Part 1

The Origin and Evolution of the Paleo Diet

Abstract

Public interest in the Paleo diet (PD) has been growing since 2002, following the publication of Dr Loren Cordain's book The Paleo Diet. The premise of this diet is rooted in the ancestral hunter-gatherer pattern of eating, including lean meat, fish/seafood, fruits, vegetables (leaves and tubers), and nuts. Many followers adopt the diet as a healthier alternative to the standard American diet. Today, however, the interpretation and practice of the PD vary. To promote an increased understanding of the PD and its potential health benefits and to also help clinicians better engage with patients following the PD, researchers should work to (1) standardize how PD is defined in the literature and (2) examine the nutrient composition of the PD.

Public interest in the Paleo diet (PD) first began in 2002 after the publication of Dr Loren Cordain's¹ book, The Paleo Diet. A Paleo-type diet has also been described as the “Stone Age diet,” “caveman diet,” “hunter-gatherer diet,” or “primal diet.” The premise of the PD aligns with what was thought to be the original ancestral pattern of hunter-gatherers—allowing for lean meat, fish/seafood, fruits, vegetables (leaves and tubers), and nuts—foods of the preagricultural Paleolithic era. The Neolithic transition from hunting and gathering to agriculture emerged as people learned early agriculture practices and moved away from foraging to farming. Foods introduced in the Neolithic period, such as dairy and cereals, alongside foods introduced with the rise of the industrial period (ie, refined oils and processed foods) are omitted from this diet. Although a true Paleolithic era diet would require scavenging for food, practice of the modern PD is closer to the agricultural practices of the Neolithic age.² Modern-day Paleo originated in the 1960s from the earlier work of scientists, such as Murdock³ and Lee⁴ (Figure). Today, many interpretations of the PD exist not only in the popular press but also among scientists. In this review, we summarize the evolutionary origins of the PD and consider how earlier interpretations of the diet align with modern-day Paleo practice. By clarifying discrepancies in the overall nutritional composition of historic versus modern-day PD, practitioners, nutrition educators, and researchers alike will better understand the evolution of this dietary pattern and its nutritional composition.

FIGURE: History of Paleolithic diet.

EVOLUTIONARY ORIGINS OF THE PD

First, this section puts the evolutionary origins of the PD in context by providing a brief overview of the Stone Age period. There were 3 phases during this period: Paleolithic period (2.5 million years ago, referred to as the Old Stone Age), followed by the Mesolithic period (c. 25,000 years ago, the Middle Stone Age), and, finally, the Neolithic period (c. 10 000 years ago, the New Stone Age). Mesolithic period tools, which were predominately “chipped or flaked tools” for hunting, evolved to more “polished tools” in the Neolithic period (c. 10 000 years ago) as evidenced by ethnoarchaeological work.⁵ Up until this time, the food Paleo and Mesolithic humans consumed was either “hunted” (big game, small animals, fish, shellfish, insects) or “gathered” (fruits, leafy vegetables, tubers, seeds, nuts), thus the emergence of the term “hunter-gatherer.” As time transitioned from the Paleolithic to Neolithic periods, hunter-gatherers learned to domesticate wild animals and plant their own crops and began consuming what they grew or raised. As we describe the evolutionary history of the PD in this review (Figure), it is important to acknowledge that different perspectives of the PD exist throughout the fields of research, such as archaeology, genetics, anthropology, and more.^6–8 In this review, we focus on the evolution of dietary habits.

The publication of Walter Voegtlin's book in 1975, The Stone Age Diet,⁹ first introduced the notion that diets of our hunter-gatherer ancestors were nutritionally superior to modern-day diets. In Voegtlin's book, “modern diets” is referencing standard dietary patterns from the early 1940s to mid-1970s. Voegtlin, a gastroenterologist, aimed to answer the question: “Is modern man actually better or worse off nutritionally than was his Stone Age forebears?”⁹ In reference to the transition to diets emerging after World War I, he wrote in his book^9(p2): “All this would have been quite all right had not this new and fashionable trend junked most of the foods comprising man's diet from the dawn of humanity, and substituted refined, processed, uncooked plant foods and cheap carbohydrates for the animal fats and proteins he was designed to digest most easily.” And “Animal fats, vitamins, and roughage are removed from our foods and vegetable oils, synthetic vitamins and expensive bulking agents are used to replace them.”^9(p3) Overall, Voegtlin assumed that the Stone Age diet, with higher proportions of fat and protein, and little to no carbohydrate, was more beneficial to overall health and well-being compared with the modern diet, further stating “many of man's physical discomforts and health problems can be traced to this modern, highly scientific, yet faulty diet.”^9(p4) These discomforts encompass “gas, heartburn, frequent explosive bowel motions, bloating, and vague abdominal discomforts, [and although] previously considered as normal annoyance, are actually the protecting of a digestive tract that is being abused.”^9(p10) He claimed, “The evidence is incontestable that man's foods still should be those he naturally selected and even today (ie, in 1970s) digests with greatest ease—protein and fat with little or no carbohydrate.”⁹ Other researchers later debated on the range of plant and animal foods in these hunter-gatherer diets, with ranges in contribution from 26% to 99% animal foods and 1% to 74% plant foods.¹⁰

While Voegtlin was the original founder of the modern PD (Figure), it was not until 1985 that the term “Paleolithic nutrition” emerged and was coined by Boyd Eaton and Melvin Konner,¹¹ 2 doctors from Harvard Medical School. Similar to Voegtlin's belief, Eaton and Konner's “Paleolithic nutrition” was based on the premise that modern humans should eat like their ancestors to combat current maladaptation attributed to industrialized diets, which leads to increased chronic disease risk.¹² The hypothesis of “evolutionary discordance,” as coined by Eaton and Konner,¹¹ emphasizes that modern agriculture and the introduction of processed foods changed faster than man's ability to genetically adapt, thus leading to increased chronic disease risk. Various genetic adaptations that are probably related to diet become common in response to evolutionary pressures that predated agriculture, such as high copy numbers of the salivary amylase gene (AMY1), which is associated with high starch intake, or the NAT2 gene, which is involved in the activation/deactivation of toxic compounds commonly found in native plants, but is now associated with cancer and drug toxicity.^7,13 It is possible that these genetic adaptations that occurred early in human history may affect man's disease risk in modern times. Although not the focus of this review, some research has supported the belief that the shift in diet from the PD to the agricultural period and beyond to industrialization was too rapid for genetic adaptations.⁷ This emergence in “evolutionary discordance” gave rise to additional academic and cultural interest in evolutionary nutrition and ultimately led to the emergence of the “Paleo diet.”

MACRONUTRIENT AND NUTRIENT COMPOSITION OF THE DIET OF OUR PALEO ANCESTORS

Eaton and Konner¹¹ were the first researchers to estimate energy contribution from animal and plant sources, along with the macronutrient content in “hunter-gatherer populations” in their landmark paper “Paleolithic Nutrition—A Consideration of its Nature and Current Implications.”

Focusing on 6 groups of hunter-gatherers living in marginal habitats (!Kung, -Kade, San, and Hadza of Tanzania in Africa; Aborigines in Australia; and Tasaday in the Philippines), the estimated energy intake from animal sources varied from 20% among the Hadza of Tanzania to 42% among the Philippine Tasaday, suggesting that the vast majority of energy for these hunter-gatherers was derived from plants.¹¹ Fruits, seeds, roots, and tubers are estimated to be the largest contributors to carbohydrate intake within these diets, but vary greatly across populations, latitude, and ecological environments.¹¹ Of note, Eaton and Konner's¹¹ estimates were based on the anecdotal hunter-gatherer subsistence data generated from the original version of Murdock and colleagues'³ 1967 “The Ethnographic Atlas.” In brief, this atlas is a collection of data that have been continuously updated since 1937 and captures ethnographic data on the cultural practices of more than 1200 societies ranging from agricultural economies to political systems.³ Murdock and colleagues' original volume laid the foundation for testing the ideas of major trophological and social science theorists and has been widely used by anthropologists and sociologists.

Eaton and Konner¹¹ generated estimates of macronutrient and micronutrient intake using an assumed plant-to-animal energy ratio of 65:35, which was derived from the work of Lee.⁴ Lee was a Canadian anthropologist who conducted field work on the subsistence ecology of !Kung bushman of the Kalahari deserts of Botswana for 15 months in the mid-1960s and was the first to report of the largely plant-based diets among these bushmen.^14,15 The energy ratio derived from Lee was justified by Eaton and Konner as a “reasonable approximation of the nutrient composition of the typical diet of our ancestors during the early history and evolution of our species.” Based on this ratio, the estimated energy intake was 34% from protein (75% of total protein from animal protein and 24% from plant protein), 45% from carbohydrates, and 21% from fat.¹¹ Because of the relatively high proportion of plant foods, the estimated average crude fiber intake was 45.7 g/d.¹¹ Over the next decade, the authors updated their estimates based on latest data from literature recognizing wild plant and animal foods utilized by recent hunter-gatherers.¹⁶ By 1997, the authors published their fourth estimate of macronutrient composition. During this period, the number of animal species included in their model increased from 21 in 1985 to 85 in 1997, whereas the number of plant foods increased from 44 to 236 over the same period. Based on the 1997 model, the estimated energy intake was 41% carbohydrate, 37% protein, and 22% fat. Despite the contribution of more data, the macronutrient composition stayed relatively similar to earlier estimates, although the average crude fiber intake was estimated to be higher (approximately 100 g/d). In 1998, other researchers estimated energy contribution from plant foods among the Australian Aboriginal people and applied various subsistence patterns ranging from 20% to 80% energy from plant foods.¹⁷ Similarly, they found that their macronutrient ratio for 65% plant foods was similar to that estimated by Eaton and Konner (Table 1), although fiber intake was higher (130 g compared with 45.7 g).¹⁷

TABLE 1 - Estimated Nutrient Intake of Paleolithic Diets

	Eaton and Konner9	Eaton et al16	Konner and Eaton19
Total energy intake, kcal	3000	3000	2200
Protein, %	34	37	33
Carbohydrate, %	45	41	25
Fat, %	21	22	42
Saturated fat, %	6	6	9
Sodium, mg	690	768	795
Calcium, mg	1580	1956	890
Vitamin C, mg	392	604	559
Total dietary fiber, g	45.7	104	47

In 2000, Loren Cordain, a health scientist with a PhD in exercise physiology, published a research paper¹⁸ that also contained estimates of the energy sources and macronutrient intake of the PD. These estimates expanded upon the work of Eaton and Konner by including data from a revised ethnographic atlas, which included data on the contribution of hunted animals and fish to energy intake. Based on his research, Cordain et al estimated that the hunter-gatherer diet was largely derived from animal sources and may have ranged from 45% to 65% of daily calories from animal-based sources, in contrast to Eaton and Konner's estimate of 35% daily calories from animal sources. Cordain and colleagues suggested that Eaton and Konner underestimated the energy contribution from animal sources because the original data collected by Lee⁴ did not capture the hunting of animals or the contribution of fish to animal intake.¹⁸ The contribution of fish intake would be especially important in camps made along coastal areas that allowed convenient access to shellfish. In 2010, Konner and Eaton¹⁹ published another revision, including a discussion of Cordain's criticism. Table 1 illustrates the variations in estimates of macronutrient content from 1985 to 2010.^11,16,19 Ultimately, this work contributed to the premise of the modern-day PD.

Discrepancies in Historic Models

There is some debate over the macronutrient content of hunter-gatherer diets because models of intake were based on historic data with poor methodology for collecting dietary intake. In fairness, it is difficult to estimate diets of hunter-gatherers in the past, as well as in the 21st century, as they are highly dependent on resources available and the environmental conditions of camp sites. Great variability in nutrient intake can be attributed to latitude, climate, season, and geography (coastal, inland, grassland, forest).^20,21 Some researchers have attempted to account for changes in seasonality by collecting data across multiple seasons. For example, based on estimates derived from multiple 24-hour dietary recalls collected over 5 seasons, energy intake estimated in the diets of traditional Inuit in Alaskan natives was high in dietary fat (50%), moderate in protein (30%–35%), and low in total carbohydrate (15%–20%), primarily owing to their main food source being seal and whale meat.²² While this may have captured seasonal differences, in a different geographic region, such as the indigenous persons of New Guinea, the people had a diet consisting of approximately 70% energy intake from carbohydrates, mostly derived from potatoes and fruit.²³ This further highlights that even accounting for seasonal differences does not capture some of the inherent variation in dietary practices between indigenous peoples.

While the Ethnographic Atlas is a valuable resource for cross-cultural behavioral understanding and for identification of foods, the extrapolation of the data to estimate macronutrient composition is not equivalent to “actual intake.” For instance, it contains relatively sparse information on classification (eg, plant species) and quantity (eg, particular unit of measurement) of foods consumed in these societies.⁸ Furthermore, it remains unclear whether communities termed as “hunter-gatherers” in the Atlas were exclusively hunter-gatherers or included other, displaced agricultural people.²⁴ The locations of these tribes were rather isolated, and food behaviors were captured within the study camps but rarely included foods consumed outside of the study camp.²⁰ Interestingly, variation in diet composition among modern-day Hadza, an ethnic group living a hunter-gatherer lifestyle in Tanzania Africa, was recently observed in order to understand variation in the gut microbiota according to cyclical, seasonal fluctuation.²⁵ The study suggested that meat is largely consumed during the dry season, whereas honey, berries, and other foods are consumed during the wet season.²⁵ In conclusion, we must recognize that the hypothetical historic diet of “hunter-gatherers” is based on data generated solely from the Ethnographic Atlas. While there is agreement among many that the PD is based on human ancestral intakes from the Paleolithic period, accurate data on the composition of original hunter-gatherer diets are limited to the data collected in the Ethnographic Atlas.

MODERN PD

The PD increased in popularity following the publication of Cordain's book, The Paleo Diet, in 2002.¹ In fact, in 2013, the PD was the most searched diet using the Google search engine, and a 2014 survey of more than 500 registered dietitians predicted that the PD would be the top diet trend of the year.^26–28 More recent data suggest that popularity of the PD may be decreasing. Based on data from 1012 American adults in the 2019 Food and Health Survey, 3% of respondents claimed to follow the PD, which is down from 7% of respondents in the same survey conducted in 2018.²⁹ This diet, among others (eg, vegan/vegetarian, low-carbohydrate, “whole food” diets; ie, Mediterranean), emphasizes avoiding refined, highly processed foods that are high in sodium, sugar, and fat. The macronutrient energy composition of the PD as defined by Cordain's research is 25% carbohydrate, 33% protein, and 42% fat. In comparison, dietary intake estimated from the 2015–2016 National Health and Nutrition Examination Survey³⁰ illustrates that typical American adults are consuming ~50% of energy from carbohydrate, 16% of energy from protein, and 33% of energy from fat. Although the PD has heavily emphasized adhering to specific recommendations around macronutrient composition, consideration of the nutrient density of this diet is important. When compared with the standard American diet, the PD is typically higher in potassium, antioxidants, micronutrients, and fiber. ^16,31 Of note, the PD has also been identified in short-term intervention studies as providing inadequate levels of iodide ³² and calcium.³³

While the Food and Nutrition Board of the Institute of Medicine provides recommended ranges (ie, 20%–35%, 10%–35%, and 45%–65% of energy from carbohydrate, protein, and fat, respectively) with respect to macronutrient intake,³⁴ there is ambiguity around recommended intakes for popular dietary patterns. While vegan and vegetarian diets provide clear exclusions on animal products, the only clear and consistent guideline for adhering to a PD is the allowance of lean meat and fish. For example, with the PD, fruit is conditional to allowing only berries or specific amounts of dried fruit, whereas vegetables are conditional to differences in legume and potato inclusion. Table 2 illustrates the variation in the implementation of the PD across randomized controlled trials.

TABLE 2 - Comparison of Paleolithic Diet Definitions Used Among Studies

	Lean Meat	Fish/Seafood	Eggs	Vegetables	Fruits	Nuts	Dairy	Cereals	Legumes	Added Salt	Refined Fat	Sugar
Lindeberg et al35a	✓	✓	✓	✓	✓	✓	X	X	X			X
Osterdahl et al33b	✓	✓		✓	✓	✓	X	X	X	X		X
Jönsson et al36c	✓	✓	✓	✓	✓	✓	X	X	X	X	X	X
Boers et al37d	✓	✓	✓	✓	✓	✓	X	X	X	X	X	X
Mellberg et al38e	✓	✓	✓	✓	✓	✓	X	X		X	X	X
Masharani et al39f	✓	✓	✓	✓	✓	✓	X	X	X
Pastore et al40g	✓	✓	✓	✓	✓	✓	X	X	X
Genoni et al41h	✓	✓	✓	✓	✓	✓	X	X	X
Otten et al42i	✓	✓	✓	✓	✓	✓	X	X	X	X	X	X
Irish et al43j	✓	✓		✓	✓		X		X

✓ = included, X = excluded, blank = not mentioned.

^aAvoid bakery products, soft drinks, and beer. Restrictions of the following: “eggs (one or fewer per day), nuts (preferentially walnuts), potatoes (2 or fewer medium size per day), rapeseed or olive oil (one or fewer tablespoons per day).”

^bAllowed “dried fruit (ad lib 2 days per week), salted seafood (1 meal per week), fat meat (1 meal per week), potatoes (2 medium size per day), honey (used in marinade once per week), cured meats (as entree once per week), mineral water (only when drinkable tap water was not available). Prohibited food: charcuterie products (eg, sausages, pate, and so on), canned food (except tomatoes, see above), and all forms of candy, ice cream, sorbet, soft drinks, juices, syrups, liquor.”

^cAvoid candy, soft drinks, and beer. Restrictions of the following: “eggs (≤2 per day), nuts (preferentially walnuts), dried fruit, potatoes (≤1 medium size per day), rapeseed or olive oil (≤1 tablespoon per day), wine (≤1 glass per day).”

^dAllowed up to 2 cups of coffee or black tea per day.

^eNo additional guidelines.

^fAllowed tree nuts, canola oil, mayonnaise, and honey. Avoid products containing potassium chloride. “Some foods, such as mayonnaise, carrot juice, and domestic meat, were not consumed by hunter-gatherers but contain the general nutritional characteristics of preagricultural foods.”

^gNo more than ½ cup of potato per day, no more than 1 oz of dried fruit per day, and no more than 4 oz of wine per day. No limit on egg consumption.

^hAllowed small amounts of olive or coconut oils. Dairy was replaced with almond milk; black coffee and tea were allowed. Corn and white potatoes were prohibited, and dried fruit was limited to 1 tablespoon per day.

ⁱException to added salt: canned fish and cold cuts. Restrictions included “eggs (1-2 per day but a maximum of 5 per week), potatoes (1 medium size per day), dried fruit (130 g/d), and nuts (60 g/d). Rapeseed or olive oil (maximum 15 g/d) and small amounts of honey and vinegar were allowed as flavoring in cooking. Participants were instructed to drink mainly still water. Coffee and tea were restricted to a maximum of 300 g/d, and red wine to a maximum of 1 glass per week.”

^jMaximum of 4 servings of nuts (raw, sprouted). Goal of ¼ teaspoon seaweed, 1 teaspoon blue green algae, 2 tablespoons olive or flax or hemp or walnut oil, and 3 tablespoons nutritional yeast. No eggs or potatoes allowed. Gluten-free grains/starches limited to 2 times per week.

For the consumer, there are several books and thousands of resources online with details on preparing Paleo meals and guidance on how to follow a PD. Most sources promote the inclusion of lean meat, fish and seafood, eggs, nonstarchy vegetables, nuts and berries, and the exclusion of grains, legumes (except green beans, snow peas, sugar snap peas), fruits, refined vegetable oils, dairy, and alcohol.⁴⁴ Depending on the resource, fruits are allowed in moderation, although not considered “essential to the PD because there are no nutrients in fruits that are not in vegetables.”⁴⁵ In some cases “grass-fed butter,”⁴⁵ “raw dairy,”⁴⁶ or full-fat pasteurized dairy ⁴⁷ are permitted in popular cookbooks, and flexibility also exists with respect to sugar and the use of artificial sweeteners, with some moderate allowance for natural sugars (honey, molasses, maple syrup, minimally processed cane sugar, whole dates, fruit juice) and Stevia.

It is clear that there is great variation with respect to what constitutes a PD, as well as how the modern PD is practiced, and this variation likely contributes to confusion among consumers and educators. In some instances, “disease-specific” PD modifications have been developed. For example, a modified PD was created by Dr Wahls for people with multiple sclerosis,⁴⁸ whereby nightshade vegetables (ie, tomatoes, eggplants, white potatoes, peppers) and eggs are excluded. Following Dr Wahl's modified PD, fiber and calcium are predicted to be 2 main nutrients of concern when compared with the recommended healthy US eating pattern.⁴⁹ The autoimmune PD was first described by Sarah Ballantyne in her book, The Paleo Approach, published in 2014, which includes limiting certain foods the authors proposed to trigger inflammatory reactions such as nuts and seeds.⁵⁰ In 2013, Kresser⁵¹ published his own Your Personal Paleo Code that allows for the reintroduction of dairy and legumes, assuming these foods can be tolerated by the individual.

Several randomized clinical trials have been conducted to evaluate the health benefits of the PD compared with other healthy diets, but the evidence is mixed, in part due to variation in study design, sample characteristics, and administration of the diets.^52–54 A recent publication by Ge and colleagues⁵⁵ on the comparison of the macronutrient composition and health effects of several popular diet programs highlights the wide variation in macronutrient composition across a number of diets, including difference in PD practices. In their review, the PDs administered in 2 intervention studies^35,38 were different in macronutrient composition. Typically, a PD is considered a low-carbohydrate diet, and the study by Mellberg and colleagues³⁸ fell within this classification—less than 40% carbohydrate (C), 30% protein (P), and 30% to 55% fat (F). However, the energy derived from carbohydrate sources was higher in the study by Lindeberg et al³⁵ (55%-60%) and thus classified as a moderate macronutrient diet. Table 3 shows the variation in the actual macronutrient composition of a modern-day PD among 10 randomized controlled studies in the review. The intervention period ranged from 2 weeks to 24 months, with the majority of the feeding designs being ad libitum diets following instruction on the principles of the PD. In addition, the majority of the intervention diets led to a reduction in caloric intake, and thus, it remains unclear if the purported health benefits of the PD in these studies were attributed to reductions in energy intake, to certain characteristics of the PD (such as avoidance of refined carbohydrates), or to nutrient composition of the PD itself. Although the historic PD purported that dietary fiber intakes were high, estimated fiber intake in most of these intervention studies was less than 30 g/d. Only 1 intervention study by Masharani and colleagues³⁹ reported a high-fiber intake (42 g daily), and this may be attributed to the diet intervention being higher in both carbohydrate (58% energy intake from C) and total caloric intake (3001 kcal).

TABLE 3 - Paleolithic Diet Literature With Reported Energy, Macronutrient, and Fiber Intakes^a

	Study Design/Country	Diet Protocol	Diet Intervention	Diet Duration	Energy Intake After Intervention, kcal/d	Dietary Fiber After Intervention, g/d	Carbohydrate, %	Protein, %	Fat, %
Lindeberg et al35	RCT, parallel, Sweden	PD	Ad libitum	12 wk	1344 (521)	21.4 (13.2)	40.2 (8.3)	27.9 (6.8)	26.9 (6.4)
Osterdahl et al33b	RCT, run-in intervention, Sweden	PD	Ad libitum	3 wk	1584 (208)	32	40 (4.9)	23.9 (5.4)	35.8 (4.1)
Jönsson et al36	RCT, crossover, Sweden	PD	Ad libitum	12 wk	1581 (295)	21 (8)	32 (7)	24 (3)	39 (5)
Boers et al37	RCT, parallel, the Netherlands	PD	Meals provided	2 wk	2080	34	32	24	41
Mellberg et al38c	RCT, parallel Sweden	PD	Ad libitum	24 mo	2000 ± 59.0 [8.37 ± 0.25] pre −401 ± 89.5 [−1.68 ± 0.38] change after 24 mo	24.6 ± 1.03 pre −2.70 ± 1.36 change after 24 mo	46.2 ± 0.67 pre −12.7 ± 1.68 change after 24 mo	17.1 ± 0.32 pre 4.79 ± 0.72 change after 24 mo	33.4 ± 0.60 pre 7.02 ± 1.40 change after 24 mo
Masharani et al39	RCT, parallel, USA (California)	PD	All meals prepared by kitchen, eaten at home or on site	3 wk	3001.5	42	58	19	27
Pastore et al40	RCT, cross-over, USA (Michigan)	PD	Ad libitum	8 mo	Women 1719.7 (262.7) Men 2101.8 (382.2)	NR	23	37	40
Genoni et al41	RCT, parallel, Australia	PD	Ad libitum	4 wk	1414 (347)	24.6 (7.2)	27.8 (8.05)	26.8 (7.2)	39.8 (9.6)
Otten et al42	RCT, parallel, Sweden	PD	Ad libitum	12 wk	2022 (1583-2268) pre −291 (−587, −66) change after 12 mo	23 (15-30) estimates at 12 mo	31 (24-39) estimates at 12 mo	24 (19–27) estimates at 12 mo	42 (37–48) estimates at 12 mo
		PD + Ex	Ad libitum	12 wk	1595 (1428-2257) pre −530 (−863, −157) change after 12 mo	14 (13-17)	27 (24-29)	26 (22-29)	45 (37–47)
Irish et al43	RCT, parallel, USA (Iowa)	Modified Paleolithic diet	Ad libitum	12 wk	NR	NR	NR	NR	NR

Abbreviations: Ex, exercise; PD, Paleo diet; RCT, randomized controlled trial.

^aMean (SD) or 95% confidence interval is reported. Prediet estimates and difference after intervention are reported in some cases because of differences in data reporting.

^bFourteen of 20 completed the dietary intervention, but because of computer problems (as reported in the publication), there are nutrition data from only 6 participants.

^cAlthough the authors explained the Paleolithic diet as 30% carbohydrate, 40% fat, and 30% protein to the participant, the actual was reported in this table.

To date, there are few data on the nutritional composition of the diet of self-identified PD followers, and ambiguity remains in modern reconstruction of the PD. Typically, observational studies do not capture self-identified Paleo followers; however, researchers have characterized study participants as adhering to a Paleo-type diet based on self-reported dietary intakes derived from 3-day records. In a recent observational study by Genoni et al,⁵⁶ study recruitment defined PD followers as those self-identifying for greater than 1 year and consuming no more than 1 serving of grains and dairy products per day during this time. This classification of the PD strictly deviates from the traditional principles to avoid grains and dairy. Because of individual differences in the interpretation of the PD by followers themselves, the authors had to create a scoring to classify individuals as “pseudo-Paleo” if they self-identified as PD followers but consumed grains and dairy products daily, further demonstrating the differences in adherence and understanding of the PD. In this study, estimated macronutrient composition (C:P:F ratio) was 17:24:53 among strict Paleo followers, 16:20:57 among pseudo-PD followers, and 38:19:37 among a control group eating a healthy diet.⁵⁶ Of the observational studies that have captured intake of PD followers, reported total daily dietary fiber intakes were in the range of 13 to 14 g/1000 kcal.^38,57–59 It is important to consider that in both intervention and observational studies, the macronutrient estimates not only stray from Cordain's modern-day hypothetical estimates (ie, 25:33:42), but also study findings relating these diets to a variety of outcomes may also be confounded by differences in macronutrient composition and the type of carbohydrate or fat consumed.

CONCLUSION

Modern followers of the PD adopt historic principles that vary substantially when translated into actual practice. The original adopters of the PD, beginning with the hunter-gatherers, lived in an era spanning from 3 million years ago to the end of the Ice Age. Defining the PD depends on temporal data sweeping over geologic time, compared with food avoidance or macronutrient guidelines that define other diets of modern practice (eg, vegan, low-carbohydrate, Atkins, ketogenic, Mediterranean). Nevertheless, modeled constructs of prehistoric and modern hunter-gatherers offer promising estimates of PD constituents. More than other diets, the dietary attributes of the historic PD necessarily shifted among seasons and terrain covered by nomadic peoples. Defining the historic diet of a PD follower requires that we extend early 1-point models to accommodate shifting food availabilities over the course of a typical year in contemporary hunter-gatherers. Developing robust models to test PD in modern settings is an area of continued interest. We need to better define PD, specifically what it means to adhere to a PD in the modern era. Congruency around the definition of a PD among researchers will not only help standardize diets allowing comparison between studies, but also enhance our understating of the health implications of long-term adherence to the PD. Understanding nutrient composition is crucial if clinicians are to advise patients about the health benefits and concerns associated with the PD.