Cutting Through the Paleo Hype: the Evidence for the Palaeolithic Diet

CLINICAL

Cutting through the Paleo hype: The evidence for the Palaeolithic diet

Christopher E Pitt

Background eneral practitioners (GPs) are groups are optimal for human health – commonly asked about popular ‘Could I eat this if I were naked with a General practitioners (GPs) are G diets. Fad diets come and go, some sharp stick on the savanna?’.9 Therefore, commonly asked about popular diets. gaining more traction within the public meat, fruit and vegetables are acceptable, The Palaeolithic diet is both highly sphere than others. One of the most but grains and dairy products are not.10 popular and controversial. controversial diets in recent times is the Such views have drawn criticism from Objective Palaeolithic diet, otherwise known as the anthropologists, who argue that there is Stone Age diet, or simply as Paleo. no blanket prescription of an evolutionarily This article reviews the published Even without controversy, the appropriate diet, but rather that human literature to establish the evidence for Palaeolithic diet has been increasing in eating habits are primarily learned through and against the Palaeolithic diet. popularity over the past few years. The behavioural, social and physiological diet has been the subject of intense mechanisms.11 Other commentators have Discussion criticism by health professionals because noted that the claims of the Palaeolithic of wellness bloggers’ and celebrity chefs’ diet are unsupported by scientific and The Palaeolithic diet remains exaggerated claims about its purported historical evidence.12 The Palaeolithic diet’s controversial because of exaggerated claims for it by wellness bloggers and effects – for example, that the Palaeolithic anthropological validity notwithstanding, celebrity chefs, and the contentious diet could prevent or cure polycystic is there scientific support of the various evolutionary discordance hypothesis ovary syndrome, autism, mental illness, health claims made of it? Pragmatically 1 on which it is based. However, a dementia and obesity. speaking, is a diet sans dairy and refined number of underpowered trials have Does the published medical literature carbohydrates beneficial, even if it is not suggested there may be some benefit support the vast and extravagant claims historically accurate? to the Palaeolithic diet, especially made by the Palaeolithic diet’s celebrity in weight loss and the correction of proponents? Should GPs recommend Published evidence on the metabolic dysfunction. Further research the Palaeolithic diets to their patients, or Palaeolithic diet is warranted to test these early findings. caution them? This article seeks to review While proponents of the Palaeolithic diet GPs should caution patients who are the published clinical research on the claim that it is evidence-based, there are on the Palaeolithic diet about adequate Palaeolithic diet. only a limited number of controlled clinical calcium intake, especially those at higher trials comparing the Palaeolithic diet to risk of osteoporosis. Why Paleo? The rationale accepted diets such as the diabetic or for the Palaeolithic diet Mediterranean diets (Table 1). The rationale for the Palaeolithic diet In 2007, Lindeberg et al performed stems from the evolutionary discordance a randomised controlled trial of the hypothesis – that human evolution ceased Palaeolithic versus Mediterranean-like diet 10,000 years ago, and our Stone Age in 29 patients with ischaemic heart disease genetics are ill-equipped to cope with our and impaired glucose metabolism over modern diet and lifestyle, leading to the 12 weeks. Both groups lost approximately ‘diseases of civilisation’.2–8 Thus, only foods the same amount of weight; however, the that were available to hunter–gatherer Palaeolithic group showed a significantly

decreased waist circumference and and no statistically significant change in slightly and there was a stark rise in CRP, improved glucose sensitivity.13 In 2009, C-reactive protein (CRP).14 although it did not reach significance. Jönsson et al expanded on the study Osterdahl and colleagues published a The authors noted that the Palaeolithic by Lindeberg et al. They published a pilot study in 2008, in which 14 healthy diet was significantly lower in calcium randomised crossover pilot study of 13 volunteers were placed on the Palaeolithic compared to the subjects’ pre-study diet.15 patients with type 2 diabetes, randomised diet over the course of three weeks. Frassetto et al performed a metabolically to either the Palaeolithic or diabetic-like diet Six participants gave a complete dietary controlled study in 2009 in nine non- over two consecutive three-month periods. assessment. Across all participants, there obese, sedentary, healthy volunteers, Compared with patients on the diabetic-like was a significant mean weight loss of comparing the Palaeolithic diet to their diet, patients on the Palaeolithic diet showed 2.3 kg over the three weeks and a mean usual diet. The Palaeolithic diet led to improved HbA1c, diastolic blood pressure, decrease in waist circumference by significant reductions in blood pressure lipid profile, weight and waist circumference, 0.5 cm. Systolic blood pressure improved with improved arterial distensibility, insulin

Table 1. Summary of selection methodology and reviewed articles*

Lindeberg et al (2007) 29 patients; IHD and impaired RCT – Paleo versus Weight – no difference; Paleo – glucose metabolism Mediterranean-like diet for decreased WC, increased glucose 12 weeks sensitivity

Osterdahl et al (2008) 14 healthy volunteers Pilot Paleo – mean weight loss of 2.3 kg, decreased WC 0.5 cm, improved SBP, increased CRP (non-significant) and significantly lacked calcium

Jönsson et al (2009) 13 patients; T2DM RCT – crossover, two Paleo – improved HbA1c, DBP, lipid consecutive three-month profile, weight, WC; no significant periods, Paleo versus change in CRP diabetic-like diet

Frassetto et al (2009) Nine non-obese, sedentary Metabolically controlled – Paleo – improved SBP, DBP, arterial volunteers Paleo versus diet as usual distensibility, insulin sensitivity, plasma lipids

Jönsson et al (2013) 13 patients; T2DM RCT – crossover, two Paleo – increased satiety for energy, consecutive three-month energy density and glycaemic load. periods, Paleo versus Difficult to adhere to. diabetes diet

Boers et al (2014) 34 patients; metabolic RCT – two weeks, Paleo Paleo – decreased SBP, DBP and syndrome versus isocaloric diet based weight (1.32 kg); improved lipid on Dutch Health Council profile. No change to intestinal guidelines permeability, inflammation, salivary cortisol

Whalen et al (2014) University of Minnesota Case controlled – incidence Similar non-significant decreased Cancer Prevention Research colorectal polyps, Paleo incidence colorectal polyps Unit data versus Mediterranean diet

Mellberg et al (2014) 70 patients; females who RCT – two years, Paleo Paleo – increased weight loss at six were obese and post- versus Nordic Nutrition months, no difference at two years. menopausal Recommendations diet Improved BP, CRP and lipid profile in both groups. No difference glucose/ insulin both groups

Bligh et al (2015) 24 healthy male volunteers Acute, double-blinded, RCT – No significant difference in glucose two different Paleo diets and and insulin responses between WHO-based reference meal different meals

*Controlled human trials of the Paleolithic diet selected for review. Papers retrieved from Pubmed search for ‘Paleolithic Diet’ limited to ‘humans’ { (‘diet, paleolithic’[MeSH Terms] OR (‘diet’ [All Fields] AND ‘paleolithic’ [All Fields]) OR ‘paleolithic diet’ [All Fields] OR (‘paleo’ [All Fields] AND ‘diet’ [All Fields]) OR ‘paleo diet’ [All Fields]) AND ‘humans’ [MeSH Terms]} CRP, C-reactive protein; DBP, diastolic blood pressure; IHD, ischaemic heart disease; Paleo, the Palaeolithic diet; RCT, randomised controlled trial; SBP, systolic blood pressure; T2DM, type 2 diabetes mellitus; WC, waist circumference; WHO, World Health Organization

36 REPRINTED FROM AFP VOL.45, NO.1–2, JAN–FEB 2016 © The Royal Australian College of General Practitioners 2016 PALAEOLITHIC DIET CLINICAL

sensitivity and plasma lipids, all unrelated to research program at the University of and cholesterol, whereas there were no body weight.16 Minnesota Cancer Prevention Research changes in fasting glucose and insulin.20 In 2013, Jönsson et al studied the satiety Unit between April 1991 and April 1994 In a double-blind randomised controlled of the Palaeolithic diet, compared with the to analyse the incidence of colorectal trial of 24 healthy male volunteers in 2015, diabetic diet, in a randomised crossover polyps versus the diet history given on Bligh et al compared the acute satiety and trial of 13 patients with type 2 diabetes. a standardised dietary questionnaire. gut hormone responses of two Palaeolithic- The Palaeolithic diet resulted in greater They found that for both the Palaeolithic type meals with a reference meal based satiety quotients for energy, energy density and Mediterranean diets, there was no on the World Health Organization’s and glycaemic load per meal. While the statistically significant change in risk.19 Preparation and use of food-based dietary Palaeolithic diet was more satiating per Mellberg et al published a longer term guidelines. Both Palaeolithic meals were calorie than the diabetic diet, they also trial on the Palaeolithic diet in 2014. predominantly plant based. The first noted that it was difficult to adhere to.17 Their study was a randomised controlled Palaeolithic meal was based on estimated In 2014, Boers et al compared the trial of 70 women who were obese and range ratios for protein and fat that was Palaeolithic diet with an isoenergetic, post-menopausal, and they compared considered typical of hunter–gatherers, healthy control diet based on Dutch Health the Palaeolithic diet to a reference and contained no cereals or dairy products. Council guidelines in 32 subjects with diet based on the Nordic Nutrition The second Palaeolithic meal was made metabolic syndrome. The Palaeolithic diet Recommendations over a two-year with identical plant-based ingredients to resulted in lower systolic and diastolic period. The end points were body fat the first Palaeolithic meal, but normalised blood pressure, and improved plasma percentage calculated by dual-energy to the reference meal for caloric and lipid profile. Despite both diet arms X-ray absorptiometry (DEXA) scan, and macronutrient equivalence. There was no being isocaloric, bodyweight decreased the level of metabolic markers. The significant difference in the response of in the Palaeolithic group, compared Palaeolithic group lost significantly more glucose and insulin between the meals.21 with the reference (–1.32 kg; P = 0.012). weight than the group on the reference No significant difference was found in diet at six months, although this was not Reconciling the research intestinal permeability, inflammation and sustained at the 24-month mark. The Looking at the studies as a whole, the salivary cortisol as secondary outcome Palaeolithic diet group lost more body Palaeolithic diet was often associated measures.18 fat and lean tissue than the reference with increased satiety, independent of Whalen et al in 2014 used data collected group. Both groups showed similar caloric or macronutrient composition, as part of a previous case-controlled improvements in blood pressure, CRP along with improvements in body weight, waist circumference, blood pressure and lipid profiles. However, the studies Table 2. Comparison of the current Australian dietary guidelines were short, heterogeneous in design recommendations24 with the Palaeolithic diet15 and underpowered. The strongest of Australian dietary guidelines Palaeolithic diet the studies was by Mellburg et al, who showed no long-term differences between Enjoy a wide variety of nutritious foods from participants on the Palaeolithic diet and the following five groups every day those on the control at 24 months.20 Plenty of vegetables, including different types In the studies that measured and colours, and legumes/beans Ad libitum fresh vegetables and fruit inflammatory markers, there was no Fruit significant difference as a result of Grain (cereal) foods, mostly wholegrain consuming the Palaeolithic diet. Adherence and/or high cereal fibre varieties, such as All cereals/grain products prohibited, and palatability were common issues bread, cereals, rice, pasta, noodles, polenta, including maize and rice raised about the Palaeolithic diet. Some couscous, oats, quinoa and barley studies reported improvements in plasma Ad libitum lean meats and poultry, fish, Lean meats and poultry, fish, eggs, tofu, nuts glucose or other markers of glycaemic eggs, tofu, nuts and seeds, but all legumes and seeds, and legumes/beans control,10,14,16 though some did not,20,21 prohibited including the study of Bligh et al, which Milk, yoghurt, cheese and/or their alternatives, was a high-quality, laboratory-controlled mostly reduced fat (reduced fat milks are not All dairy products prohibited study. It showed no significant difference suitable for children younger than two years) in glucose and insulin levels for the Ad libitum water (mineral water allowed if tap Drink plenty of water Palaeolithic diet, compared with their water unavailable) reference meal.21

Other factors should be considered Author 15. Osterdahl M, Kocturk T, Koochek A, Wandell when thinking of the Palaeolithic diet in the Christopher E Pitt MBBS, FRACGP, General PE. Effects of a short-term intervention with a Practitioner, Carseldine, QLD. [email protected] paleolithic diet in healthy volunteers. Eur J Clin broader patient context. Modelling of the Nutr 2008;62:682–85. Competing interests: None. cost of the Palaeolithic diet suggests that 16. Frassetto LA, Schloetter M, Mietus-Synder Provenance and peer review: Not commissioned, M, Morris RC Jr, Sebastian A. Metabolic and it is approximately 10% more expensive externally peer reviewed. physiologic improvements from consuming a than an essential diet of similar nutritional paleolithic, hunter-gatherer type diet. Eur J Clin value, which may limit the Palaeolithic Acknowledgements Nutr 2009;63:947–55. I would like to thank Drs Edward Osborne and 1 7. J önsson T, Granfeldt Y, Lindeberg S, Hallberg diet’s usefulness for those on a low Genevieve Yates for their constructive editorial AC. Subjective satiety and other experiences income.22 Calcium deficiency also remains feedback. of a Paleolithic diet compared to a diabetes diet in patients with type 2 diabetes. Nutr J a significant issue with the Palaeolithic References 2013;12:105. diet; the study by Osterdahl et al in 2008 18. Boers I, Muskiet FA, Berkelaar E, et al. 1. Duck S. Paleo diet: Health experts slam chef showed that the calcium intake of the Paleo Favourable effects of consuming a Palaeolithic- Pete Evans for pushing extreme views. Sunday type diet on characteristics of the metabolic diet was about 50% of the recommended Herald Sun. 7 December 2014. syndrome: A randomized controlled pilot-study. dietary intake.15 Uncorrected, this could 2. Konner M, Eaton SB. Paleolithic nutrition: Lipids Health Dis 2014;13:160. Twenty-five years later. Nutrition in clinical increase a patient’s risk of osteoporosis.23 19. Whalen KA, McCullough M, Flanders WD, practice: Official publication of the American Hartman TJ, Judd S, Bostick RM. Paleolithic and Patients on the Palaeolithic diet should be Society for Parenteral and Enteral Nutrition Mediterranean diet pattern scores and risk of 2010;25:594–602. referred to an accredited practising dietitian incident, sporadic colorectal adenomas. Am J 3. Eaton SB, Eaton SB 3rd, Konner MJ. Paleolithic Epidemiol 2014;180:1088–97. for individualised medical nutrition therapy nutrition revisited: A twelve-year retrospective 20. Mellberg C, Sandberg S, Ryberg M, et al. Long- (Table 2). on its nature and implications. Eur J Cli Nutr term effects of a Palaeolithic-type diet in obese 1997;51:207–16. Overall, conclusions about the postmenopausal women: A 2-year randomized 4. Eaton SB, Konner M. Paleolithic nutrition. trial. Eur J Clin Nutr 2014;68:350–57. effectiveness of the Palaeolithic diet should A consideration of its nature and current 21. Bligh HF, Godsland IF, Frost G, et al. Plant- be considered cautiously. Positive findings implications. New Engl J Med 1985;312:283–89. rich mixed meals based on Palaeolithic diet 5. Kuipers RS, Luxwolda MF, Dijck-Brouwer DA, principles have a dramatic impact on incretin, should be tempered by the lack of power et al. Estimated macronutrient and fatty acid peptide YY and satiety response, but show of these studies, which were limited intakes from an East African Paleolithic diet. Br J little effect on glucose and insulin homeostasis: by their small numbers, heterogeneity Nutr 2010;104:1666–87. An acute-effects randomised study. Br J Nutr 6. Eaton SB, Konner MJ, Cordain L. Diet-dependent 2015;113:574–84. and short duration. Nevertheless, there acid load, Paleolithic [corrected] nutrition, and 22. Metzgar M, Rideout TC, Fontes-Villalba M, appears to be enough evidence to warrant evolutionary health promotion. Am J Clin Nutr Kuipers RS. The feasibility of a Paleolithic diet for low-income consumers. Nutr Res further consideration of the Palaeolithic 2010;91:295–97. 7. O’Keefe JH Jr, Cordain L. Cardiovascular disease 2011;31:444–51. diet as a potential dietary option in the resulting from a diet and lifestyle at odds with 23. Warensjo E, Byberg L, Melhus H, et al. management of metabolic diseases. our Paleolithic genome: How to become a Dietary calcium intake and risk of fracture and osteoporosis: Prospective longitudinal cohort Larger independent trials with consistent 21st-century hunter-gatherer. Mayo Clin Proc 2004;79:101–08. study. BMJ 2011;342:d1473. methodology and longer duration are 8. Eaton SB, Eaton SB 3rd, Sinclair AJ, Cordain 24. National Health and Medical Research Council. Australian dietary guidelines. Canberra: NHMRC, required to confirm the initial promise L, Mann NJ. Dietary intake of long-chain polyunsaturated fatty acids during the paleolithic. 2013. in these early studies. Claims that the World Rev Nutr Diet 1998;83:12–23. Palaeolithic diet could treat or prevent 9. Audette RV, Gilchrist T. Neanderthin: Eat like a conditions such as autism, dementia and caveman to achieve a lean, strong, healthy body. 1st St Martin’s Press edn. New York: St Martin’s, mental illness are not supported by clinical 1999. research. 10. Lindeberg S. Paleolithic diets as a model for prevention and treatment of Western disease. Conclusion Am J Hum Biol 2012;24:110–15. 11. Turner BL, Thompson AL. Beyond the Paleolithic The Palaeolithic diet is currently over- prescription: Incorporating diversity and flexibility hyped and under-researched. While the in the study of human diet evolution. Nutr Rev 2013;71:501–10. claims made by its celebrity proponents 12. Knight C. ‘Most people are simply not designed are not supported by current evidence, to eat pasta’: Evolutionary explanations for the Palaeolithic diet may be of benefit in obesity in the low-carbohydrate diet movement. Public Underst Sci 2011;20:706–19. the management of various metabolic 13. Lindeberg S, Jonsson T, Granfeldt Y, et al. A derangements. Further research is Palaeolithic diet improves glucose tolerance warranted to test these early findings. more than a Mediterranean-like diet in individuals with ischaemic heart disease. GPs should caution patients on the Diabetologia 2007;50:1795–807. Palaeolithic diet about adequate calcium 14. Jönsson T, Granfeldt Y, Ahren B, et al. Beneficial effects of a Paleolithic diet on cardiovascular risk intake, especially those at higher risk of factors in type 2 diabetes: A randomized cross- osteoporosis. over pilot study. Cardiovasc Diabetol 2009;8:35.