Follow me on

3,500 Cal Per Pound Is More Like 4,500 Cal


Note: see comment #1

One of the most persistent notions about equating caloric deficit to weight loss is the 3500 Cal “rule”.

I have previously posted about why this is nonsense and not exactly helpful when it comes to thinking about clinical weight loss or weight (you’re dealing with physiology NOT physics!).

Now, Nicholas Gwerder, a student from the University of Sacramento, in his Master Thesis, has apparently reviewed the literature on this and concludes that if anything, one pound of weight loss comes closer to 4,500 calories.

Gwerder reaches this conclusion by analysing data from 28 studies in which he compares the theoretical weight loss to the actual weight (and fat) lost in these studies.

Although, I do not have access to Gwerder’s Master Thesis, here is what he says in the summary:

“The energy equivalent of body weight loss varied considerably, dependent upon the constituent portions of fat, water, protein, carbohydrate and mineral lost. Adipose tissue also varied with type and was dependent upon the composition of lipid, water, and protein. The most valid theoretical equivalent for a pound of fat was calculated at 4,423.90 kilocalories based on in vivo extraction of human intracellular lipid samples.”

Thus, as Gwerder points out, the 3,500 per pound notion tossed around (including by a number of guidelines and associations)

“…severely underestimates the caloric values needed to achieve desired fat mass loss. This use of the proper caloric value for fat mass loss has the potential to improve exercise and nutrient recommendations for achieving healthy body fat values.”

Thus, if this number holds true, a daily 500 Cal deficit maintained over 10 weeks will not give you a 10 pound weight loss, but rather only about 7.5 lbs.

All the same, in practice over time this never really works out, not just because of the individual variability (Gwerder notes about 20% variation in this relationship) but because as you reduce your caloric intake, your individual metabolic requirements will very quickly shift to living off fewer calories, which means that pretty soon into your diet, the initial 500 Cal deficit is no longer a deficit (thank your physiology). This is the feared weight-loss plateau – the frustration of every dieter.

So, whether 3,500 or 4,500 Cal per pound, the relationship between calorie restriction and weight loss is not linear and thus extrapolating the amount of expected weight loss based on this deficit seldom works out in practice.

Indeed, I know from my patients that this “rule” is a matter of endless frustration and seldom helpful.

Managing weight is not simply about energy in and energy out.

Edmonton, AB


  1. The following important correction to the above reached me from Dr. Mike Lean:

    Dear Arya, You have clearly confused my very smart PhD student, George Thom (you will meet him one of these days!).

    Your man’s PhD was NOT estimating the calories per kg of weight loss (which is 3,500 kcal/lb (or 7,000 kcal/kg as we prefer), but estimating the calories per kg of lipid (which is 4,500 kcal/lb (or 9,000 kcal/kg) for us normal folk.

    The most valid theoretical equivalent for a pound of fat was calculated at 4,423.90 kilocalories based on in vivo extraction of human intracellular lipid samples.

    Those figures were first properly worked out by John Garrow in his book, which is doubtless on your shelves.

    Retract or face ridicule!

    Hope all is well.


    Post a Reply
    • Thanks Mike for pointing this out – and I am indeed grateful for the clarification (as I noted in my post, I did not have access to the actual thesis). That said, I still don’t think that the 3,500 (or whatever the actual figure may be) is particularly helpful in counseling my patients.

      Always good to know that I have such a bold and attentive readership.

      Cheers, Arya

      Post a Reply
  2. This comment reached me from George Thom:

    here is what I have from Garrow as an explanation for the 1 kg = 7000 kcals rule of thumb:

    “the excess weight in obese people comprises 75% fat and 25% fat free tissue (which is about 75% water and 25% protein). Since fat has an energy value of 9000kcal (37MJ)/kg, and fat free tissue about 1000 kcal (4 MJ)/kg, this mixture in adipose tissue has an energy value of 7000 kcal (29 MJ)/kg.”

    Within the book this extract is taken from Garrow does go on to discuss how an energy imbalance of say 40 kcal sustained over 10 years could (theoretically) lead to 20 kg weight gain, but only within the context of new equilibrium’s being reached incrementally (thus at a new equilibrium, the increase in calorie intake would need to be re-established for weight gain to continue), notwithstanding adaptive changes in energy expenditure opposing weight change and changes in diet induced thermogenesis. It is for these latter reasons that I believed that this rule of thumb was outdated thinking as there isn’t a linear relationship between calorie reduction and weight loss and this doesn’t recognise the dynamic changes that are occurring as an individual loses (or gains) weight.

    Widely debated topic and was featured in the NEJM Myths, Presumptions and Facts paper on obesity from a couple of years ago and also in a few others (one attached). I think the 3500 kcal rule originated from – Max Wishnofski (1958).

    Post a Reply

Submit a Comment

Your email address will not be published. Required fields are marked *