We're always being told that alcohol is loaded with calories and that we'll put on the pounds if we drink it. But it's all a myth based on bad science, says Tony Edwards
One fact that's been drummed into us about alcohol is that it's high in calories and therefore fattening. It's a message harped on by the medical profession who, having scared us with their health propaganda, like to add the coup de gr^ace by appealing to our vanity. To fight the flab, they say, ditch the drink.
The NHS Direct website asks rhetorically: Did you know a glass of wine can contain as many calories as a slice of cake, and a pint of lager has about the same calorie count as a small sausage roll?
The trouble with the statement is that the message is just plain wrong. There is no evidence whatever to support the notion that alcohol by itself adds on weight. Period.
Yes, alcohol does contain a load of calories-in fact, almost as much as fat-but the fact remains that alcohol isn't fattening.
In the late 1990s, researchers from Brigham and Women's Hospital in Boston, Massachusetts, embarked on a survey of around 20,000 middle-aged women whose drinking habits and weights were tracked for almost 13 years. At the beginning of the study, all of the women started out with low-to-medium weights (roughly, UK dress sizes 8 to 12). However, by the end of the survey period, around 9,000 women had put on significant amounts of weight, and some had even become clinically obese.
Conventional medical theory would dictate that, all other things being equal, the fatties would most likely be the ones who drank alcohol. But they weren't-in fact, quite the reverse. The fatties were the women who didn't drink, and the skinnies were the heaviest drinkers.
In this study, it was the women who didn't drink who turned into size 18s, not the women who boozed. Drink prevented weight gain.1
There are at least a dozen other studies of alcohol and weight that generally confirm the findings of the Boston study. Not all of them show such a strong connection between alcohol intake and lack of weight gain, but the general trend is in the weight-loss direction, particularly for women.
Cause and effect
But there's a major caveat with all this evidence: it doesn't conclusively show that alcohol doesn't add on weight. The data show only a correlation between alcohol and weight-or rather, a lack of weight. It doesn't actually prove a causal connection, so over the past decade, a number of studies have subjected the alcohol and weight connection to the kind of testing required to establish the efficacy of a new drug.
One of the simplest studies was done by sports scientists in the US, who asked this common-sense question: does drinking a couple of glasses of wine each day put on weight or not? A total of 14 men were studied for 12 weeks, during which time they either drank a third of a bottle of red wine every day for six weeks and then abstained for the next six weeks, or vice versa. The results were as clear-cut as they were boringly revealing: the addition of two glasses of red wine to the evening meal did not change the men's body weight either way.2
But that still wasn't enough to convince sceptical nutritionists. After all, the central tenet of the calorie theory is this: for weight to remain stable, calories in must equal calories out. So, the sceptics argued, if taking in extra calories from alcohol doesn't put on weight, it must mean that alcohol somehow makes people eat less, meaning that they simply substitute food calories for alcohol calories.
In 1999, Dutch physiologists at Maastricht University tested 52 people to see if the sceptics were right. Predictably, they weren't: alcohol actually made people eat more.3
Another theory that a team of Swiss physiologists at the University of Lausanne wanted to test was whether alcohol might cause the body to heat up and so dissipate most of its energy as heat (in medicospeak, it's called 'thermogenesis').4 Although they found that alcohol increased the men's body temperature, it was only by a relatively small amount, so the thermogenesis theory couldn't begin to explain the lack of weight gain.
OK, they said, let's look at fat metabolism: could alcohol somehow be affecting that and so prevent weight gain? Again the answer was no. In fact, if anything, the Lausanne results showed that alcohol should technically lead to weight gain, rather than loss, as they found it "reduces lipid oxidation"-in other words, it doesn't burn fat.
So why doesn't alcohol's huge amount of energy (or calories) not translate into weight gain?
The calorie theory
Let me remind you what the problem is: alcohol is high in calories, which is one reason why we're officially discouraged from drinking. Yet the clinical evidence is clear that alcohol doesn't put on anything like the amount of weight which ingesting so many calories should theoretically produce.
Let's take a look at a chart of what's actually in alcoholic drinks in terms of three major nutrients and calorie content.
White wine (dry)
White wine (sweet)
The first thing that strikes anyone who's been watching their weight by cutting down on carbs is that the carbohydrate content of drinks is very low despite their high calorie content. For example, red wine's 3 g of carbs (at 4 calories/g) is only 'worth' 12 calories, but red wine supposedly 'contains' 85 calories. Where do the extra 73 calories come from?
Even more strikingly, although spirits contain not a single molecule of fat, carbohydrate or protein, they have a massive calorie score.
What's going on? Where do all these calories come from? To answer those questions, we must take a dive into history . . . back to the late 19th century, when calories were first 'invented'.
Of the many scientific terms describing energy, few of us know more than a handful: volts, amps, watts and calories. The three electrical ones are all named after real people and by the same token, we really ought to call the calorie the 'clem', after the 19th-century French scientist Nicolas Cl'ement-Desormes.
Cl'ement wasn't a nutritionist, but an engineer. His main interest was steam engines and he wanted to find a more accurate measure of heat energy than horsepower. A patriotic Frenchman, he chose the system invented by his countrymen-the metric system-and defined the amount of energy required to raise the temperature of 1 kg of water by 1 degree centigrade as a 'calorie' (in Latin, calor = heat).
However, Cl'ement abandoned the concept in the 1880s, precisely the time that American agricultural chemist Wilbur O. Atwater decided to take it up.
Ostensibly a philanthropist, Atwater was worried by the poverty of the common people, most of whose income was spent on food. He wanted to determine which foods gave the most energy bang for the buck so he could recommend the cheapest foodstuffs to shovel down the throats of America's labouring classes to fuel their toiling muscles.
Atwater wanted to sound scientific, so he stole Cl'ement's 'steam-engine calorie' idea and turned it into a 'food-calorie' concept. To his way of thinking, the human body and its moving parts was indeed a kind of steam engine, with food as the coal and the stomach as the furnace.
To measure the energy of a food, Atwater decided to treat it just like coal-burn it to ash and measure how much heat it produced. Thus was born the so-called 'bomb calorimeter', a mini-furnace surrounded by a jacket of water, a device still in use today. To calibrate the amount of heat produced, Atwater nicked Cl'ement's formula: the amount of heat to raise 1 kg of water by 1 degree C = 1 calorie.
Atwater tested a wide variety of foods in his calorimeter and discovered that the three main food types followed a consistent pattern, with the most heat (in his terms, calories) being produced by fats, which had more than double the heat output of carbohydrates and protein.
Atwater was also a leading member of the Temperance movement, the forerunner of the anti-alcohol Prohibitionists in the 1930s, so he decided to see what dirt he could dish out on alcohol via his bomb calorimeter.
The outcome was predictable. Anybody familiar with the stories of stranded motorists filling their empty petrol tanks with whisky knows that alcohol is highly combustive. Indeed, alcohol made from distilling agricultural crop waste was a rival to petrol until the 1930s, and today's biofuels, of course, are also alcohol-based.
So when Atwater put alcohol into his bomb calorimeter, it burned like a firecracker, giving off huge amounts of heat-in other words and once again, in his self-invented terminology, loads of 'calories'.
Calories are only a theory
We have understandably come to think of calories as little packets of energy somehow embedded in food. But they are not real: they are simply a measure of how much heat is given off when foods are burned.
You may well ask: what relationship does burning food in a furnace have to what happens to food in the body? It turns out that not a single scientific study has ever been done to answer such an obvious question-and for obvious reasons: it's just not feasible.
Digesting and assimilating food energy is a complex biological process that is impossible to replicate using laboratory equipment, especially a crude 19th-century device like a bomb calorimeter.
Take another example: nuts. Chock-full of fats and oils, nuts are among the top 10 most calorific foods; in fact, they're often recommended to people who are dangerously underweight. However, in the last 20 years, study after study has consistently shown that nuts not only don't put on weight, they may even help reduce it.
Nutritionists have tried to explain these findings by suggesting that nuts promote 'satiety'-they make you feel full-so you eat less food in general. But at least two studies have shown this can't be the whole answer: when people are fed diets with identical numbers of calories (in the jargon, 'isocaloric'), they put on less weight when the diet contains nuts.5
So we now have two examples of high-calorie substances-nuts and alcohol-with no effects on weight. But there are other reasons to doubt the science behind the calorie theory.
For decades, the food/nutrition/diet industry triumvirate has rammed down our throats the message that fat is fattening and low-fat is slimming, hence the plethora of low-fat versions of virtually every processed food. But it now turns out that the low-fat message is just plain wrong, and here's just one example of a clinical trial that proves it.
In 2003, an American team of nutritionists tested two low-calorie slimming diets head-to-head in a group of obese women. One was exactly the kind of weight-loss diet recommended by the nutritional establishment and food industry: low in fats (30 per cent of total calories) and relatively high in carbohydrates. The other was the reverse: relatively high in fats and low in carbohydrates. Both diets had roughly the same number of calories. The low-fat dieters lost 3.9 kg (8.5 lb), but the high-fat dieters lost more than twice that amount: 8.5 kg (19 lb). Even more startling was that this group had more than half their calorie intake from fatty foods.6
So where did the low-fat idea come from? From the entrenched belief that fat contains loads of calories. And who told us that? Our old friend with the crude 19th-century food furnace, the self-taught American nutritionist Wilbur Atwater.
To recap, when Atwater burned foods in his bomb calorimeter, he observed that fatty foods burned more fiercely than carbohydrate and protein foods, giving off twice as much heat or calories. As anyone with a barbecue knows, fat burns very hot and catches fire easily. That's the reason Atwater derived the high figure of nine calories per gram of fat compared with four calories per gram of protein and carbohydrate.
So far so good, but so what? This still tells you nothing about what happens to food in the body. After all, burning 1 g of coal creates almost 10 times the amount of heat as burning 1 g of fat, but no one says coal 'contains' 70 calories and that you'll put on masses of weight if you eat it.
The problem is that Atwater, whose basic theories still underpin modern nutritional science, made a fundamental error in thinking that because food burns, it 'contains energy', and this energy is what powers the human body. Nutritional science post-Atwater went on to claim that if you ingest this energy (calories), but don't use it all up running the body's processes, the leftover calories will be deposited as fat. Hence modern nutrition's mantra: to maintain a steady weight, calories 'in' must balance calories 'out'.
You don't need a PhD in logic to see that this reasoning is chock-full of empty presuppositions, one of which is: how do you know how much 'energy' any particular substance delivers to the body?
Take the coal idea again. Coal burns far hotter than fat and so 'contains' way more calories. So why don't we put pulverized coal in cattle feed to fatten them up? The reason isn't simply because it's unpalatable, it's that coal can't be digested by the body-in the jargon, it's not 'bioavailable'.
The case of the disappearing rat
0% alcohol= 186g
5% alcohol = 185g
10% alcohol =175g
20% alcohol =170g
40% alcohol =157g
Scientists testing five groups of normally fed rats found that the more alcohol the rats drank, the less weight they put on. The top figure shown here refers to the percentage of alcohol added to each group's water, and the bottom figure the rats' final weight in grams. Rats on a 40 per cent booze/water mixture ended up 15 per cent lighter than those just drinking water, even though their calorie intakes were identical.
Back to booze
Now you may think this is all a massive digression from alcohol, but it's not. To reiterate the problem: alcohol contains lots of calories, but drinkers don't put on weight. Why?
If you've been following my argument, you should already have the answer and it's this: the calories in alcohol are probably not bioavailable, just like the energy in coal isn't.
But that's a line of reasoning only a few experts can contemplate even raising as an issue. "The paradox of increased alcohol-induced energy intake with no clear correlation between alcohol intake and body weight has led to the curious concept that alcohol energy has a low biological value," said Lausanne University's Dr Eric J'equier in 1999.7
But why do so few nutritionists raise the alcohol calorie issue? Because the proposition that calories from alcohol may be different from food calories is utter heresy. Nutritionists insist that 'a calorie is a calorie' and proclaim this as unchallengeable dogma. As a result, the heresy is not only buried, but also poorly researched.
Fortunately, some scientists have had the cojones to examine the alcohol bioavailability question. In 2004, nutritionists at Brazil's Fluminense Federal University reported on a series of highly controlled experiments feeding alcohol to laboratory rats.8 They observed that the more alcohol the rats drank, the less weight they put on, even though their total calorie intake remained the same.
Clearly, if the rats were so plastered they forgot to eat, you'd expect them to lose weight. But the researchers measured each rat's calorie intake and demonstrated that each animal's combined alcohol-plus-food caloric intake was the same as its food-only caloric intake.
These were identical groups of rats kept under identical conditions and ingesting the same total amount of calories from food and alcohol. The only difference was the amount of booze in their water bottles.
The graphic above shows the results when the rats were offered unlimited amounts of food, so you'd expect them to put on weight. Yet when alcohol was added to their water bottles, their weight gain was severely affected-and in a stepwise manner.
Drinking a 5 per cent alcohol/water mixture reduced their weight gain by 1g, but they basically stopped growing when the alcohol content was increased to
10 per cent.
Raising the alcohol/water mixture to 20 per cent caused the rats actually to lose about 5 g of weight. Increasing the alcohol content to 40 per cent made the rats lose even more weight, ending up 29 g thinner than the rats on plain water.8
It bears repeating that the rats' calorie intake remained constant no matter what the strength of the alcohol/water mixture they imbibed. This suggests that the alcohol calories were not assimilated by the rats' bodies-in other words, alcohol calories are not bioavailable. QED.
Now you could dismiss this study on the grounds that rats aren't people and may have different metabolisms from ours. So let's look at another graph (see opposite page: A liquid diet).9
This graph, buried away in a recondite paper by the 'father' of alcohol research himself, the late Professor Charles Lieber, shows what happened to the average weight of 12 people confined for over three weeks under "metabolic-ward conditions".
For the first week, the subjects were given a standard hospital-type diet. On the seventh day, food items of a known caloric value were removed from the diet and substituted by alcohol of the same caloric value (isocaloric). On each subsequent day, more food was removed and substituted by alcohol while maintaining the subjects' original identical calorie intakes.
The test ended after around three weeks, by which time the caloric value from food had been halved, so that the subjects were ingesting half their total calories from alcohol.9
In theory, the subjects' weight should have remained exactly the same throughout, but it didn't: the more alcohol substituted for food, the more weight they lost. As the graph shows, they ended up losing almost 1.5 per cent of their original weight, a staggering figure considering it took just a few weeks.
According to the calorie theory, this graph is impossible, but now you know why it's not.
Arch Intern Med, 2010; 170: 453-61
J Am Coll Nutr, 1997; 16: 134-9
Am J Clin Nutr, 1999; 69: 205-12
N Engl J Med, 1992; 326: 983-7
Am J Clin Nutr, 2003; 78: 647S-50S
J Clin Endocrinol Metab, 2003; 88: 1617-23
Am J Clin Nutr, 1999; 69: 173-4
Braz J Med Biol Res, 2004; 37: 841-6
Am J Clin Nutr, 1991; 54: 976-82
Tony Edwards is a science broadcaster and author.
This excerpt is from Tony's new book The Good News About Booze (London: Premium Publishing, 2013) which explodes virtually every myth about alcohol and health. Available from Amazon andwww.thegoodnewsaboutbooze.com
Special discount for readers
Tony Edwards is offering
WDDTYreaders a signed copy of his new book The Good News About Booze at a discounted price of lb10, including p&p, which represents a 20 per cent saving off the list price.
To claim your copy, send a cheque for lb10 made out to Tony Edwards to:
TGNAB, WDDTY Publishing
Ltd, Unit 10 Woodman
Works, 204 Durnsford
Road, London SW19
8DR. Don't forget to specify to whom you'd like the book dedicated. Please allow 28 days for delivery.