Despite my leanings towards natural medicine, I still read conventional medical journals each week, including the British Medical Journal. Over the last few weeks this journal has been running a series of articles about obesity that will be published next year as a book geared towards the education of medical professionals.
This week’s article focuses on ‘Strategies for preventing Obesity’. In it, just like previous articles in the series, the nutritional focus is on eating less fat. This, despite the fact that low-fat eating has been proven to be ineffective for the purposes of weight loss. I think the fat-phobia that grips our dietetic and medical establishments needs challenging. What really is the evidence regarding the relationship between dietary fat and obesity, and why do health professionals persist dishing out dietary advice which simply isn’t supported by the science?
Why should eating fat cause us to be fat?
The idea that eating fat causes us to be fat is well-embedded in most of our psyches and that of the medical establishment. But why? One reason for why fat has a fattening reputation is because it is, well, fat. The idea here is that the fat we eat is therefore needs little or no chemical transformation before it is dumped as fat under our skin and elsewhere in our body.
Fat is also incriminated in the burgeoning rates of obesity seen in so many countries because, weight for weight, it contains about twice a many calories as carbohydrate and protein. So, in theory, eating a lot of fat means generally means consuming a lot of calories – something which obviously increases our risk of piling on the pounds.
This obsession with calorie balance seems to have blinded most health professionals to the fact that different types of calories are dealt with differently in the body. Some types of calories burn more efficiently than others. We can think of the body’s metabolism as a bit like a fire which, say, easily burns small bits of dry wood, but has more difficulty with larger, soggy pieces of wood. Could it be that some individuals are very efficient metabolisers of fat, and are therefore relatively resistant to fat accumulation in the body? If we think of the body’s metabolism as a sort of internal fire, maybe for some people is like putting small, dry pieces of wood on that fire. It could also be that other types of fuel, even tough they contain the same amount of energy (calories) may not burn very efficiently, and stand a bigger chance of causing gains in weight.
So is a calorie really a calorie?
One way to test this theory is to test the weight loss effects of diets that contain the same number of calories, but are made up of varying amounts of fat, carbohydrate and protein. If it’s only calories that count, then diets that contain the same number of calories should have the same effect on weight, irrespective on the composition of the diets.
In one study, the effects of two diets of different composition were assessed in a group of women over a 6-month period. Each of these diets contained exactly the same number or calories, but one had 35 per cent of calories coming from fat, where the other had just 21 per cent . On analysis of the results, it was found that in postmenopausal women those eating the higher-fat diet actually lost significantly more weight than those on the lower-fat regime (an average of 7.7 kg versus 4.7 kg). And this is not the only study to find that eating more fat seems to bring more effective weight loss compared to diets with less fat and the same number of calories.
And what about appetite?
Apart from its failure to consider the type of calorie consumed, another reason why the calorie principle is fundamentally flawed concerns the effect foods have on appetite. Some foods, while perhaps being low in calories, may do little to sate the appetite, and may therefore lead to the consumption of more calories later on. On the other hand, a relatively calorific food may be highly effective in terms of quelling appetite, and may therefore lead to the consumption of less later on.
There is indeed some evidence that the eating of intensely calorific food may lead individuals simply eating less of other foods. In one study, scientists fed individuals 50 g (320 calories) of almonds each day for a period of six months to see what effective this had on weight . Despite all this almond-eating, weight did not increase significantly. Food diaries and interviews revealed that when individuals increase their consumption of nuts, they automatically compensated by eating less of other foods.
The appetite-quelling effects of a food seems to be related to the speed and extent it releases sugar into the bloodstream (known as its ‘glycaemic index’ or ‘GI’). Basically, the lower a food’s GI, the more satisfying it is. Foods relatively rich in fat such as meat and eggs are naturally low in carb, and therefore have very low GIs. This will obviously reflect well on their ability to sate our appetite.
Studies also show that, calorie for calorie, the most appetite suppressing component of the diet is protein. High fats foods like meat and eggs are very rich in protein, and may therefore help to reduce hunger because of this.
Put another way, eating high-carb, low-fat, low protein foods (like the starchy carbs we’re encouraged to base our diet on) may just cause us to eat more in the long term, and gain more weight as a result.
So, does eating less fat actually lead to weight loss?
The effects of different types of foods on calories on metabolism and appetite should clearly cause us to question the wisdom of eating a low-fat diet. But let’s not leave it there, let’s also look at the studies which assess the effect low-fat diets actually have in the real World.
Before we look at the results of such studies it is important for us to bear in mind that the best tests of a low fat diet will compare it with another diet which does not explicitly restrict fat (sometimes referred to as a ‘control’ diet). Generally, in these studies, individuals assigned to eat the low fat diet are given more instruction and support than those in the ‘control’ group. This will generally result in results skewed in favour of the low-fat diet.
The mass of evidence from several studies looking at low-fat eating was assessed in a review published by the highly-respected international group of researchers known as the Cochrane Collaboration in 2002 . The researchers were particularly interested in the ability of participants to sustain weight loss over a long period of time.
The average amount of weight lost in low-fat and control diets were assessed at 6, 12 and 18 months. Both low-fat and control diets were found to lead to weight loss which declined in time. Tellingly, though, the control diets out-performed the low-fat diets at every stage. And perhaps most importantly of all, at 18 months, those instructed to eat low-fat diets had, on average, actually gained weight, while those eating control diets had lost some. In the long term, it seems that all the sacrifice and deprivation usually entailed on a low-fat diet gets people precisely nowhere.
Low fat fallacy
So, why the appetite for low-fat eating? Well, I suppose some of this has something to do with the herd instinct that pervades the health professions. Sometimes it’s easier in life to trot out conventional ‘wisdom’ than have an original thought and risk rocking the boat and jeopardizing one’s ‘status’ and career prospects.
But of course the factor that may have an important influence here is money. Many researchers and scientists receive ‘funding’ from the pharmaceutical and/or food industries. Could this affect the sort of research and recommendations they churn out? Well, of course it could. So that researchers cannot be accused of conflict of interest, certain journals (including the BMJ) insist that authors of studies declare these interests.
One of the editors of the recent obesity series in the BMJ is Mike Lean – professor of nutrition at the University of Glasgow in Scotland, UK. He also co-authored this week’s edition of the series in which the ‘benefits’ of low-fat eating are trumpeted yet again (despite the evidence which shows this approach is simply ineffective).
To see what Professor Lean’s conflicts of interest are, we are directed to the first edition of the series. No date is given for this, so to find it I had to trawl back through back copies of the BMJ until I found it (for interested parties, it appears in 23rd September 2006 edition). Once I unearthed it, I discovered that full and frank disclosure was not to be found there at all. This is what the declaration states: ML has received personal and departmental funding from most major pharmaceutical companies involved in obesity research, and from several food companies. A full list can be seen on http://www.food.gov.uk/science/ouradvisors/ACR/
My question is, why not provide the list there and then? All the other authors did, but not Prof Lean. And what a shame this is for truth-seekers who don’t have on-line access. Why make us jump through a hoop to get the information anyway?
Things get even more curious when one find that the link provided does not actually take you to Prof Lean’s conflicts of interests at all. Click on it, and you find yourself at a page within the website of the Government’s Food Standard’s Agency (FSA). The FSA is advised on policy by a committee called the Advisory Committee on Research (ACR). This body is chaired by Professor Lean. The FSA website lists the ‘interests’ of each of the members of the ACR.
Now, to get from the link provided by the BMJ to the actual information on conflicts of interest takes a full three more clicks, and the route to it is far from obvious. More hoops! I suspect some of you may get lost on the way so to make it easier for you here’s the link to the actual document: http://www.food.gov.uk/multimedia/pdfs/acrdeclarationofinterestsaug06. There, that wasn’t so difficult, was it? I’ve got a good mind to send this link to Prof Lean to encourage him to use it in the future.
Those who manage to navigate what could be a very torturous route through to the document itself will learn that Professor Lean has quite a few ‘interests’, including funding from food companies such as Walkers and Unilever. These companies hawk, amongst other things, reduced fat or low-fat foods. In fact, earlier this year Walkers had a big marketing push focused on their reduced-fat crisps. And Unilever doesn’t just appear as an interest under Prof Lean’s name, but also under someone else on the ACR called Dr Peter McClure. Well, actually Dr McClure doesn’t just have an ‘interest’ in Unilever – it turns out he is a full time employee of this company. Now why would a committee whose job it is to advise on food policy from a ‘scientific’ perspective require the services of a full-time employee of a food company? Not just any company, mind, but the same one which provides funding to the chairman of the committee. Cosy.
Is it really so difficult to find researchers and advisors who do not have financial links with industry?
I suppose all of this would be much more palatable if the ideas promoted by such people stood up to scientific scrutiny. But, as we’ve seen in the case of the advice about ‘importance’ of cutting back on fat to combat obesity, this simply isn’t the case.
Now, if you haven’t already done this, take a look at the first reference below which refers to the study in which it was found that a higher fat diet was found to be more effective for weight loss than another diet containing the same number of calories and but less fat. Now focus on the beginning of this reference which tells us who the lead author of the study was. Recognise this name? Yup, that’s the very same Professor Lean who continues to trumpet the ‘benefits’ of low-fat eating. Surely shum mishtake?
1. Lean M E, et al. Weight loss with high and low carbohydrate 1200 kcal diets in free living women. Eur J Clin Nutr 1997;51:243-248
2. Fraser GE, et al. Effect on body weight of a free 76 Kilojoule (320 calorie) daily supplement of almonds for six months. J Am Coll Nutr. 2002;21(3):275-83
3. Pirozzo S, et al. Advice on low-fat diets for obesity. Cochrane Database Syst Rev. 2002;(2):CD003640