver, 2014). A trend toward decreasing ASF intake could further reduce the intake of this food group.

In terms of plant-based milk alternatives, it is important to consider the variation in their nutritional profiles and that most do not provide the same nutrients as cow's milk. A study comparing the nutrient composition and carbon footprint of cow's milk and plant-based beverages (e.g., soy, oat, almond, coconut, and rice beverages) found that the protein and EAA content of cow's milk was higher. Although the carbon footprint of cow's milk was higher compared to plant-based beverages when expressed per serving, when expressed based on index of nutritional value (i.e., ability to contribute to meeting EAA requirements), the carbon footprint of cow's milk was lower than that of all plant-based drinks examined, except for soy beverage (Singh-Povel et al., 2022). These findings reflect the importance of considering the nutritional value of food choices when reporting environmental impact and making broader conclusions regarding sustainability.

The sustainability of food systems can be measured across four domains: health, environmental, economic, and societal (Drewnowski, 2017). Each domain has respective metrics. For example, nutrient profiling models estimate the nutrient density of foods. Life cycle assessments (LCA) evaluate environmental impacts of foods relative to land, water, and energy use. Choices related to dietary protein may be influenced by culture. Assessments of food consumption patterns across populations can be used to understand the cultural and societal importance of such foods (Drewnowski, 2017).

The complexity lies in integrating metrics across domains to capture a holistic impact of food production. A study examining the relationship between the energy and nutrient content of foods and associated greenhouse gas emissions (GHGE) found that many foods with low GHGE had relatively low nutritional value; meat and dairy products, which were more nutrient-dense, had higher GHGE values per 100 g but lower values per 100 kcal. This raises the question as to whether the higher GHGE cost of some foods could be offset by their higher nutritional value (Drewnowski et al., 2015). Another analysis expressed GHGE of ASF and PBF relative to EAA and found the perceived environmental advantage of plant-based protein production to be smaller than previously estimated. Expressing land use relative to EAA also negated some perceived advantages of plant-based proteins (Tessari et al., 2016). When evaluating the environmental impact of animal- and plant-based foods, different conclusions can be drawn between assessments based on protein quantity and those that account for protein quality. For example, GHGE for milk production has been estimated as ~400% higher than for plant production when expressed as per ton of gross protein consumed. This difference was reduced to 59% when expressed based on kilograms of digestible lysine consumed to account for protein quality. Milk production was also the most efficient production system in terms of water use when expressed on a digestible lysine basis (Moughan, 2021).

Incorporating nutritional-based functional units (FU) in LCAs is one way to harmonize the environmental and nutritional impacts of food production and dietary patterns. They may include nutrient quantity, calories (i.e., per 100 kcal), amount of individual nutrient (i.e., grams of protein), composite scores of several nutrients, and nutrient quality (i.e., Digestible Indispensable Amino Acid Score). One methodology incorporated protein quality and quantity into LCAs to more comprehensively compare ASF and PBF in terms of protein content and quality and environmental impacts (Berardy et al., 2019). Another methodology has introduced an emissions per unit nutrient density metric to examine GHGE from food production to compare different types of food products based on their nutritional value rather than according to a singular nutrient or specific attribute like weight (Doran-Browne et al., 2015). Nutritional-based FU may be helpful in ensuring protein quality is not overlooked in the effort to deliver on healthy diets from sustainable food systems.

Conclusions regarding the environmental impact of food products can vary depending on the metrics used, each of which has strengths and limitations. Deciding which approach to use may depend on context – for example, nutrients of concern differ across populations and countries, as do trade-offs between the nutritional contribution and environmental impact of foods. Utilizing a variety of metrics to make comparisons between findings may allow for more comprehensive assessments to inform public health guidance.

It is critical that the dialogue surrounding food systems transformations consider the multiple domains of sustainability – health, environmental, economic, and societal. Traditionally, assessments of the sustainability of food production and consumption have focused on the environmental dimension. There is a lack of evidence on how shifts in food systems and dietary patterns will impact other dimensions of sustainability, which are all interconnected. Research is needed on the impact of consuming plant-based or -exclusive diets on health outcomes among population sub-groups with unique nutritional needs, such as older adults, so that the most vulnerable can make well-informed dietary choices. Evidence is also lacking on the ability of populations with low intakes of legumes, nuts, and seeds to increase consumption of these foods and on the affordability and availability of such PBF across regions and population groups of different socioeconomic and cultural backgrounds. Changes in the food market, including development of ultra-processed foods, lab-grown meat, plant-based beverages, and animal protein alternatives like insects and mycoproteins, require further exploration to evaluate their role. Without such evidence, the feasibility of recommendations that have shifted toward plant-based or -exclusive diets remains unclear.

There is also a need for more robust assessments and standardized metrics for food systems that capture the complexity of sustainability and the trade-offs across the domains. The utilization of a variety of metrics can help address the limitations and constraints of each individual metric and allow for the presentation of a more complete picture. This can provide more comprehensive information for decision-makers and the public seeking to understand how to optimize sustainable production and consumption of both ASF and PBF. Studies focused on evaluating the environmental impact of dietary patterns should consider the nutritional value of food choices and the nutrient requirements of a population, with attention placed on the dietary needs of population sub-groups, particularly those that are at at-risk. While nutritional-based FU can help achieve this nuance, additional questions must be asked to determine which FU would be the best to use, which can vary depending on the overall goals of the study. Further work is needed to expand the use of nutritional-based FU to include more types of dietary and environmental data, and economic considerations like affordability and accessibility.

It is recognized that plant-based diets may be the preferred dietary choice for many. However, it is important to consider how diets can be optimized in terms of meeting intake requirements for protein, AA, and key micronutrients like vitamins A and D, B-vitamins, calcium, iron, and zinc. PBF and ASF contain different quantities and combinations of nutrients and thus play complementary roles in the diet (Comerford et al., 2021). As ASF provide relatively higher quality protein it is important to consider their contribution to optimal health and nutrition outcomes. Moreover, it is critical to take a holistic perspective on the linkages between health, the environment, and socioeconomic factors when assessing the sustainability of food production systems, food choices, and dietary patterns to inform dietary recommendations.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

MP, JR-C, and GM developed article conceptualization. MP and GM supported the original draft preparation. All authors contributed to the review, editing, and subsequent draft preparation. All authors have read and agreed to the published version of the manuscript.

This study was funded by National Dairy Council, Rosemont, IL, United States. The funder was involved in the conceptualization, development, and publication of this review.

MP, JR-C, and GM are employees of National Dairy Council, Rosemont, IL, United States.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.