6.1c. The analysis shows that the human appetite maintains absolute protein intake relatively tightly, with non‐protein energy intake varying more passively with dietary macronutrient ratios.
Source: Adapted from Raubenheimer and Simpson [19].
Further evidence for protein leverage has come from analysis of population [66–68] and cohort data, including most recently in children with obesity [69]. Discussion of the theoretical foundations of the protein leverage hypothesis as well as clarification of some misconceptions that have arisen since its first publication can be found in Raubenheimer and Simpson [55].
Because protein comprises the minor proportion of total energy intake in humans (~15%), regulating to a target amount of protein means that even a 1% decline in dietary protein can “leverage” a large increase in total energy intake as fats and/or carbohydrates [53,54]. However, dilution of protein in the diet can be more substantial than this across certain sectors of a population; for example, Martínez Steele et al. [68] found in an analysis of the NHANES data (2009–2010) (further discussed below) that dietary protein varied between 18.2 and 13.3% across quintiles of the USA population. As predicted by protein leverage, total energy intake rose (in this case by 8.5%), whereas absolute protein intake remained near‐constant.
Figure 6.6 Effect on protein leverage of increased protein requirements relative to non‐protein energy. T1 represents the reference intake target, and T2 an increased P requirement due, for example, to increased amino acid‐based gluconeogenesis. On the same protein‐dilute diet (blue nutritional rail), the increased P requirement amplifies protein leverage, such that the over‐ingestion of NPE is greater for T2 than T1 (NPE + 2 > NPE + 1).
Source: Adapted from Raubenheimer and Simpson [55].
Just as even small variations in dietary percentage protein among different individuals or sectors within populations can have large effects on total energy intake, the same is the case for variation in the target intake for protein. The higher the protein target, the greater the over‐consumption of energy required on a given low‐protein, high‐energy diet to reach the protein target (Fig. 6.6). Consequently, increases in protein requirements across the life‐course, with lifestyle, health status, ethnicity, genotype, and with epigenetic influences are predicted to increase the risk of overweight and obesity in low percent protein environments [53,70,71]. Some implications of this are discussed below.
Finally, why might the human appetite system have evolved to regulate protein intake so tightly and to prioritize protein intake above that of other macronutrients and total energy? The answer likely lies in the need to balance the costs of eating too little protein (e.g. to growth and reproduction) against the costs of consuming too much protein (e.g. accelerated rates of aging and later life health impacts) [72,73].
Human nutritional ecology
The evidence thus suggests that there is nothing peculiar nor unique in the biological responses of humans to varying diet composition. Like other species, from insects to primates, macronutrients exert strong control over human ingestive behavior, with specific appetites interacting to select an intake target. The selected intake target, comprising ~15% of energy from protein, falls midway within the range recorded for other species of living apes, which provides reason to suspect that it is a natural part of human biology. When eating diets with macronutrient ratios that prevent the target ratio from being eaten humans, like many other species, prioritize meeting the regulatory target for protein, whether this involves over‐eating fats and carbohydrates (on protein‐dilute diets) or under‐eating fats and carbohydrates (on high‐protein diets).
A strong benefit of this model, which underpins the protein leverage hypothesis of obesity, is that it uses human biology as a guide to identifying the most important ecological questions around the global obesity epidemic. Importantly, there already are public health frameworks in place with which to integrate the protein leverage model and help answer those questions to illuminate the nutritional ecology of human obesity.
Some relevant frameworks from public health nutrition
An increasingly prominent framework in public health research is that of the “food environment.” This concept first emerged in ecology to refer to the abundance and nature of foods in the environment in which an animal forages [e.g. 74]. In public health, the term was adopted in the context of socio‐ecological theory, around the realization that health‐related behaviors are determined by the interaction of personal and environmental factors [75,76]. It continues to be defined in this way, referring not just to foods available, but also to the factors within the environment that influence the accessibility and choice of foods: “Food environments comprise the foods available to people in their surroundings as they go about their everyday lives and the nutritional quality, safety, price, convenience, labeling and promotion of these foods” [75].
Food environments can be thought of as the consumer interface with a broader system, the “food system,” which comprise “all the elements (environment, people, inputs, processes, infrastructures, institutions) and the activities that relate to the production, processing, distribution, preparation, and consumption of food, as well as the output of these activities, including socioeconomic and environmental outcomes” [77]. In this respect they are closely parallel to the concept of the “ecosystem,” developed in ecology as a holistic framework for understanding the influence of both biological (such as competition and predation) and non‐biological (such as climate and soil nutrients) factors on organisms [78,79].
A second relevant public health framework is the “nutrition transition” concept [80,81]. This framework focuses on the stages through which human societies transition in terms of food and beverage consumption and patterns of physical activity. Although conceived within a public health context, it has an ecological‐evolutionary underpinning in that it emphasizes mismatch between evolved human biological traits, such as preference for sweet tastes, and changes to human food environments that in terms of health outcomes are incompatible with these traits, such as abundant availability of sweet foods in obesogenic food environments.
Both of the above concepts are highly relevant to the protein leverage hypothesis because they frame the important question this hypothesis highlights around the ecological drivers of obesity: what has changed in human food environments concurrent with the rise of obesity that might dilute protein, causing the strong protein appetite to leverage over‐consumption of energy? There is now widespread agreement that this is related to the industrialization and globalization of food systems, in which diets rich in legumes and other vegetables, fruits, and coarse grains are replaced by diets low in fiber and micronutrients, and rich in highly processed oils and carbohydrates, and increased consumption of animal‐sourced food [82]. In the United States and many other higher‐income countries, this transition began in the 1960s, and has happened more recently in many lower‐middle‐income countries.
A third public health framework that can play a significant role in understanding how nutrition transitions create obesogenic environments is the NOVA food classification system. The NOVA system classifies foods into four categories according to the extent and purpose of industrial processing [83]. Group 1 foods are unprocessed or minimally processed, including natural foods that have been altered by, for example, drying, roasting, boiling, pasteurization, or the removal of unwanted parts. Group 2 comprises processed culinary ingredients, which are seldom consumed alone but combined with group 1 foods in the preparation of meals, for example salt, vegetable oils, honey, sugar, starches, and butter. Group 3, “processed foods,” are relatively simple products made by adding group 2 items, such as sugar, salt, or oil to unprocessed (group 1) foods, with the main aim of increasing their durability or enhancing their sensory qualities. Group 4, “ultra‐processed