Author: Chip Bennett

For background information, see the Introduction.

Part 1: Is It Really Low-Carb?

The study purports to evaluate “the effects on mortality of habitual low carbohydrate–high-protein diets that are thought to contribute to weight control.” – ostensibly, this study evaluates low-carbohydrate/high-protein diets designed for weight control (that is, weight loss). In other words, this study purports to evaluate low-carb weight-loss diets. (I may sound repetitive here, but I have a purpose). What, then, does this study consider to be “low-carbohydrate/high-protein”?

According to the Results section:

With respect to the choice LC/HP score that relies on energy-adjusted components, at the high extreme of the distribution around 20% of energy intake was derived from proteins, whereas around 25% was derived from carbohydrates. At the low extreme of the distribution, around 10% of energy intake was derived from protein, whereas more than 50% was derived from carbohydrates.

Thus, for the purposes of this study, “low-carbohydrate/high-protein” means 25% carbohydrate and 20% protein.

Yes, you read that right: this study claims that a diet that includes 25% of its caloric intake from carbohydrate is “low-carb”, and that a diet that includes 20% of its caloric intake from protein is “high-protein”.

For a 2,000 kCal daily intake, these values equate to 500 kCal, or 125g, of carbohydrate per day – and bear in mind, this is the most extreme low-carb limit in the study. (Likewise, these values equate to 400 kCal, or 100g, of protein per day, for a 2,000 kCal daily intake. This is the extreme high-carb limit in the study.)

So, my first question is this: what mainstream low-carb diet in any way resembles this macro-nutrient intake?

To put the bounds of the study into perspective, here are the mean macro-nutrient intake values for the population:

In this population, the mean intake of protein was 76 g/day with standard error of the mean (s.e.m.) 0.16 g/day, the mean intake of carbohydrates was 208 g/day with s.e.m. 0.44 g/day, and the mean intake of lipids was 109 g/day (28% saturated, 15% polyunsaturated, 48% monounsaturated fatty acids and 9% other components of the lipid group) with s.e.m. 0.25 g/day.

Let’s explore those numbers for carbohydrate intake: mean intake 208 g/day, s.e.m. 0.44 g/day. Standard Error on the Mean (s.e.m.) is equal to the standard deviation (σ) divided by the square root of the number of samples (total population size). The total population size is 22,944; the square root of 22,944 is 151.5. Thus:

s.e.m. = 0.44 g/day = σ / 151.5

σ = 0.44 g/day * 151.5 = 66.7 g/day

Thus, carbohydrate intake for the study was a mean of 208 g/day with a standard deviation of 66.7 g/day.

Standard Deviation is the measure of the distribution around the mean. We can probably safely assume that our population is normally distributed (bell-shaped curve, with the peak at the mean, and exactly half of the population on either side of the peak). In a normally distributed set of data, 2/3 of all data lie within 1 σ of the mean, and 90% of all data lie within 2 σ of the mean.

Therefore, given the values for mean and standard deviation, we know that, statistically speaking, 2/3 of the study participants had a carbohydrate intake between 141 and 274 g/day, and that 90% of the study had a carbohydrate intake between 75 and 340 g/day.

How accurate is that estimate? The accuracy depends on the normality of the distribution.

According to Table 1 of the study, only 552 men (5.9%) and 2,218 women (16.3%) – a total of 2,770 participants (12.1%) had a carbohydrate intake of less than 140 g/day.

Based on the mean and standard deviation above, we can estimate that since 2/3 of the study participants had a carbohydrate intake between 141 and 274 g/day, that 1/3 were outside of that span – half of which (1/6, or 16.7%) had a carbohydrate intake of less than 140 g/day.

If anything, the distribution is biased against the low-carb side of the curve – meaning that the true number of participants at any given standard deviation is actually less than predicted by the normal curve.

The two most popular mainstream low-carb diets are the Atkins Diet and Protein Power (the South Beach diet does not consider itself to be “low-carb” and is excluded here).

The Atkins Diet allows anywhere from 20g carbohydrate per day in the Induction phase to 60-90g carbohydrate per day in the Maintenance phase (with allowance for more than 90g per day for active exercisers).

Protein Power allows anywhere from 40g carbohydrate per day at the Intervention level to 80g carbohydrate per day at the Maintenance level (according to Table 13.6 of my Protein Power Lifeplan book, page 355).

That means that less than 5% of the study participants had a carbohydrate intake anywhere near what is specified by mainstream low-carbohydrate diets.

In other words, this study in no way resembles either mainstream, low-carbohydrate diet. Therefore, any conclusions to which this study might lead do not apply to mainstream low-carbohydrate diets.

Coming Soon – Part 2: Is It Really High-Protein?

Introduction

Recently, Jimmy Moore at Livin’ La Vida Low-Carb linked to an epidemiological Greek Study that purported to compare mortality rates of Low-Carb/High-Protein and High-Carb/Low-Protein diets, along with a challenge from Dr. Steven Acocella – a vocal critic of low-carb diets – to refute it under the assumption that it is a legitimate study:

I will not editorialize on the study, but simply listen to you and your reader’s comments. I will say that there’s no reason that we need to dispute the efficacy of the study itself. Let’s go from the position that the study is not flawed. Let’s discuss the science and findings.

The conclusions of the authors are reproducible and consistent. What do you all think? If you do post this study I applaud your willingness to explore the science and not ignore nor dismiss it.

Many of Jimmy’s readers have been up to the challenge, and I would like to throw in my two-cents’ worth.

First, let’s examine the abstract, to get a basic understanding of the study. Here is the study objective:

We have evaluated the effects on mortality of habitual low carbohydrate–high-protein diets that are thought to contribute to weight control.

As the objective states, the object of the study was to evaluate the effect on mortality rate of so-called low-carbohydrate/high-protein diets – ostensibly, low-carbohydrate weight-loss diets.

The study setting is the general Greek population. Here are the subject methods:

Follow-up was performed from 1993 to 2003 in the context of the Greek component of the European Prospective Investigation into Cancer and nutrition. Participants were 22 944 healthy adults, whose diet was assessed through a validated questionnaire. Participants were distributed by increasing deciles according to protein intake or carbohydrate intake, as well as by an additive score generated by increasing decile intake of protein and decreasing decile intake of carbohydrates. Proportional hazards regression was used to assess the relation between high protein, high carbohydrate and the low carbohydrate–high protein score on the one hand and mortality on the other.

In other words, this is an epidemiological study based on the Greek-population subset (about 23,000 people) of the EPIC study from 1992 to 2003. Analysis is based on separating this subset into various decile groups based on descending protein intake, ascending carbohydrate intake, and a sum of the two deciles.

Understanding these groupings is key to understanding the study, so let’s take some time with the explanation. For the three analysis groups, participants were separated into decile groups – that is, groups of 1/10 of the total – based on the given criterion. So, for the descending protein intake group, all 23,000 participants are ordered based on protein intake, and the highest 2,300 participants are placed in the first decile (and so on, for all 23,000 participants). Likewise for descending carbohydrate intake (except, in this case, the 2,300 lowest carbohydrate-intake participants are placed in the first decile). Finally, for the third analysis group, each participant’s two decile “scores” (the first decile is scored a “1” and the tenth decile is scored a “10”) are added together, and groups are defined from a “score” of 2 (low-carb/high-protein) to 20 (high-carb/low-protein). Make sense?

Moving on: The study claims the following results:

During 113 230 persons years of follow-up, there were 455 deaths. In models with energy adjustment, higher intake of carbohydrates was associated with significant reduction of total mortality, whereas higher intake of protein was associated with nonsignificant increase of total mortality (per decile, mortality ratios 0.94 with 95% CI 0.89 –0.99, and 1.02 with 95% CI 0.98 –1.07 respectively). Even more predictive of higher mortality were high values of the additive low carbohydrate–high protein score (per 5 units, mortality ratio 1.22 with 95% CI 1.09 –to 1.36). Positive associations of this score were noted with respect to both cardiovascular and cancer mortality.

Finally, the study makes the following conclusion:

Prolonged consumption of diets low in carbohydrates and high in protein is associated with an increase in total mortality.

Sounds pretty bad for low-carb diets, eh? Perhaps – or perhaps not. We need to dig deeper into the study in order to decide.

Read on, in Part 1: Is It Really Low-Carb?