How to calculate a client's calorie needs from their intake

Four steps calculate a client's calorie needs from an intake form

Age, sex, height and weight go into the Mifflin-St Jeor equation and a resting metabolic rate comes out. You raise that for everything the client does between waking and sleeping, take a slice off it for the rate of fat loss you want, and you have a target. Then you find out how wrong it was, because the first three steps are arithmetic and only the fourth one is measurement.

We would not defend the first number past the second weigh-in. Set it from the equation, then let the scale edit it.

Step one: Mifflin-St Jeor turns four answers into a resting rate

Mifflin and others built the equation from 498 healthy adults, 247 women and 251 men aged 19 to 78, 264 of them normal weight and 234 obese, with resting energy expenditure measured by indirect calorimetry. The two forms they published are the ones still in use:

• REE (males) = 10 x weight (kg) + 6.25 x height (cm) - 5 x age (y) + 5
• REE (females) = 10 x weight (kg) + 6.25 x height (cm) - 5 x age (y) - 161

REE is resting energy expenditure, which most software and most coaches now call resting metabolic rate. Take a client who is 34, female, 168 cm and 78 kg. That is 780 plus 1,050 minus 170 minus 161, or 1,499 kcal a day. A man of 41 at 182 cm and 96 kg comes out at 1,898. Our free macro calculator opens on the first of these two clients and shows every step of the sum with her numbers in it.

Those figures are what the client burns doing nothing at all. Nobody eats at that number, and nobody should be given it as a target.

The equation lands within 10 % for most people and misses for some

The American Dietetic Association ran a systematic review of the four equations most used in clinical practice: Harris-Benedict, Mifflin-St Jeor, Owen and the WHO/FAO/UNU set. Frankenfield and others published the result in the Journal of the American Dietetic Association in 2005, and it is the reason Mifflin-St Jeor is the default in almost every piece of nutrition software.

Of these equations, the Mifflin-St Jeor equation was the most reliable, predicting RMR within 10% of measured in more nonobese and obese individuals than any other equation, and it also had the narrowest error range.

Best of four is not the same as right. On the 34 year old above, a 10 % band around 1,499 runs from about 1,350 to 1,650, and the review is clear that errors outside that band exist. It also records who was missing from the research: older adults and ethnic minorities living in the United States were underrepresented both when the equations were built and when they were tested.

The review's expert panel asks for clinical judgement about when an estimated rate can be accepted for a given person. A nutrition coach does not have that judgement to give. When a client's intake form carries a thyroid medication, a pregnancy, a recent unexplained weight change or a history of disordered eating, the calorie target is not the first thing they need, and a referral to their doctor or a registered dietitian is.

Step two: the jump from resting to total is where the error lives

Mifflin's paper reports resting energy expenditure and stops there. Everything above it, the walk to the car and the shift on their feet and the two hours of lifting a week, is the coach's estimate, and it is the largest number in the sum. It is also the one nobody measures.

The obvious shortcut is to ask the client what they eat now. Lichtman and others took ten obese people, nine women and one man, all with a history of losing no weight on a diet they reported as low in calories, and measured them for 14 days with indirect calorimetry and body composition analysis. Their resting metabolic rates were within 5 % of predicted, so nothing was wrong with their metabolism. What was wrong was the reporting. Food intake came in 47 % low on average, with a standard deviation of 16, and reported activity 51 % high, with a standard deviation of 75.

Ten people selected for one trait is not the general population, and nobody should turn 47 % into a correction factor. The useful lesson is narrower and it holds: a reported intake is a floor, not a maintenance figure. If a client says 1,800 kcal and their weight has not moved in three months, their real maintenance is 1,800 or more, and you can set a first target against that without pretending to know the true number.

Step three: set a rate of loss, not a deficit

A deficit in kcal is the wrong thing to choose first, because the number that decides whether a client keeps their muscle is the speed the weight comes off. Helms and others reviewed the contest preparation literature in 2014 and put the rate at approximately 0.5 to 1 % of bodyweight a week to maximise muscle retention. That figure comes from lean, resistance-trained dieters rather than from desk workers, and it is still the most useful rate a coach has. For a 78 kg client it is 0.4 to 0.8 kg a week.

Convert it once, then stop converting. The same review gives the classic arithmetic: a pound of body fat yields about 3,500 kcal, so a daily deficit of 500 kcal theoretically produces about a pound a week. A pound is 0.45 kg, which is 0.6 % of 78 kg, so a 500 kcal deficit is the right order of magnitude for this client. The review does not leave the sum standing on its own.

However, a static mathematical model does not represent the dynamic physiological adaptations that occur in response to an imposed energy deficit.

Step four: two weeks of weigh-ins correct the number

Ask for a daily morning weight and read the seven day average, never the day. Two of those averages, 14 days apart, tell you what the first three steps could not.

• The average falls 0.5 to 1 % of body weight in the week: the target is right, leave it alone.
• The average moves less than a quarter of a kilo: that is noise, and the target is maintenance. Take 10 % off and look again in another 14 days.
• The average falls faster than 1 % : put 10 % back before performance and lean mass start paying for it.

The 10 % step is ours, not a research finding. We suggest it because it is large enough to show up above the noise in a fortnight and small enough that a client who overshoots has not wrecked their week.

Expect the target to keep drifting after that. Helms and others report that in overweight dieters, where metabolic adaptation was observed at all, falls in energy expenditure ran from 79 kcal a day to 504 kcal a day beyond what the weight loss alone predicts, and they say plainly that the intake a diet starts at will need adjusting as body mass drops. A target set in January is not a target in March.

A 34 year old at a desk starts on 1,800 kcal and a daily weigh-in

She is 34, 168 cm, 78 kg, at a desk all day, lifts twice a week, walks a dog, reports about 1,800 kcal and a flat weight for three months, and wants to lose fat.

Mifflin-St Jeor gives 1,499 kcal at rest. Her reported 1,800 with stable weight makes 1,800 a floor for her maintenance, not the ceiling. So the first target is 1,800 kcal with a daily weigh-in, which is a small real deficit if her reporting is low and a break-even week if it is not. Fourteen days later the seven day averages read 78.1 kg and 77.6 kg: 0.5 kg, or 0.6 % of body weight, inside the range Helms and others ask for. Nothing changes. Had the averages read 78.1 and 78.0, the target would have gone to 1,620 and the clock would have restarted.

Protein comes out of the budget before anything else

The calorie figure is a budget, and the first line of it is fixed. Protein for fat loss and muscle gain sets that line in grams per kg, and a client in a deficit needs more of it rather than less, which is exactly when the budget is tightest. Set protein against body weight or lean mass, subtract it, then spend what is left on fat and carbohydrate.

Twenty clients is the same four steps, twenty times

One client is four minutes of arithmetic and a calendar reminder. Twenty clients is that on Sunday night, plus the weigh-in averages, plus the three targets that moved since last week. Software is worth paying for at the point where the arithmetic stops being the hard part and the bookkeeping starts, and that is what Peas & Pans is for.

Set the first number from the equation and put it in front of the client. The next two weigh-ins are the real calculation.