Why The Case For Public Health Recommending A Low-Carb Diet Is Flawed

In 2016, the Public Health Collaboration [PHC] in the U.K. published the National Obesity Forum Report entitled, ‘Eat Fat, Cut the Carbs, and Avoid Snacking to Reverse Obesity and Type 2 Diabetes.’

The report made several bold statements about the role of fat in the diet, in particular that ‘eating fat does not make you fat.’ Within the report, several hypotheses were advanced in favour of their position, including:

1: Low-carbohydrate, high fat [LCHF] diets are superior to low-fat diets for weight loss;

2: The reason LCHF diets are superior is due to their satiating effect and lack of impact on insulin levels (i.e. weight loss is dependent on lowered insulin levels);

3: That public health should recommend a low refined carbohydrate, high healthy fat diet (National Obesity Forum, 2016).

The report also repeated the “saturated fat is not associated with heart disease”, which I’ll address in the next post so this one does not become overly convoluted. I want to stay on point with the above hypetheses.

Interestingly, point no.10 in the report stated: “Evidence Based Nutrition Should Be Incorporated In to Education Curricula For All Healthcare Professionals.”

I wholeheartedly agree. So, let’s put the PHC report’s hypotheses to the evidential test, taking each point 1-3 above and examining each in turn.

1: LCHF Diets Are Superior to Low-Fat Diets for Weight Loss

In a meta-analysis of 9 trials of free-living [meaning diet isn’t controlled, normal life circumstances] comparing LCHF to diets high in carbs and low in fat [HCLF], unrestricted [meaning no conscious calorie limitations] low-carbohydrate diets resulted in an average of 2.5kg and 4kg greater weight loss over 12 and 24-weeks, respectively, than hypocaloric [calories matched] low-fat diets (Buchholz & Schoeller, 2004).

Studies like this have led to the suggestion the LCHF diets have a “metabolic advantage” i.e. greater weight loss compared to an isocaloric low-fat diet (Feinman & Fine, 2003). This “metabolic advantage” was said to result from increased availability of fatty acids in the absence of carbohydrates to use as fuel, suggested to amount to an additional 300-600kcal per day of energy expenditure on LCHF diets (Ibid.).

Other meta-analysis have found significantly greater reductions in weight on low-carbohydrate diets compared to low-fat diets over 1-2 years (Hu et al., 2012). However, meta-analyses can often be misleading depending on the quality of the studies included. In this particular meta-analysis, it is difficult to draw any conclusions due to wide variation in the carbohydrate content of the low-carbohydrate diets [4-45%] and low-fat diets [10-30%] of included studies (Ibid.).

A more recent meta-analysis confining inclusion to ad libitum [meaning eat as much as you like] LCHF trials with <20% of energy coming from carbohydrate found significantly greater weight loss over 6 to 24-months compared to energy-restricted low-fat diets with <30% energy from fats (Mansoor et al., 2015). However, this meta-analysis is indicative of the biggest variable between results from LCHF trial s and HCHF trials: dietary protein intake. When you look at the data in this meta-analysis, the LCHF trials included averaged 30-35% energy from protein compared with 15% in the low-fat diets (Ibid.).

This is critical to the outcome effects of each diet, and a significant confounder of the results. Why? Because it is well-established in the literature that the “metabolic advantage” to LCHF diets is attributable to higher protein intake, not lowered carbohydrate and/or increased dietary fat, due to the increased satiety and diet-induced thermogenesis of higher protein diets [you burn off 1/4 of the protein you eat in the process of digestion] (Schoeller & Buchholz, 2005).

In January of this year, a comprehensive recent meta-analysis was released, which appears to have concluded this argument. Hall & Guo (2017) confined their analysis to controlled feedings studies of diets with equal calories and with equal protein, but differing in content of carbohydrate to fat. The results illustrate that in controlled situations with matched protein intake, varying the ratio of dietary fat to carbohydrate does not make a significant difference in weight loss (Hall & Guo, 2017). See the tables from the paper:

 

hall-guo

hall-guo-2

Figure: Meta-analysis of controlled isocaloric feeding studies with constant dietary protein and varying ratios of carbohydrate to fat. Studies are ordered from top to bottom according to the largest difference in carbohydrate between diet comparisons. Effect size (ES), upper and lower 95% confidence limits (UCL and LCL, respectively) are indicated for the differences in daily energy expenditure (A) and rate of body fat change (B). The pooled weighted mean difference across studies demonstrated small differences in daily energy expenditure (26 kcal/d, P <.0001) and body fat change (16 g/d, P <.0001) favoring lower fat diets (Hall & Guo, 2017).

 

2: LCHF Diets Induce Satiety & Lower Insulin Levels

One of the noted features of ad libitum low-carbohydrate diets is that subjects spontaneously reduce energy intake (Mansoor et al., 2015). However, the PHC report attributed – incorrectly – this satiating effect to dietary fat, when the research is clear it is more appropriately attributed to the satiating effect of higher protein content than increased fat (Schoeller & Buchholz, 2005).

In free-living subjects, the evidence suggests increasing percentage of fat in the diet increases dietary energy density, promoting increased weight gain (Rolls, 2009; Stubbs et al., 1995). Studies on appetite and energy intake show dietary fat has less satiating effect compared to carbohydrate and protein (Schoeller & Buchholz, 2005; Subbs et al., 1995; Astrup et al., 2000). Thus, the hypothesis that eating more fat induces satiety is not supported by the literature.

That weight loss is dependent on lowered insulin levels is a cornerstone argument in the LCHF paradigm, based on the hypotheses that lowered insulin increases circulating fatty acids for oxidation (Manninen, 2004). Two metabolic ward studies dismiss this concept. In the first, subjects were fed isocaloric diets for 4-weeks on 50% carbohydrate, 35% fat, 15% protein before crossover to 5% carbohydrate, 80% fat, 15% protein very low-carbohydrate ketogenic diet [VLCKD] for a further 4-weeks (Hall et al., 2016). In the VLCKD phase insulin levels decreased by 47%, yet there was no difference in the quantity of circulating energy or energy expenditure (Ibid.).

The second study by the same group compared isocaloric diets with a 30% energy deficit; the comparison diets contained 29% carbohydrate vs. 8% fat, and led to equal total weight loss, dismissing the hypothesis that decreased insulin levels are metabolically advantageous for rate of fat loss (Hall et al., 2015).

However, while metabolic ward studies disprove the “carbohydrate-insulin hypothesis” of obesity, it has to be accepted that they lack applicability to free-living situations. So while the PHC report is incorrect on the science from the perspective of LCHF diets being superior for weight loss, that fat is a more satiating macronutrient, and that lowered insulin is required for weight loss, there is still compelling argument to make that what matters is free-living situations.

3: Public Health Recommendations Should Emphasize an Ad Libitum Low Carbohydrate, High Healthy Fat Diet

Advocates of LCHF diets have argued that long-term fat consumption within a range of 18-40% total calories has little effect on driving increased weight gain (Willet & Leibel, 2002). The latest National Diet and Nutrition Survey data shows that the population average intake of dietary fat is 36% (Public Health England, 2014). This is a crucial point to grasp: LCHF advocates suggest that a 40% energy intake from fats is the upper threshold for calories from fat, yet the population average is already at the high end of the range suggested by LCHF proponents.

In sum, the population are already eating a relatively high fat diet.

NDNS data also shows average population carbohydrate consumption at 50% of calories (Public Health England, 2014). This leaves about 15% of energy coming from protein. What is interesting about these population averages, is that they are not far from the targeted macronutrient intakes on a population level, which are: 10% protein, 55% carbohydrate, 35% fat.

So if the population target levels of macronutrient intake [10%P/55%C/35%F] are not enormously dissimilar from actual dietary intake in the population [13%P/50%C/37F], then why has obesity risen from 15% in 1993 to 26% in 2014 (HSCIC, 2016).

Because the driver isn’t one macronutrient. This data suggests that the primary driver is increased food energy supply and energy-density (Swinburn, Sacks & Ravussin, 2009; Rolls, 2009). There is substantial evidence in support of the fact that energy-density has a greater effect on total calorie intake than any macronutrient variations (Rolls, 2009).

The idea that there is on macronutrient that is inherently fattening simply lacks any support in nutrition science. In studies on overfeeding, weight gain is identical whether subjects are overfed by predominantly carbohydrate [78%] or fat [58%], where energy is matched (Lammert et al., 2000).

The primary goal for public health is thus a reduction in overall dietary energy density across the population. To date, this has not been achieved. The PHC report (2016) believes that the introduction of the NACNE 1983 dietary guidelines are at fault, and that recommending a low-fat diet to the population shifted consumption to carbohydrate-based foods, which precipitated obesity.

This is an absurd assertion, evident once you have read the guidelines, and taken an objective look at the science. While the NACNE guidelines did recommend that population fat intake be targeted at 35%, the guidelines also recommended reducing sugar, increasing fibre, vegetables and wholegrain carbohydrates (Foster & Lunn, 2007).

What has happened in the interim? Sugar has increased, refined carbohydrates predominate in the diet, and fibre and vegetable intake has declined (Foster & Lunn, 2007; Public Health England, 2014). The PHC argument is predicated upon an assumption that the population did in fact reduce their dietary fat intake in favour of higher carb foods. They didn’t: fat intake stayed at around 37%. People started eating more sugar and refined carbohydrates anyway. The goal of achieving even 35% fat intake in the population has never been achieved.

If population targets for vegetable, fruit, fibre, and complex, unrefined carbohydrates had been met, we would arguably not need to be engaging in this “fat vs. carbs” debate, as even with fat at its current threshold of 37%, those variables would still confer health benefits. The Mediterranean diet is a perfect example of this: higher in fat [35-45% energy], but fat composition emphasising plant-fats and fish, while the dietary carbohydrates are rich in legumes, vegetables, fruit, and consequently dietary fibre.

The bottom line is this: public health nutrition guidelines have never been implemented. On that basis alone, it is both premature and unsupported by the evidence to recommend LCHF diets at a population level.

Conclusion

Eating fat does not make you fat, and neither does eating carbohydrates, independent of total energy intake (Buchholz & Schoeller, 2004; Astrup et al., 2000; Rolls, 2014; Lammert et al., 2000).

Increased energy density is a greater driver of obesity than dietary macronutrient composition (Swinburn, Sacks & Ravussin, 2009). Thus, a macronutrient-based argument for public health recommendations is redundant; the primary aim for public health nutrition is a reduction in population energy intake simpliciter.

In this respect, it is imperative that we shift to food-based recommendations to help implement key aspects of dietary guidelines for vegetable, unrefined wholegrain and fibre consumption. This must be paired with strategic legislative initiatives targeting the food environment, industry regulations, and shifting population behaviours toward more “default” healthful options.

 


 

References:

Astrup, A., Grunwald, G., Melanson, E., Saris, W. and Hill, J. (2000). The role of low-fat diets in body weight control: a meta-analysis of ad libitum dietary intervention studies. International Journal of Obesity, 24(12), pp.1545-1552.

Buchholz, A. and Schoeller, D. (2004). Is a calorie a calorie?. Am J Clin Nutr, 79(suppl), pp.899S-906S.

Feinman, R. and Fine, E. (2003). Thermodynamics and Metabolic Advantage of Weight Loss Diets. Metabolic Syndrome and Related Disorders, 1(3), pp.209-219.

Foster, R. and Lunn, J. (2007). 40th Anniversary Briefing Paper: Food availability and our changing diet. Nutrition Bulletin, 32(3), pp.187-249.

Hall, K. and Guo, J. (2017). Obesity Energetics: Body Weight Regulation and the Effects of Diet Composition. Gastroenterology.

Hall, K., Bemis, T., Brychta, R., Chen, K., Courville, A., Crayner, E., Goodwin, S., Guo, J., Howard, L., Knuth, N., Miller, B., Prado, C., Siervo, M., Skarulis, M., Walter, M., Walter, P. and Yannai, L. (2015). Calorie for Calorie, Dietary Fat Restriction Results in More Body Fat Loss than Carbohydrate Restriction in People with Obesity. Cell Metabolism, 22(3), p.531.

Hall, K., Chen, K., Guo, J., Lam, Y., Leibel, R., Mayer, L., Reitman, M., Rosenbaum, M., Smith, S., Walsh, B. and Ravussin, E. (2016). Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men. American Journal of Clinical Nutrition, 104(2), pp.324-333.

HSCIC, (2016). Statistics on Obesity, Physical Activity and Diet. Health and Social Care Information Centre.

Hu, T., Mills, K., Yao, L., Demanelis, K., Eloustaz, M., Yancy, W., Kelly, T., He, J. and Bazzano, L. (2012). Effects of Low-Carbohydrate Diets Versus Low-Fat Diets on Metabolic Risk Factors: A Meta-Analysis of Randomized Controlled Clinical Trials. American Journal of Epidemiology, 176(suppl 7), pp.S44-S54.

Lammert, O., Grunnet, N., Faber, P., Bjørnsbo, K., Dich, J., Larsen, L., . . . Quistorff, B. (2000). Effects of isoenergetic overfeeding of either carbohydrate or fat in young men. British Journal of Nutrition, 84(2), 233-245. doi:10.1017/S0007114500001471

Manninen, A. (2004). Is a Calorie Really a Calorie? Metabolic Advantage of Low-Carbohydrate Diets. Journal of the International Society of Sports Nutrition, 1(2), p.21.

Mansoor, N., Vinknes, K., Veierød, M. and Retterstøl, K. (2015). Effects of low-carbohydrate diets v. low-fat diets on body weight and cardiovascular risk factors: a meta-analysis of randomised controlled trials. British Journal of Nutrition, 115(03), pp.466-479.

National Obesity Forum, (2016). Eat Fat, Cut the Carbs, and Avoid Snacking to Reverse Obesity and Type 2 Diabetes. National Obesity Forum in association with Public Health Collaboration.

Public Health England, (2014). National Diet and Nutrition Survey: Results from Years 1, 2, 3 and 4 (combined) of the Rolling Programme (2008/2009 – 2011/2012). London: Public Health England.

Rolls, B. (2009). The relationship between dietary energy density and energy intake. Physiology & Behavior, 97(5), pp.609-615.

Schoeller, D. and Buchholz, A. (2005). Energetics of Obesity and Weight Control: Does Diet Composition Matter?. Journal of the American Dietetic Association, 105(5), pp.24-28.

Stubbs, R., Ritz, P., Coward, W. and Prentice, A. (1995). Covert manipulation of the ratio of dietary fat to carbohydrate and energy density: effect on food intake and energy balance in free-living men eating ad libitum. Am J Clin Nutr, 62(2), pp.330-7.

Swinburn, B., Sacks, G. and Ravussin, E. (2009). Increased food energy supply is more than sufficient to explain the US epidemic of obesity. American Journal of Clinical Nutrition, 90(6), pp.1453-1456.

Willett, W. and Leibel, R. (2002). Dietary fat is not a major determinant of body fat. The American Journal of Medicine, 113(9), pp.47-59.