Energy Requirement Calculator (Mifflin, Harris-Benedict, Schofield)
Energy Requirement Calculator: Mifflin-St Jeor, Harris-Benedict, Schofield and Henry
What five published predictive equations yield in kcal/day for one set of inputs — Mifflin-St Jeor 1990, the 1919 Harris-Benedict original, the Roza & Shizgal 1984 revision, Schofield 1985 and the Henry 2005 Oxford equations — shown side by side so the disagreement between them is visible, with the activity and stress multipliers applied as separate, individually sourced steps.
These figures come from published predictive equations and reference ranges. They are not a measurement of the person in front of you, and they are not a feeding prescription — they are a starting estimate for a dietitian or clinician to weigh against the patient, the local policy and, where it is available, measurement.
Predicted resting energy expenditure, and what the multipliers do to it
a 40-year-old man, 70 kg, 175 cm, Mifflin-St Jeor selected, no activity factor and no stress factor
The five equations, the two multipliers, and the one exact conversion
- W
- body weight in kg — actual weight, which is what all five regressions were fitted on
- H
- height in CENTIMETRES. Mifflin-St Jeor and both Harris-Benedict forms take cm; Schofield and Henry do not take height at all. Entering metres here is the commonest way to be wrong by about a thousand kcal on this page
- A
- age in years. In Mifflin and both Harris-Benedict forms it is a linear term that keeps subtracting for ever; in Schofield and Henry it only selects which band’s coefficients are used, and the oldest band is open-ended
- S
- sex in the single-line Mifflin form: 1 for male, 0 for female. Because the paper also prints 166 − 161 = +5 for men and −161 for women, the one-line and the two-line versions of Mifflin-St Jeor are the same equation
- k, c
- the weight coefficient and the constant of the banded equations, in kJ/day for Schofield and MJ/day for Henry. Both band at 18–30, 30–60 and over 60 for adults, with 30 and 60 themselves falling into the OLDER band — which is why the answer steps down at those two birthdays
- 4.184
- the exact number of kilojoules in a thermochemical kilocalorie, which is the kilocalorie used for food energy. It is a definition, not a measurement, so the kcal and kJ rows on this page can never disagree
Worked example
a 40-year-old man, 70 kg, 175 cm, Mifflin-St Jeor selected, no activity factor and no stress factor
Mifflin-St Jeor as the paper fitted it: 9.99 × 70 = 699.3, plus 6.25 × 175 = 1,093.75, minus 4.92 × 40 = 196.8, plus 166 for male, minus 161. That is 699.3 + 1,093.75 − 196.8 + 166 − 161 = 1,601.25 kcal/day, which the headline rounds to 1,601.
Check it against the rounded form the same paper prints for men: 10 × 70 + 6.25 × 175 − 5 × 40 + 5 = 700 + 1,093.75 − 200 + 5 = 1,598.75. The two differ by 2.5 kcal/day. Both are Mifflin-St Jeor 1990; neither is a misquotation of the other.
Now the other four, same person. Harris-Benedict 1919: 66.473 + 13.7516 × 70 + 5.0033 × 175 − 6.755 × 40 = 1,634 kcal/day. Roza & Shizgal 1984: 88.362 + 13.397 × 70 + 4.799 × 175 − 5.677 × 40 = 1,639. The 1984 revision moves this answer by 4 kcal/day — a reminder that the two are close for a middle-aged adult of average build, and that the arguments about which is “the” Harris-Benedict matter less than the arguments about Harris-Benedict against Mifflin.
Schofield, weight only, male 30–60 band: 48 × 70 + 3,653 = 7,013 kJ/day = 7.013 MJ/day, which at exactly 4.184 kJ per kcal is 1,676 kcal/day. Henry/Oxford, male 30–60 band: 0.0592 × 70 + 2.48 = 6.624 MJ/day = 1,583 kcal/day.
So the five answers for one 70 kg man are 1,583, 1,601, 1,634, 1,639 and 1,676 kcal/day. The spread is 93 kcal/day, 5.8% of the Mifflin figure, and the two extremes are the two weight-only equations — Henry low, Schofield high, which is exactly the direction the Henry reanalysis predicts, because the Italian series that dominated Schofield's database had an unusually high BMR per kilogram. Picking one of the five without seeing the other four is picking a number out of a 93 kcal range and calling it a requirement.
Now the multipliers, which is where the page stops being arithmetic. Take him as confined to bed (1.2) with a stress factor of 1.35 for skeletal trauma, as the conventional tables have it: 1,601.25 × 1.2 × 1.35 = 2,594 kcal/day. The activity factor added 320 kcal and the stress factor another 672. The page prints all three figures — 1,601, 1,921 and 2,594 — because the last of them is 62% larger than the first and the reader is entitled to see where that came from.
And then the awkward part. That same combination — Harris-Benedict, activity 1.2, trauma 1.35 — was tested against indirect calorimetry in mechanically ventilated trauma patients, and the adjusted equation overpredicted measured total energy expenditure by about 619 kcal/day, while the unadjusted equation underpredicted by about 289. A separate review of ten years of measured expenditure in hospitalised patients on nutrition support found an average stress factor of about 1.25. So the honest reading of 2,594 is not “he needs 2,594 kcal”; it is “this is what this equation and these two conventional multipliers produce, and the multipliers are the part with the weakest claim on the number”.
Per kilogram, 1,601 kcal/day is 22.9 kcal/kg/day resting and the multiplied figure is 37.1 kcal/kg/day. That is the check a dietitian applies in one second and it is why this page prints it: 37 kcal/kg/day is above the range most general adult guidance describes, so the per-kilogram row flags the multiplied figure before anyone acts on it. For the fluid that goes with this energy figure, see the adult fluid requirement page.
The banded weight-only equations, both sets, as published
| Adult age band | Schofield 1985 — men (kJ/day) | Schofield 1985 — women (kJ/day) | Henry 2005 — men (MJ/day) | Henry 2005 — women (MJ/day) |
|---|---|---|---|---|
| 18–30 | 63 W + 2,896 | 62 W + 2,036 | 0.0669 W + 2.28 | 0.0546 W + 2.33 |
| 30–60 | 48 W + 3,653 | 34 W + 3,538 | 0.0592 W + 2.48 | 0.0407 W + 2.90 |
| over 60 | 49 W + 2,459 | 38 W + 2,755 | 0.0563 W + 2.15 | 0.0424 W + 2.38 |
Five equations, one person: where the disagreement lives
| Person | Mifflin-St Jeor | Harris-Benedict 1919 | Roza & Shizgal 1984 | Schofield 1985 | Henry 2005 | Spread |
|---|---|---|---|---|---|---|
| Man, 70 kg, 175 cm, 40 y | 1,601 | 1,634 | 1,639 | 1,676 | 1,583 | 93 kcal (5.8%) |
| Woman, 70 kg, 175 cm, 40 y | 1,435 | 1,461 | 1,464 | 1,414 | 1,374 | 90 kcal (6.3%) |
| Man, 85 kg, 180 cm, 25 y | 1,856 | 1,967 | 1,949 | 1,972 | 1,904 | 116 kcal (6.2%) |
| Man, 120 kg, 180 cm, 50 y | 2,083 | 2,280 | 2,276 | 2,250 | 2,291 | 208 kcal (10.0%) |
| Woman, 50 kg, 160 cm, 85 y | 920 | 1,032 | 1,038 | 1,113 | 1,076 | 192 kcal (20.9%) |
| Man, 70 kg, 175 cm, 110 y | 1,257 | 1,162 | 1,242 | 1,408 | 1,456 | 294 kcal (23.4%) |
Where each multiplier actually comes from — the provenance table
| Multiplier as usually printed | What it can be traced to | What the evidence says about it |
|---|---|---|
| Activity 1.2 (confined to bed), 1.3 (out of bed) | Long et al., JPEN 1979;3(6):452–6 — a method paper that measured resting expenditure by indirect calorimetry in groups of surgical, trauma, septic and burned patients and then set out this activity-plus-injury adjustment of Harris-Benedict. The abstract describes “a previously measured activity and injury factor” without printing the table, so the figures circulate chiefly through textbooks citing that paper. | The 1.2 for bed rest is the value tested in the Academy of Nutrition and Dietetics evidence analysis of mechanically ventilated trauma patients, where Harris-Benedict adjusted by 1.2 and a 1.35 trauma factor overpredicted measured expenditure by about 619 kcal/day. The direction of that error is consistent across the adjusted-equation literature. |
| Stress 1.2 minor surgery, 1.35 skeletal trauma, 1.6 major sepsis, 2.1 severe burn | The same 1979 Long paper is the citation given almost universally. This page does NOT offer these four as a dropdown, because the author could not verify the table against the primary text, the values differ between the textbooks that reproduce them, and the measurements behind them were made on small numbers of patients by the indirect calorimetry of the 1970s. | The one modern figure the author could verify is an average overall stress factor of about 1.25 across ten years of indirect-calorimetry measurements in hospitalised patients on nutrition support (Barak, Wall-Alonso & Sitrin, JPEN 2002;26(4):231–8). That is a long way below 1.6, and a very long way below 2.1. |
| Nothing — the resting value alone | The published regression, with a standard error of estimate the original paper states: 153 to 167 kcal/day for Schofield’s adult male bands, 108 to 119 for the female ones, and about ±213 and ±201 kcal/day for the Roza & Shizgal men and women. | This is the only row on this table where the uncertainty is a number someone published rather than a number nobody can find. It is also the default this page ships with. |
| A measured figure | Indirect calorimetry on the patient in front of you. | The reference standard, named as such in both the ESPEN and the ASPEN/SCCM critical-care guidelines. Comparisons of predictive equations against it in intensive care have reported accuracy below 40%. Where a measured figure exists, nothing on this page should be used in preference to it. |
What a predictive equation is, what it is not, and why this page prints five of them
An energy requirement is not a measurement, and this page does not pretend otherwise. There is no prescription upstream to read a figure off, the way a drip rate comes off a drug chart. What exists instead is a small set of published regressions, each fitted to a particular population decades ago, each with a standard error its own authors reported, and each giving a different answer for the same person. This page reports what those equations yield for the inputs you give it, names each one, and prints the gap between them. The decision about what to feed somebody is a clinical one that belongs to a dietitian or a clinician, informed by this estimate along with weight history, intake, losses, biochemistry and the direction the patient is travelling in.
The spread between the equations is the most useful thing here, and most calculators hide it. For a 70 kg, 175 cm, 40-year-old man the five equations on this page give 1,583, 1,601, 1,634, 1,639 and 1,676 kcal/day. A reader shown only one of those five has no way to know they were handed a number from a 93 kcal range, and no way to know that the range widens to over 20% in an 85-year-old woman of 50 kg. Worse, the ranking changes: Schofield is the highest of the five for that man and nearly the lowest for a woman of identical height, weight and age, because its female 30–60 weight coefficient is 34 kJ/kg against 48 for men. Any statement of the form “equation X runs high” is therefore wrong as stated; it runs high for some people and low for others, and the only way to know which is to compute all of them, which is cheap.
Which Harris-Benedict? There are two, and they are confused constantly. Harris and Benedict published their original equations in 1919, with the constants 66.473 for men and 655.0955 for women. Roza and Shizgal refitted the same form in 1984, giving 88.362 and 447.593. Both are offered here, labelled by year, because a page that says “revised Harris-Benedict” without a year is ambiguous between them — and because a surprising number of sites using that phrase have in fact implemented Mifflin-St Jeor, which is a different equation by different authors from a different dataset. For most adults the practical difference between the 1919 and 1984 forms is a handful of kcal; the choice that actually moves the answer is Harris-Benedict against Mifflin-St Jeor, and in people with obesity that choice is worth a couple of hundred kcal a day.
Mifflin-St Jeor has two forms, and both of them are in the 1990 paper. The regression as fitted is 9.99 W + 6.25 H − 4.92 A + 166 S − 161 on 498 healthy adults aged 19 to 78, measured by indirect calorimetry; the paper also prints a rounded sex-specific version, 10 W + 6.25 H − 5 A + 5 for men and −161 for women. Those two differ by a couple of kcal a day and this page shows both, because which one a calculator has implemented is almost never stated and the difference, while trivial, is otherwise unexplainable to anyone comparing two sites. Note also that Mifflin-St Jeor predicts RESTING energy expenditure while the other four predict BASAL metabolic rate; these are near neighbours rather than the same quantity, and treating the five rows as five estimates of one number is a small abuse of all of them.
Schofield, Henry, and the Italians. Schofield’s 1985 equations were built from 114 published studies and 7,173 data points, were adopted by the FAO/WHO/UNU report of the same year, and became the standard weight-only equations in British dietetic practice. About 47% of that database — 3,388 data points — came from Italian subjects, whose BMR per kilogram was higher than that of any other group in it, and the equations have since been shown to overestimate BMR in a range of other populations, by over 20% in one Ceylonese series. Henry’s 2005 reanalysis, built from 10,552 BMR values with the Italian series excluded altogether and tropical populations far better represented, produced the Oxford equations, which SACN adopted for UK population requirements; they use the same age bands and the same weight-only form, and they generally sit below Schofield. Both sets band adults at 18–30, 30–60 and over 60, which has a consequence worth stating plainly: the answer steps down on two birthdays. A man of 70 kg gets 1,746 kcal/day from Schofield at 29 and 1,676 at 30, and 1,676 at 59 and 1,408 at 60 — a 268 kcal step for one day of ageing. That is an artefact of banding, not a physiological event, and it is a good reason to read the other rows when somebody is near a band edge.
The multipliers are the weak part, and the page says so where you can see it. Activity and stress factors are multipliers of widely varying provenance. The activity pair offered here — 1.2 confined to bed, 1.3 out of bed — is traced to a 1979 method paper by Long and colleagues, which is also the citation given for the stress factors that circulate in every textbook of clinical nutrition. This page deliberately does not offer those stress factors as a condition-to-number dropdown. The values differ between the books that reproduce them, the author of this page could not verify the table against the primary text, and the one place where the combination was tested against measurement it overshot: Harris-Benedict adjusted by 1.2 for bed rest and 1.35 for trauma overpredicted measured total energy expenditure in mechanically ventilated trauma patients by about 619 kcal/day, while the unadjusted equation underpredicted by about 289. A review of a decade of indirect calorimetry in hospitalised patients on nutrition support put the average stress factor at about 1.25. So the stress factor here is a box you type a number into, with the provenance table above it, and the page shows the resting value, the post-activity value and what each multiplier added, so that a multiplied figure is never the only number on the screen.
Indirect calorimetry is the reference standard, and that is worth one sentence even on a page like this one. Both the ESPEN and the ASPEN/SCCM critical-care guidelines name measured energy expenditure as the standard against which estimates are judged, and comparisons of predictive equations against it in intensive care have reported accuracy below 40%. Where a measurement is available, it displaces everything here. Where it is not — which is most wards, most of the time — the useful output is not a single number but a number with its equation named, its multipliers visible, and the four alternatives printed beside it. For the fluid that is prescribed alongside, use the adult fluid requirement page, which applies the 1 mL per kcal rule to a figure from here; for the protein that goes with this energy figure, the protein requirement page, and for delivering it, the enteral feed rate page or the parenteral macronutrient page; for the running total of what has actually gone in and come out, the fluid balance chart page. Refeeding risk is not assessed here and is not reproduced anywhere in this group — it is covered by the hypophosphataemia and refeeding risk interpreter. Nitrogen balance from a measured urinary urea nitrogen is a separate page, and paediatric energy and fluid are not on this page at all.
Frequently asked questions
Which equation should I use?
This page will not choose for you, and the reason is in the comparison table: the ranking of the five changes with the person. What can be said is narrower and more useful. Mifflin-St Jeor was fitted on a sample that deliberately included obese subjects and is the one usually preferred in obesity, where the 1919 Harris-Benedict tends to read high. Schofield and Henry take weight only, which makes them usable when height is unavailable or unreliable — a common situation in an acutely unwell inpatient — and Henry is the later reanalysis of the same data that SACN adopted in the UK. Beyond that, the honest advice is to use whichever equation your local policy names, so that your figures are comparable with your colleagues’, and to look at the spread on this page before you act on any single row.
Is the original Harris-Benedict or the 1984 revision the “real” one?
Both are real, published equations and the page labels them by year rather than calling either one “revised”. Harris and Benedict published in 1919; Roza and Shizgal refitted the same algebraic form in 1984 on a different dataset. For a middle-aged adult of average build they differ by a handful of kcal a day. The practical problem is not choosing between them, it is that “revised Harris-Benedict” is used on the internet to mean the 1984 refit, the 1919 original with rounded coefficients, and Mifflin-St Jeor, which is neither. If you are reconciling your number with somebody else’s, the first question is which equation they actually ran, not which year they called it.
Where do the activity and stress factors come from?
The activity pair offered here, 1.2 for a patient confined to bed and 1.3 for one out of bed, is traced to Long and colleagues, JPEN 1979, which is also the citation given for the familiar stress factors. The page states that provenance rather than presenting the numbers bare. The stress factor is a free numeric field and not a condition-to-number dropdown, on purpose: the author could not verify the conventional table against the primary paper, the values differ between the textbooks that reproduce them, and where the combination has been tested against measurement it overshot by hundreds of kcal a day. The provenance table on this page sets out, factor by factor, what each one can be traced to and what the evidence says about it. If the honest answer for your unit’s factor is “this is local convention”, that is still better than a number dressed up as a measurement.
Why does the answer jump when I change the age by one year?
Because Schofield and Henry band by age rather than carrying age as a continuous term, and the bands for adults are 18–30, 30–60 and over 60, with 30 and 60 themselves falling into the older band. For a 70 kg man, Schofield gives 1,746 kcal/day at 29 and 1,676 at 30, then 1,676 at 59 and 1,408 at 60 — a 268 kcal step. That discontinuity is an artefact of how the equations were published, not a physiological event, and it is one reason to look at the Mifflin and Harris-Benedict rows as well when somebody is close to a band edge: those three carry age as a smooth linear term and move by only a few kcal per year.
Why is the per-kilogram figure shown, and what counts as sensible?
Because it is the check a dietitian applies in a second and the one that catches a unit error or a runaway multiplier before anybody acts on the number. A resting value for an adult usually lands somewhere around 20 to 25 kcal/kg/day, and the general adult guidance most often quoted for a total runs 25 to 30 kcal/kg/day, with the critical-care guidelines describing a lower band in the acute phase. The page flags below 20 and above 35 kcal/kg/day in the result note. Those flags are sanity checks on the arithmetic, not targets: this page has no view on what anybody should be fed.
Does this page tell me what to feed somebody?
No. It reports what a named published equation yields for the inputs given, what your multipliers did to that output, and what four other equations give for the same person. Translating that into a feeding plan involves the weight trend, actual intake, losses, biochemistry, refeeding risk, route, tolerance and the patient’s own wishes, and it is a decision the dietitian or clinician owns. The clearest marker of that boundary is the default state of this page: with both multipliers at 1.00 the headline is simply the equation’s own output, which is the only figure here with a published uncertainty attached to it.
Why does it refuse an age under 18?
Because Schofield and Henry have separate paediatric bands below 18 with quite different coefficients, because Mifflin-St Jeor was fitted on adults aged 19 to 78, and because a page that silently applied an adult equation to a twelve-year-old would be producing a plausible wrong answer rather than an error. Paediatric energy estimation is a different calculation; this page declines it rather than approximating it. The paediatric maintenance FLUID case is covered separately by the paediatric maintenance fluid page in the medical set.
Why kilojoules as well as kilocalories?
Because Schofield and Henry were published in kJ/day and MJ/day respectively, and converting them to kcal is a step at which errors enter. The conversion used here is exactly 4.184 kJ per kilocalorie, which is the definition of the thermochemical kilocalorie used for food energy — not 4.1868, which is the international steam-table calorie and a different definition. Because it is a definition rather than a measurement, the kcal and kJ rows on this page cannot disagree, and the MJ/day rows let you check the banded equations against their original papers in the unit those papers used.
What about indirect calorimetry?
It is the reference standard, named as such in both the ESPEN and the ASPEN/SCCM critical-care guidelines, and where a measured figure exists it should be used in preference to anything on this page. Published comparisons of predictive equations against measured expenditure in intensive care have reported accuracy below 40%, and the adjusted-equation approach — a predictive equation multiplied by conventional activity and stress factors — has performed worse than the unadjusted equation in at least one well-known comparison, overshooting by about 619 kcal/day where the unadjusted equation undershot by about 289. That asymmetry is the strongest single argument for showing the resting value, the post-activity value and the final figure separately, which is what this page does.
Related calculators
References
- Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr 1990;51(2):241–7. Primary source for the equation implemented here. The regression as fitted is REE = 9.99 × weight + 6.25 × height − 4.92 × age + 166 × sex (male 1, female 0) − 161, R² = 0.71, from 498 healthy subjects (247 female, 251 male) aged 19–78 measured by indirect calorimetry; the paper also prints the rounded sex-specific form (10 / 6.25 / 5, +5 for men and −161 for women). Both forms are computed on this page. Coefficients verified against the paper’s abstract and independently against the Endotext table “Estimating Resting Metabolic Rate” (Dwyer JT, Melanson KJ, Sriprachy-anunt U et al., Dietary Treatment of Obesity, Endotext, updated 2015), which prints 9.99 / 6.25 / 4.92 explicitly.
- Harris JA, Benedict FG. A Biometric Study of Basal Metabolism in Man. Carnegie Institution of Washington, Publication no. 279, 1919. The 1919 ORIGINAL coefficients implemented here: men 66.473 + 13.7516 W + 5.0033 H − 6.755 A, women 655.0955 + 9.5634 W + 1.8496 H − 4.6756 A, with W in kg, H in cm and A in years. A 1919 publication, so the equation itself is long out of copyright as well as being a method rather than an expression.
- Roza AM, Shizgal HM. The Harris Benedict equation reevaluated: resting energy requirements and the body cell mass. Am J Clin Nutr 1984;40(1):168–82. The 1984 REVISION implemented here: men 88.362 + 13.397 W + 4.799 H − 5.677 A, women 447.593 + 9.247 W + 3.098 H − 4.330 A, with a reported confidence range of about ±213 kcal/day for men and ±201 for women. These eight coefficients are the ones most often mislabelled online, and were cross-checked against two independent compilations before being implemented.
- Schofield WN. Predicting basal metabolic rate, new standards and review of previous work. Hum Nutr Clin Nutr 1985;39 Suppl 1:5–41. Source of the banded weight-only equations implemented here, in kJ/day: men 63 W + 2,896 (18–30), 48 W + 3,653 (30–60), 49 W + 2,459 (over 60); women 62 W + 2,036, 34 W + 3,538, 38 W + 2,755 for the same bands. Standard errors of estimate 641, 700, 686 kJ/day for men and 497, 465, 451 for women. Adopted by the FAO/WHO/UNU 1985 report Energy and protein requirements (WHO Technical Report Series 724), whose own text records that adults over 60 were under-represented in the database and that this part of it needed expansion. Schofield also published weight-and-height variants; this page implements the weight-only set.
- Henry CJK. Basal metabolic rate studies in humans: measurement and development of new equations. Public Health Nutr 2005;8(7a):1133–1152 (volume, issue and page range verified against the publisher’s record). Source of the Oxford equations implemented here, built from 10,552 BMR values with the Italian series excluded, in MJ/day: men 0.0669 W + 2.28 (18–30), 0.0592 W + 2.48 (30–60), 0.0563 W + 2.15 (over 60); women 0.0546 W + 2.33, 0.0407 W + 2.90, 0.0424 W + 2.38. Also the source for two facts this page states: that 3,388 of the 7,173 data points in the Schofield database — about 47% — were Italian subjects with a higher BMR per kilogram than any other group in it, and that the FAO/WHO/UNU equations overestimate BMR in a range of other populations, by 22.4% in one Ceylonese male series and about 12.8% in an Indian one.
- Long CL, Schaffel N, Geiger JW, Schiller WR, Blakemore WS. Metabolic response to injury and illness: estimation of energy and protein needs from indirect calorimetry and nitrogen balance. JPEN J Parenter Enteral Nutr 1979;3(6):452–6. The citation given almost universally for the activity and stress factors applied to Harris-Benedict. PROVENANCE NOTE, stated here because it is part of what this page had to establish: the abstract describes adjusting the Harris-Benedict value upward “using a previously measured activity and injury factor” but does not print the multiplier table, and the author of this page could not obtain the primary text to verify the individual values. The two activity factors offered on this page (1.2 confined to bed, 1.3 out of bed) are the pair consistently attributed to this paper across the clinical nutrition literature and the pair used in the published validation cited below. The familiar stress factors attributed to the same paper (1.2 minor surgery, 1.35 skeletal trauma, 1.6 major sepsis, 2.1 severe burn) are NOT offered as options on this page, because they could not be verified at source and because they differ between the textbooks that reproduce them.
- Academy of Nutrition and Dietetics Evidence Analysis Library, critical illness worksheet on the best method to estimate resting metabolic rate (andeal.org, worksheet 252007), summarising a comparison in mechanically ventilated trauma patients. Unadjusted Harris-Benedict underpredicted measured total energy expenditure with a bias of about 289 kcal/day; Harris-Benedict adjusted by an activity factor of 1.2 for bed rest and a trauma stress factor of 1.35 OVERPREDICTED it by about 619 kcal/day. Cited for those two figures, which are the strongest evidence this page found on whether the conventional multipliers do what they claim.
- Barak N, Wall-Alonso E, Sitrin MD. Evaluation of stress factors and body weight adjustments currently used to estimate energy expenditure in hospitalized patients. JPEN J Parenter Enteral Nutr 2002;26(4):231–8. A decade of indirect-calorimetry measurements in hospitalised patients requiring nutrition support, from which an average overall stress factor of about 1.25 relative to Harris-Benedict was derived. Cited as the one modern, traceable alternative to the 1979 multiplier table, and as the reason this page flags a combined multiplier of 1.5 or more.
- Oshima T, Berger MM, Singer P and colleagues as reviewed in Journal of Intensive Care 2019;7:article 71 (energy expenditure in critical illness). Cited for two statements made on this page rather than for any coefficient: that indirect calorimetry is named the reference standard for energy expenditure in both the ESPEN and the ASPEN/SCCM critical-care guidelines, and that comparisons of predictive equations against measured expenditure in intensive care have reported accuracy below 40%. Guideline recommendation content is cited, not reproduced.
- Thermochemical kilocalorie: 1 kcal ≡ 4.184 kJ exactly, by definition, and this is the kilocalorie used for food energy. The international steam-table calorie (4.1868 J) and the 15 °C calorie (about 4.1855 J) are different definitions and are not used here. A defined conversion factor is not anybody’s intellectual property, which is why the Schofield and Henry equations can be printed on this page in both their published MJ or kJ form and in kcal without the conversion introducing any uncertainty of its own.
- Licensing position taken for this page, recorded deliberately: the equations above are methods and are reproduced with attribution, which is normal scholarly use. No part of this page is built on NICE text, criteria or tables, and the ESPEN and ASPEN/SCCM guidelines are cited for the existence and direction of a recommendation rather than quoted. The 1919 Harris-Benedict publication is additionally out of copyright. No criteria list is reproduced anywhere on this page; the refeeding criteria that would ordinarily appear beside an energy calculator are deliberately absent and are covered on a separate interpreter page.
CalcEngines health calculators are for education and for checking arithmetic that has already been decided elsewhere. They are not medical advice, they do not decide what to give, and they do not replace the judgement of a doctor, nurse, midwife or dietitian who knows the person in front of them. Every figure depends on the values you enter and on the assumptions stated on the page — check it against the prescription, the product label and your local policy before acting on it.