For some time now there has been debate in some nutritional circles about the relative merits of different ‘diets’. Certainly, there has been at least two factions here. One argues for a low-fat, high-carbohydrate diet which usually includes plenty of starchy carbohydrates such as bread, potato, rice, breakfast cereals and pasta. Others (including me) argue that all that carbohydrate is bad news for many people, and in general terms it helps to put more emphasis on fat and protein in the diet.
The Atkins and other similar diets have often been classed as ‘high-protein’ diets. Some believe them to be high-fat too. Actually, when individuals adopt such diet they don’t tend to eat much more in the way of protein or fat – they just eat less carbohydrate. So, while the percentage of protein and fat will generally go up, the absolute amount of these ‘macronutrients’ stays more-or-less the same.
One of the rationales for eating a diet relatively rich in protein is that it is the macronutrient which, calorie-for-calorie, sates the appetite most effectively. What appears to be true is that individuals who adopt this type of diet almost always spontaneously eat less (often, several hundred calories a day less). This may have something to do with protein, but it might also have something to do with fat (which many people find is important for sating the appetite properly). It might also have something to do with the better blood sugar stability these diets tend to afford, which will help guard against episodes of low blood sugar which can trigger ‘false hunger’ and food cravings (usually for carbohydrate).
There is also a theory that higher protein diets may assist weight loss through increased ‘thermogenesis’. After eating, the metabolism will generally enjoy a ‘boost’ – a bit like what happens when you put fuel on a fire. The thermogenic effect of protein is greater for protein then it is for carbohydrate or fat. The effect is not huge, but might help weight control (and other things) over time.
This week, the European Journal of Clinical Nutrition published a review of the impact of protein content on a range of health parameters. The study was a ‘meta-analysis’ – a pooling together of similar studies. In this case, the results of 74 studies were lumped in together. These studies varied quite a lot in design. For example, in some studies individuals were left to fend for themselves food-wise, while in others they were supplied with food. Meta-analyses seem like a good idea, but something the fact that they lump together quite-different studies can make it hard to draw firm conclusions. Also, as usual, there can be a gap between what the diet was designed to test and what it actually tested, because not all people are utterly compliant of course. Many will misreport (deliberately or unconsciously) what they ate too.
Anyway, leaving these issues aside, what did the review find.
Well, overall, the prescribed diets has these macronutrient compositions (percentage of calories contributed by each macronutrient):
protein – 18 per cent
carbohydrate – 55 per cent
fat – 26 per cent
protein – 27 per cent
carbohydrate – 38 per cent
fat – 32 per cent
The diets in which higher protein was prescribed led to significant improvements, compared to lower protein, in a range of parameters, including:
Systolic blood pressure (the higher of the two blood pressure readings)
Diastolic blood pressure (the lower of the two blood pressure readings)
HDL cholesterol levels
Fasting insulin levels
No parameter measured was improved by the lower-protein diet.
Overall adverse effects was the same between groups, and there was no difference detected in terms of bone and kidney health (high-protein diets are often claimed to be ‘bad for the bones’ or kidneys though, actually, there is no evidence for this for health people).
In general terms, one would have to chalk this up as a win for higher-protein, lower-carbohydrate diets. But actually, the authors of this review were quite dismissive of the results. They tell us that: “Higher-protein diets probably improve adiposity, blood pressure and triglyceride levels, but these effects are small and need to be weighed against the potential for harms.”
First of all, why “probably”? Their own meta-analysis shows that higher-protein diets do indeed induce these changes. And the authors left some of the other benefits out too. But what’s this “potential for harms” all about? The meta-analysis actually found these diets to be as safe as lower-protein ones, overall, so what could they be referring to?
When authors’ conclusions don’t match their very own results it sometimes pays to look for potential conflicts of interest (this can help explain authors’ bias). Well it turns out that one of the study authors is an employee of Barilla – an Italian company that makes pasta. Two of the other authors received “grant support” from Barilla to conduct the review.
Now, the relative popularity of ‘high-protein’ diets has not been particularly good news for companies like Barilla, essentially because these diets are lower in carbohydrate, and specifically encourage restraint in the consumption of starchy ‘staples’ including pasta.
My suspicion (this is just speculation) is that Barilla decided to go looking for evidence which was unsupportive of higher-protein and lower-carbohydrate diets. If so, it seems Barilla’s efforts here back-fired a bit it seems, so what to do? How about pouring cold water on the clearly positive findings and refer to the “hazards” of higher-protein diets for which there is no evidence?
1. Santesso N, et al. Effects of higher- versus lower-protein diets on health outcomes: a systematic review and meta-analysis. Eur J Clin Nutr Epub 18th April 2012