Academic ResearchCaloriesObesityPublic Health

The Calorie Theory – prove it or lose it

This is extracted from: “The Obesity Epidemic: What caused it? How can we stop it?”

During June and July of 2009 I approached the British Dietetic Association (BDA), Dietitians in Obesity Management (DOM), the National Health Service (NHS), the National Institute for Clinical Excellence (NICE), the Department of Health (DoH), the National Obesity Forum (NOF) and the Association for the Study of Obesity (ASO) to ask all of these expert organisations for proof of the 3,500 formula (also known as the calorie theory).

British Dietetic Association (BDA)

On 10 June, I sent the following query to the BDA: “I am doing some research on obesity and I would be most grateful if you could help me. Please can you explain where the ‘One pound of fat contains 3,500 calories…’ comes from?”

I received a prompt and pleasant reply, “Unfortunately we do not hold information on the topic that you have requested.” It was suggested that I contact a dietitian. I happened to be with several dietitians at an obesity conference later that month, so I asked fellow delegates and no one knew where the 3,500 formula came from. No one knew where the ‘eatwell’ plate proportions came from. One dietitian said to me “You’ve made us think how much we were just ‘told’ during our training, with no explanation. A group of us over there don’t even know where the five-a-day comes from.” (This may help)

NHS

So, after the conference, on 29 June, I sent the following email to the NHS: “I am an obesity researcher and I am trying to find out the rationale behind the statement: “One pound of fat contains 3,500 calories, so to lose 1lb a week you need a deficit of 500 calories a day”. This specific reference is a verbatim quotation from the British Dietetic Association’s Weight Loss Leaflet Want to Lose Weight and keep it off? The BDA reply was ‘we do not hold information on the topic.’ As this formula is the foundation of all current weight loss advice, it is critical to be able to prove it. Please can you let me know where this formula comes from and the evidence for it?”

On 30 June I received another prompt reply: “Unfortunately our Lifestyles team do not hold this information and are unable to assist you with your enquiry. I would suggest you contact the Department of Health to see if they can help.”

On 1 July I forwarded the email exchange with the NHS to the Department of Health. I had to chase on 6 July and then got a response saying they would get back to me within 20 working days. Meanwhile, I had also written to NICE (1 July) and they responded on 2 July saying they would get back to me within a maximum of 20 working days.

National Obesity Forum (NOF)

On 2 July I sent the same email to the NOF and the ASO and received a very prompt email back from the NOF (two hours later) suggesting that I contact the ASO. I thanked the NOF for this, but pointed out that their own web site quoted the 3,500 formula verbatim and also had the following classic example: “one less (sic) 50 calorie plain biscuit per day could help you lose 5lbs (2.3kg) in a year – and one extra biscuit means you could gain that in a year!” I have heard nothing back from the NOF since. I sent Dr. David Haslam, NOF chair, an email on 6 July attaching An Essay on Obesity, which I had written, for his interest and comment. On 10 July I also sent Dr. Haslam the exchange I had had with the ASO, so that information could be shared.

Association for the Study of Obesity (ASO)

The ASO response was the most helpful by far, but it still completely failed to prove the 3,500 formula. My query was circulated to board members and two kindly replied:

One reply was: ”Basic biology tells us that 1kg pure fat, converted to energy = 9000 kcal, 1lb pure fat = 0.453 kg = 4077 kcal. The approximation to 3500 kcal is made on the basis that ‘adipose tissue’ is not 100% fat (some water and some lean tissue). Hence to lose 1lb pure fat = 4077 kcal deficit, or 1lb fat tissue in the body = approx 3500kcal deficit. This equates to 500kcal per day to lose 1lb in a week. This has been supported by numerous studies using whole body calorimetry.” There were no sources put forward, for these “numerous studies”. I asked on 21 July and again on 11 August for “even one obesity study that proves this formula” and have received nothing back.

You can make 1lb = 2,843 calories or 3,752 calories without much effort. (This error could have a weight impact of a two stone loss or a six stone gain each year if any of this nonsense were true.) (Please see footnote for calculation).

The ASO member uses the word “approximation”, as do many references to the calorie formula, so there may be some acknowledgement of the number of variables. However, the diet advice that follows takes no account of this word “approximation”. If we take just one variation – the difference between 3,555 and 3,500 equates to five to six pounds a year (footnote). The NOF cautions that eating, or not eating, one biscuit a day could cause a person to gain, or lose, five pounds in a year. Well, the formula being inaccurate can also do this without any biscuit involvement at all. In fact, if 3,555 is correct and not 3,500 (notwithstanding the fact that there is no proof for either formula), this would have made a difference of 172 pounds over the past 30 years (the obesity epidemic period). Fortunately the error would be ‘in our favour’ so we should have all been able to eat nearly 11,000 biscuits and get away with it, or be over 12 stone lighter.

The second reply from the ASO was ‘evidence’ from NICE and a web link to the full NICE document Management of obesity: Full Guidance, December 2006. The specific proof offered was one study (Table 15.14) of 12 subjects, given a deficit of 600 calories a day, where the outcome was “a change of approximately -5 kg (95% CI -5.86kg to -4.75kg, range -0.40 kg to -7.80 kg) compared with usual care at 12 months. Median weight change across all studies was approximately -4.6 kg (range -0.60 kg to -7.20 kg) for a 600 kcal deficit diet or low-fat diet and +0.60 kg (range +2.40 kg to -1.30kg) for usual care”.

So, let me understand this, the people on the 600 calorie-a-day deficit (the NICE recommendation) were 5 kilograms (11 pounds) lighter than those not doing this “at 12 months.” Applying the basic maths formula, these 12 people should each have lost 600*365/3,500 = 62.57 pounds of fat. Not an ounce (of fat) more or less. AND, there should have been no range of results – everyone should have lost exactly the same (that’s what happens with a mathematical formula). The least anyone lost (let’s put it all into pounds) was 0.8 pounds and the most anyone lost was 17.2 pounds. Even the highest weight loss was 45 pounds lower than it should have been. This is also all about fat – we haven’t even started looking at muscle or water loss. This is also a study of 12 people. There are 1.1 billion overweight people in the world and we can’t prove a formula using 12 of them.

There were 15 other studies in Table 15.14, 10 of which had data for where a calorie deficit had been created over a specified period of time. This enabled me to analyse what the weight loss should have been (using the 3,500 formula) and what the average weight loss actually was (from the study data). Again, in every single study, there was a wide range of results (which means that the formula failed per se). In all of the other ten studies, the actual weight loss was multiples away from what the weight loss should have been. The smallest gap between actual weight loss and ‘should have happened’ weight loss (according to the formula) was 28.7 pounds (we continue to ignore water to try to give the formula a chance). At the other extreme, the biggest difference between the fat that should have been lost and the fat that was lost was 143.9 pounds.

Department of Health (DoH)

I was still digesting the immense implications of all this when the DoH reply arrived, on the 21 July, saying “The Department is unaware of the rationale behind the weight formula you refer to.” Pause for a second – the UK government Department of Health, has no idea where their founding piece of diet advice comes from. They kindly suggested another lead, (Dietitians in Obesity Management UK (DOM UK) – a specialist group of the British Dietetic Association), which I followed up on 24 July.

National Institute of Clinical Excellence (NICE)

I chased NICE on 27 July, as the 20 working days were ‘up’ in my calendar. I appeared to have been passed between NHS and NICE during July and a helpful woman called me back to say she had found the right department to deal with the query. A couple of days later, the reply came “Whilst our guidance does contain reference to studies involving 500 calorie deficit diets we do not hold any information about the rationale behind the statement ‘one pound of fat contains 3,500 calories, so to lose 1lb a week you need a deficit of 500 calories a day’.” That is to say – although we are an evidence based organisation, we have no evidence.

Dieticians in Obesity Management (DOM)

On 10 August I received a response from DOM UK: “I have asked our members and this answer was returned. It’s a mathematical equation, 1gram of fat is 9kcal, therefore 1kg fat equals 9000kcal. There are some losses but 1 lb of fat is approximately 4500kcal divide that by 7 days and its (sic) approximately 643kcal hence the deficit.” I went back to DOM UK, on the 10 August, to request an answer to the second part of the calorie theory – if that is how 600 calories is derived (and I have never before seen the 3,500 become 4,500), how can we then say with such confidence that each and every time this deficit is created one pound will be lost.

On the 18 August, I received a reply from one of the DOM UK Committee members: “My understanding is that it comes from the thermodynamics of nutrition, whereby one lb of fat is equivalent to 7000kcals, so to lose 1 lb of fat weight per week you would need an energy deficit of 7000kcals per week, or 500kcals a day. In or around that, depending on whether or not you use metric system and your clinical judgement, some people use a deficit of 600kcals a day and others 500kcals a day. There is good evidence that this level of deficit produces weight differences of approx 5kg at 1 year.”

This time the 3,500 deficit ‘needed’ has doubled to ‘7,000’ calories. Or, to put it another way, one pound of fat has become 7,000 calories. You can start to see what I have experienced as a researcher – how widely this formula is used as fact and yet how little it is understood and how few people know how to use their own ‘fact’.

So, in the example from Dietitians in Obesity Management, key proponents of the calorie formula, one year of a 600 calorie a day deficit will produce a weight loss of approximately 11 pounds – not the 62.5 pounds of fat alone that should be yielded.

When I pointed this out and suggested “I really think we need to fundamentally review the basis of current diet advice and stop saying ‘to lose 1lb of fat you need to create a deficit of 3500 calories’”, the final reply I got was “I guess a key to all of this is that weight loss doesn’t appear to be linear, any more than weight gain is.”

At last, an admission that the formula has no basis of fact.

The organisations approached have been helpful and accessible, but none is able to explain where the 3,500 comes from, let alone to provide evidence of its validity.

A request

I have a simple and reasonable request. I would like proof of this formula – that it holds exactly every single time – or I would like it to be banished from all dietary advice worldwide.

Any proof needs to source the origin of the formula. Then the proof needs to hold in all cases. There needs to be overwhelming, irrefutable and consistent evidence that each and every time a deficit of 3,500 calories is created, one pound of fat is lost.

Since, we already have overwhelming evidence that such proof cannot be provided, it is not enough that we quietly stop using this formula – it is too widely assumed to be true for us to just sweep it under the carpet. We need to issue a public statement saying that it does not hold and should not be used again. We need to tell people that they will not lose one pound of fat for every deficit of 3,500 calories that they create. We need to tell people that there is no formula when it comes to weight loss and we have been wrong in giving people the hope that starvation will lead to the loss of 104 pounds each and every year, in fat alone.

Footnote

This calculation is done as follows: It assumes that a person can maintain weight at a daily intake of the calories assumed to equal one pound of fat. If we think one pound equals 3,500 calories and in fact one pound equals 2,843 calories, over a year, 657 ‘extra’ calories a day, simply from the formula ‘being wrong’, would add up to 239,805 extra calories and this, divided by 2,843 gives 84 pounds, or six stone. Adjust the calculations for women more typically maintaining at 2,000 calories a day and men more typically at 2,600 calories a day and the inaccuracy of the formula still creates wide disparity.

161 thoughts on “The Calorie Theory – prove it or lose it

  • I loved the article and as a person who understands real science, not science as we wish it were, the idea of multiple variables makes perfect sense and the fact that if you want to have true weight loss you really cannot rely just on C-in/C-out as it is not an easy thing to control since too many variables go into what is a calorie in terms of weight gain/loss, vs the pure scientific definition. I bang my head when I read those who go super negative and try to prove you wrong the in/out hypothesis correct. It is accurate, not correct, which in proper argumentation is well understood to be very different. Since it is super hard to figure out what, exactly, might be a calorie in to my body as it may get flushed or just simply burned right away and not be part of what I need to “burn” off to lose, I find controlling glucose and insulin much easier and more production as that is one of the single most important variables, while still not being the only variable. Having struggled most of my adult life with weight it seems to work best to limit carbs and cut sugar completely. We do not need sugar anyway (table sugar that is), as with most things there are many types of sugar and our body does need some of them. Also cutting fat actually was an issue for me as we need some fat to be able to uptake certain other nutrients and vitamins as they are fat soluble. I really enjoyed you on the Peak Human podcast and hope to read some more of your work. It is very insightful and refreshing to get the other side of the coin with lots of science backed ideas.

    • Hi Aoife
      Thanks so much for your lovely comment on this and the Peak Human podcast.
      Good luck with your ongoing journey – it sounds like it’s going in the right direction :-)
      Best wishes – Zoe

  • I love when people challenge basic “facts” — I do all the time, but for some reason haven’t done so with this topic!!

    Having said that.. I think there are two arguments here.

    1) The calorie per lb of body fat is *not* an exact number, and perhaps those other factors should be given when the equation is described. Basically, upgrade people’s information from Idiot Light, to Temperature Gage (referencing the dumbing down of our cars indicators to “Needs attention” on off lights).

    2) What is required to actually have weight loss. As some people have mentioned (About 4 years ago!), metabolic rate has a lot to do with it. At its base level, the generic statement that 1lb of fat is (approximately ;) ) 3500 calories is true. But the accuracy of the deficit calculation is in question. A BMR of 1800 for a man means he is still gaining weight if he eats 2000 cals thinking he’s got a 500 cal deficit. In addition, there are other factors on the calories in and calories out parts of the equation. For calories in, some carbs take more energy to convert than others, and both require less than fat. Someone with insulin problems may lower their BMR with carbs due to insulin release, while raising it with both fat and protein. There are foods such as beans that lower carb intake. There are some gut flora that can partially digest fiber making it not the zero calories that are assumed. On the calories out part of the equation, Ketosis can create a situation where only 1/2 the calories are burned, while the other half are pissed out. Fat is also used to create cholesterol and cell walls and more, and things like healing can use up some. Fat Cals are not even at a pure 9 cals per gram — short chains can average closer to 6.

    My point is that while I agree that 3500 burned does not necessarily mean 1 lb body weight loss, I’m pretty sure we can be a little nuanced and understand that for every pound lost, 3500 cals is also lost. Burning the fat cals does reduce the fat.. The questions are; did you burn actual fat (as opposed to protein or carbs); did you maintain a caloric need (2500 for men, as an example) making your “deficit” real?

    Thanks again for attempting to track this down, and for holding both Government and other “Experts” feet to the fire with respect to both science and public policy.

  • I can’t tell you how much your information here today helped me!! I’m writing my thesis for my graduate program and this post was exactly what I needed to drive my point home about the inexact science. Thank you so much for sharing and for doing what you do!

  • Body fat is 87% lipid, and lipid is 9 kcal/g giving 1Kg containing .87*9*1000 = 7830 kcal

    1Kg is 2.2 lb so there SHOULD be 7830/2.2 = 3559 kcal/lb of body fat … which is where the APPROXIMATION of 3500 comes from, i.e a close multiple of 7 for easy calculations.

    3500 is a deficit of 500 per week
    3549 gives 507 per week (1 kcal per day!!)
    3559 gives 508.4 per week.

    The values are sound if the assumptions are sound, and the calculation is so trivially easy that I find it hard to see how so few know it … a quick google will reveal it in fact.

    The maximum error over a year between 3500 figure and 3559 is only 416 kcal … less than a single day’s dieting.

    You mentioned the single study had wide variation … but also point out that it is a single study, with insufficient sample. We also do not know if the calorie intake was enforced or left to the subjects (meaning variation most likely due to not sticking to the eating plan).

    We also do not know what exercise the subjects were doing, and weight losses are given as gross … so neglecting body composition changes (like perhaps increased muscle mass).

    I may have missed any reference to a control group as I skimmed through … but that would have been useful.

  • Pingback: Water and Fat – A Waist of Time

  • a calorie is a calorie, from a thermodynamics viewpoint. The issue is how calories are metabolized.

    The 3500Kcal/Lbs is an approximation. Will it hold true for everyone, no, but the same is true for any diet. Genetic factors will play a huge role, both in our starting point and how we respond to different diet and exercise regimes. One size does not fit all, one solution will not fit all.

    Truth is the energy density of Fat Mass is well established at 9.45KCal/g, in many weight loss studies the proportions of FM to FFM loss is around 80/20, with the FFM having a energy density of around 1KCal/g (lots of water here) so 1Kg loss would be 800g of FM – 7560KCal and 200g FFM – 200Kcal so 7760KCal/Kg loss. There will be variability here, especially around the 80/20 ratio and also the FFM energy density based on dietary sources of energy.

    Of course creating that 7760Kcal deficit becomes more of the issue and that is where the variations between individuals really kick in. a 100kg 1,95m tall male will respond very differently to a 100kg 1,6m tall female, before all the genetic components are taken into account. Add in the issues around triggers for satiety and consumption monitoring by the body and this advice soon goes adrift. Here the “a calorie is not a calorie” argument is stronger due to the issues around the nutritional availability of energy, especially when using label marking as a guide to calorie consumption.

      • The First law of thermodynamics does exactly that, it’s how a calorie is defined, as a unit of energy that can be converted to heat. It’s how a bomb calorimeter works.

        Talk about available energy and non-closed systems and you might have something when it comes to the value of a calorie in nutritional terms, but the Gram calorie has a clear scientific definition, and will always be the same. Saying otherwise undermines any argument you may make from a scientific viewpoint.

  • Gee, Dr. Zoe, for a scientist, you sure are imprecise in your terminology. When you talk about pounds of fat, are you referring to fat intake or (human) body fat?
    3500 (kilo)calories per pound of body fat is a reasonable estimate. I don’t believe any authority would characterize this figure as anything other than an estimate.
    Furthermore, anyone counting calories is probably “doing it wrong”. If one eats real food, not the junked up, carb-loaded crap that the food industry has trained us to crave, the vast majority of us will maintain a healthy weight (and improved health and lower healthcare costs!) just by listening to our stomach.

  • It doesn’t surprise me at all that all these organisations are so clueless about the origin of the “facts” they are stating, but I love the way you’ve shown them up!

    I was looking for info. on a slightly different aspect of this. Whatever we assume the energy value of a pound of body fat to be, there must be an energy cost in breaking it down and making it available to the body. So if, for argument’s sake, the energy value is in fact 3,500, what would be the net amount available to fuel exercise once it had been broken down? I can’t find info. on this anywhere – just loads of articles reeling out the 500 a day = 1lb of fat a week thing!

  • Hello,
    I have a question about the calculation and a comment for consideration. Sorry, I haven’t read the posts and replies, so may be it’s already been discussed.

    Question: in the article that linked me to this blog, you say 3500 calories is wrong, and point out that it is an estimate, but I wasn’t sure of the conclusion. Is it just to be aware that the actual amount could be higher or lower? I wasn’t sure if the point was to suggest people should not be intimidated by that number of calories, or if the idea is that people should not think it’s that easy. Or, if it was simply to point out that number is not exact.

    Comment: I have read that if you drastically reduce your calorie intake, say from 2500 to 1200 calories, it only takes your body about 2 week to adjust your metabolism to the new caloric intake, and you stop losing weight, even if you maintain the new lower caloric intake. In addition to that, (a lever in the opposite direction if you will), when you do some level of cardio, your body’s metabolism continues to burn calories at rest as a result of that cardio for roughly 2 hours.

    Consideration: I don’t have any background in physiology or biology or medicine or health, so I don’t really know what is being referred to by “metabolism”. But, if the whole conversation around 3500 calories is just to have a ‘guess-timate’ to plan around, that seems ok. But, it would be interesting to see a quantitative definition (or process by which to define) how these impact the calories burned.

    Example: If you get your heartrate over 80% for 10 minutes, is the metabolic after effect 1 hour? 2 hours? What about for 20 minutes? 30? I have found I can be very efficient at losing weight if I jump rope for 20 minutes, (1 minute of jumping alternated with 1 minute of twists or bends etc), 3x per day. My assumption is that the metabolic after effect means I get 3x the impact than if I did one workout for 1 hour. Secondary to that, based on the comment above, if you typically consume 2500 calories per day, it would be interesting to see how, in a reduced calorie diet, how much the reduction impacts metabolism. could you drop 500 calories and not impact metabolism? Does the metabolic adjustment occur immediately, or on a 1 week delay and then it speedily adjusts? Resulting example: could a person drop say 800 calories out of their diet every other week, enjoy the weight loss benefit, then eat their typical diet the following week as a step to prevent metabolic adjustment? How much of a calorie drop could be performed this way?

    Final: So, getting back to the 3500 estimate. If the after-effect metabolic impact was known, quantitatively, as well as the auto-adjust of metabolism from reduced calories, then perhaps those two could factor into how many calories burned vs that 3500. And for a final example for illustration: say your metabolism is such that your body burns 3000 calories per day. Say, you know that if you do a 10 minute cardio workout, you will get 1 hour of 20% increased metabolism. Say you know that you can drop 500 calories maximum per day without impacting metabolism, but you plan to drop 800 per day, because you want to just drop some weight this week for some meeting at the end of the week, and you will eat normally next week. say it’s a 5% drop in metabolism per day once you drop below that 500 calorie threshhold. You end up with a rough formula with the following considerations: calories in vs calories out, auto adjust metabolic impact, calories burned during exercise, number of exercise events, after-effect to metabolism, water loss. As for the 3500 calories. if the whole point is to know how many calories your treadmill should display, for a given goal, then in order to plan your weight loss, then you would not subtract the water loss you discuss in your blog, because that’s still weight lost. But if you burned 3500 calories, the actual loss (which could then be expressed as ‘effective’ calories burned, or even fat lbs burned), due to the discussed metabolic effects, could be greater.

  • Thank you for taking the time to do the research on this however those of us that have been fighting the obesity battle since we were small children have come to the conclusion that the experts have no idea what they are talking about and that we are all individuals and our bodies burn fat and metabolize calories differently. Just look around you at skinny people that eat 4000 calories a day and stay skinny without exercising and overweight people that go crazy watching their calorie intake and hit the gym everyday struggle to lose just five pounds.
    Our DNA has a lot of control over how we process calories. Do your best to eat real food, watch portion size, exercise every day and reduce your stress levels.
    Done!
    Don’t let anyone tear you down or tell you that you are a bad person because you don’t look like a supermodel.

  • Pingback: The Calorie Theory: Drop It Like It’s Hot | Valerie's Voice: For the Health of It

  • Energy itself is NOT ANYTHING. Energy is a PROPERTY of thingd. NO caloric energy is ever turned i to human fat tissue. Fat tissue is MATTER.

    • Hi Michael – great comment! So sorry I missed it – I missed loads for some reason!
      Best wishes – Zoe

  • Pingback: ¿En Realidad Que Causó la Epidemia de Obesidad y Cómo Puede Revertirse? | Parbiomagnético taptc

  • There are 454 grams to one pound
    There are 9 calories per gram of fat
    454 * 9 = 4,086 calories to a pound of fat
    More than the 3,500 rule of thumb used to lose weight
    You most likely want to lose fat and not muscle
    I fail to see why anyone does not know these things – did you ever take a science class?
    Have a great day

  • Pingback: What Really Caused the Obesity Epidemic, and How Can It Be Reversed? : Conscious Life News

  • You seem to encourage eating less carbohydrates and more good animal protein. I am considering becoming a vegetarian after reading about animal protein raising the cancer occurence risks as explained in the China study book for example. I am confused on how to manage carbohydrates without increasing the animal protein intake.

    • Hi Lara
      My dietary advice is 1) eat real food 2) three times a day (or fewer) – don’t graze and 3) if that doesn’t get you to natural weight/optimal health – manage carbs more carefully.

      Real food is meat, eggs and dairy (ideally from grass living animals); fish; nuts and seeds; fruits in season. I don’t advise anyone to eat wheat (as Wheat Belly and Grain Brain books nicely explain) Oats/quinoa/brown rice are OK for many people but not all. Legumes (beans/pulses etc) are generally OK. Anyone T2D/insulin resistant will likely need to avoid any grains and possibly legumes/fruit too. Hence – advising real food naturally features animal foods. Don’t just think of these as protein – protein is in everything other than oils/lard or sucrose. meat/fish/eggs are fat and protein. Nuts and seeds are fats/carbs and protein – it’s a common mistake to not know the macro nutrient content of foods.

      As for the China Study – read this first (http://rawfoodsos.com/the-china-study/) I can understand vegetarianism for animal reasons (although there are still misconceptions with objectives – https://www.zoeharcombe.com/2011/08/the-vegetarian-myth-lierre-keith/) Never go veggie because you think it’s healthier. It simply isn’t. https://www.zoeharcombe.com/2014/04/healthy-whole-grains-really/

      Hope this helps
      Best wishes – Zoe

  • Pingback: What Really Caused the Obesity Epidemic, and How Can It Be Reversed? | Dharma Healing Center

  • Pingback: The 1200 Calories a Day Myth

  • I see there are a lot of comments…I haven’t read them all so I don’t know if someone has already said this, my apologies if someone has. The premise that 3500ish calories equates to a lb of fat cannot be used in isolation. First, as noted, one needs to have some idea of how many calories their body burns typically without added exercises. Second, they need to know what foods to eat each day or over a week to meet all of their nutritional needs. Then they need to meet their nutritional needs while being slightly under their total calorie burn (I.e. a small calorie deficit)…but not everyday…but rather on average across the course of a week. Third, they must get the proper amount of sleep to allow their body to rebuild itself. Fourth, they must incur enough exercise to prevent their body muscles, bone and brain etc…from atrophy. These are the main energy using parts of the body and affect the normal calorie burn. After taking all of this into account…over the course of a year…they can lose fat…. To take a stab at estimating the fat loss potential under the these stated conditions…3500 calories is a ballpark estimate per pound…but doesn’t stand on its own…that is…restricting calories alone will not achieve this…not even close. None of the conditions above are impossible to know and/or achieve…but most people need some guidance…most people do tend to have some misconceptions about what proper nutrition is and how to achieve it…especially if they have medical issues on top of this. Also, most people don’t really know how many calories they burn in a day or week. There are estimating methods though…some more accurate than others. I appreciate your blog though…as your point is valid…you cannot just use the 3500 calorie deficit as a formula for fat loss in and of itself.
    Cheers,
    Michael

    • Hi Tony
      Many thanks for this – my hubby caught it by chance a few months ago. It was interesting to see the shift (not shared or known about by the vast majority of dieticians however). And it still missed the point in that dieticians still think there is a formula and still want a formula. Every dieter wants a formula – but here’s the truth – there isn’t one! Tell people the truth – not what they want to hear :-)
      Best wishes – Zoe

  • Pingback: Square One Wellness - Choices, Choices, Choices!

  • There are generally 9 calories in a gram of fat (lipid). This is determined by burning it in a bomb calorimeter and measuring the output of heat. There are 453.592 grams in 1 pound. Human fat is around 87% lipid.

    453.592 * 9 * 0.87 = 3551.62536

    3500 is a nice, round number so people use that, in the same way that people use the rule of 72 and not the rule of 69.3. Furthermore, playing loose with the numbers is fine because there’s going to be a little variation in each quantity.

      • Hi Zoë,

        The fully cited & peer reviewed paper outlining that claim is by Wishnofsky, M; Titled: Caloric Equivalents of Gained or Lost Weight and published in The American Journal of Clinical Nutrition in 1958.

        It can be viewed here: http://ajcn.nutrition.org/content/6/5/542.full.pdf

        It draws heavily on the 1929/1930 experimental results of Strang, J. M., McClugage, H. B., and Evans who found those numbers in a clinical setting, but also discusses some results of other researchers using different methods.

        Best,
        Brandon

        • Hi again
          I quote the Wishnofsky paper on the link I sent you and here (https://www.zoeharcombe.com/2015/04/where-does-the-3500-calorie-theory-come-from/) where an earlier 1918 reference is offered. Unless an earlier ref can be found, Lulu Hunt Peters wins the source but still fails to prove the calorie theory.

          I’m thus still interested in from whence the calorie myth came “to lose 1lb you need to create a deficit of 3,500 calories” (i.e. first used, where, by whom) and any proof for it.
          Best wishes – Zoe

          • Ohhh I’m sorry, I misunderstood. You’re not going to find the earliest reference unless you speak German, as the earliest works in this field are by two German Scientists – Max Rubner & Carl von Noorden.

            Here’s a short mention on wikipedia: http://en.wikipedia.org/wiki/A_calorie_is_a_calorie#History

            But for real examples of how much weight you lose per calorie, you have to go to the source. Here’s an example of CV Noorden doing the typical calculation: https://books.google.com/books?id=69ULSGmLQm0C&pg=PA92
            It roughly translates as:

            A diabetic woman at 32years of age and 55kg of weight needs 1925 calories per day. Over five days, she eats a deficit of 210 calories. He measured how much Nitrogen was in her urine, allowing him to determine how much energy she burned from protein. From Rubner’s work, he has 9.3 calories per gram of fat which meant 19.4 grams of fat were burned.

            If you convert those SI units to calories per pound, he’s arrived at 3670 calories per pound.

          • Hi Brandon
            Thanks again! You’re very kind to keep helping. The German lead is much appreciated – and the translation.

            Correct me if I’m wrong, but it doesn’t sound like the calorie theory is quoted in this (“to lose 1lb you need to create a deficit of 3,500 calories”)? Surely this whole calorie formula belief system didn’t come from an n=1 experiment where 210 calories were almost accounted for?!

            Best wishes – Zoe

          • PS That book is (C) 1901

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.