Affiliate disclosure
Book titles on this page link to Amazon. As an Amazon Associate, DataField.Dev earns from qualifying purchases — at no additional cost to you.
Chapter 5 — Further Reading
Where the equations came from
The Mifflin-St Jeor equation was published in the early 1990s in the American Journal of Clinical Nutrition, developed because the older Harris-Benedict equations (originally from 1919, revised since) were performing poorly in contemporary populations.
Worth reading the original if you can access it, for one reason: the paper reports its own error. The authors are clear about the spread of individual results around the predicted value. That information is present in the source and absent from every calculator built on it, which is a small but perfect illustration of what happens to uncertainty as it travels. For: students, dietitians, and anyone who has ever trusted a calculator.
Comparative validation studies of predictive equations. Search PubMed for "predictive equations resting metabolic rate validation" and read a comparison paper. You'll find that no equation is reliably accurate at the individual level, that all of them degrade at the extremes of body composition, and that the field knows this perfectly well. For: clinicians and coaches who quote these numbers to clients.
On organ-level metabolic rates
The figures in §5.2 — brain ~240, liver ~200, heart and kidneys ~440, muscle ~13, adipose ~4.5 kcal/kg/day — come from a body of organ-tissue metabolic work, much of it associated with analyses by Marinos Elia and subsequently refined by imaging-based body composition studies.
Search for "organ tissue energy expenditure" or "organ-specific metabolic rate" reviews. These are the numbers that quietly demolish the "muscle burns tons of calories" claim, and reading how they were derived is a good exercise in appreciating how hard the measurement is. For: anyone who found §5.2 surprising — which is most people.
On metabolic adaptation
The six-year follow-up of televised weight-loss competition contestants (published in Obesity, 2016, by Fothergill and colleagues, including Kevin Hall). Read the paper itself rather than the coverage, and read at least two critiques alongside it. The study's central strength — it measured RMR and body composition rather than estimating them — and its central weaknesses — small, no control group, extreme intervention — are both visible in the paper. For: everyone who has heard "metabolic damage" cited. Which is everyone.
Contrast reading: the National Weight Control Registry (nwcr.ws) and the literature on
successful weight-loss maintainers. People who maintain substantial loss through more moderate means
do not consistently show adaptation of that magnitude — which is the other half of the picture, and
the half that rarely makes the news.
For: balance. Read this immediately after the study above.
The broader adaptive thermogenesis literature. Search "adaptive thermogenesis weight loss review." The magnitude is genuinely contested — estimates range from trivial to a few hundred kcal/day — and reading the disagreement teaches more than any single figure would. For: students and clinicians.
On the constrained energy expenditure question
Herman Pontzer's research and his book Burn (Avery, 2021).
Pontzer's doubly-labelled-water work across populations, including with the Hadza, underpins the proposal that total daily energy expenditure responds less to physical activity than a simple additive model predicts. Genuinely unsettled; the book states it more confidently than the evidence currently supports. Read as a serious hypothesis being argued well. For: the curious. Note that if it's right, it strengthens Chapter 4 §4.6.
Research on energy expenditure across the human lifespan. Search for large doubly-labelled-water analyses examining how expenditure changes with age. The finding that adjusted expenditure is relatively stable through much of adult life, declining meaningfully only later, is directly relevant to §5.5 and directly contradicts the folklore. For: anyone over thirty who has blamed their metabolism.
On thyroid and the medical differential
The American Thyroid Association (thyroid.org) — patient-facing and professional resources on
hypothyroidism, including honest material on the relationship between thyroid function and body
weight, which is more modest than commonly assumed in both directions.
For: anyone who suspects a thyroid problem, and anyone who has been told they have one.
NIH MedlinePlus (medlineplus.gov) for plain-language, non-commercial summaries of PCOS,
Cushing's syndrome, sleep apnea, and medication side effects. Free, reliable, and free of the
supplement advertising that surrounds most search results on these topics.
For: everyone. Send this link instead of arguing.
On the supplement claims
NIH Office of Dietary Supplements fact sheets on weight-loss supplements specifically
(ods.od.nih.gov). Genuinely even-handed, regularly updated, and it goes through the individual
ingredients — green tea extract, garcinia, chromium, L-carnitine, raspberry ketones, bitter orange —
one at a time, saying plainly where evidence is absent.
If you read one thing before buying anything from the metabolism aisle, read this. For: everyone. The single best free resource for Case Study 2.
FDA warning letters and consumer updates on weight-loss products (fda.gov). The recurring
finding — that some products marketed as natural weight-loss aids have been found to contain
undeclared pharmaceutical ingredients — is more persuasive than any argument about evidence.
For: anyone still tempted.
On green tea extract and liver injury: the NIH LiverTox database (livertox.nih.gov) is a
free, authoritative resource on drug- and supplement-induced liver injury. Look up green tea extract
and garcinia. This is the sort of information that exists, is free, and is never on the box.
For: everyone who assumes "natural" means "can't hurt."
A note on calculators
Every online TDEE calculator you will find uses one of about four equations, all of which have ±10% individual error, and none of which will show you that error.
If you want one, use one — they're all much the same. Just do Exercise C5 first: put identical inputs into three of them and look at how far apart the answers are. The disagreement between calculators is itself the error bar, displayed for free.