Appendix E — Master Claim Index

Every claim this book assessed, with its verdict and the chapter that argued it.

How to use this

⚠️ This is a LOOKUP TABLE, not a reading list. It exists for one job, described in Chapter 38 §38.7: before you act on a claim, check whether it is already in here with a verdict attached.

⚠️ A surprising amount of what will arrive as news over the next decade is already below. ⚠️ Chapter 37 §37.6b explains why: refutation is not distributed the way the claim was, the audience turns over, and the kernel underneath is usually still true — so claims come back.

⚠️ Two cautions.

⚠️ First: a verdict here is a summary of an argument, not a substitute for it. The chapter reference is part of the entry because the reasoning is where the value is (Chapter 38 §38.2).

⚠️ Second: this index is a snapshot. ⚠️ Chapter 37 §37.9 named the claims most likely to move, and Chapter 38 §38.8 gives the procedure for when one does — check WHICH LEVEL was wrong, check whether it was MARKED, adjust rather than abandon.

⚠️ A 🟡 that turns out false is the system working. A ✅ that turns out false is a real error, and you should downgrade your confidence in whatever else came from the same reasoning.

The verdict scale

Verdict	What it means
✅ Well supported	⚠️ Multiple lines of good evidence agree. Act on it.
🟢 Probably true	⚠️ The evidence leans this way and is not conclusive. Reasonable to act on.
🟡 Unclear / it depends	⚠️ Genuinely uncertain, or the answer depends on who and how much. Do not let anyone tell you otherwise.
🟠 Probably false	⚠️ The evidence leans against it. Not disproved, but do not build anything on it.
❌ Not supported	⚠️ Tested and not supported, or supported by nothing.
⚗️ Untested	⚠️ Nobody has properly tested it in humans. Not the same as false.

The distribution

Verdict	Count	Share
✅ Well supported	133	28.4%
🟢 Probably true	97	20.7%
🟡 Unclear / it depends	91	19.4%
🟠 Probably false	67	14.3%
❌ Not supported	69	14.7%
⚗️ Untested	11	2.4%
TOTAL	468
Affirmative (✅ + 🟢)	230	49.1%
Uncertain or negative	238	50.9%

⚠️ Chapter 38 §38.5 discusses this table rather than celebrating it, and the discussion matters more than the numbers.

⚠️ In short: ✅ is the largest single category, which is not what a sceptical book expects to end with. ⚠️ The explanation, first given at Chapter 29 §29.16, is that a large share of those ✅s are clinical, legal, definitional and physiological FACTS rather than dietary FINDINGS — "ingredients are listed in descending order by weight" is a ✅ about labelling law, not a discovery about food.

⚠️ The scepticism in this book was never about whether facts exist. It was about which claims about DIET are supported — and in that narrower category the distribution is much less affirmative.

⚠️ You can check that for yourself, and Chapter 38's exercise S4 asks you to: sample ten ✅ entries below and classify each as a dietary finding or another kind of fact.

Part 1 — By verdict

⚠️ Use this half when you want to know what the book concluded about a CATEGORY — particularly the ❌ and 🟠 sections, which Chapter 38 §38.5 identifies as where most of this book's financial value sits.

✅ Well supported — 133 claims

Claim	Ch
Folic acid before conception prevents neural tube defects	13
Vegans need B12 supplementation	13
Vitamin D supplementation in the genuinely deficient	13
Iodine adequacy in pregnancy matters for neurodevelopment	14
Iron supplementation helps the iron-deficient, including without anemia	14
Oral rehydration solution for diarrhoeal illness	15
Creatine monohydrate improves strength, power and lean mass	16
Cooking tomatoes increases lycopene bioavailability	17
Soda taxes reduce purchases of taxed drinks	18
Sugar-sweetened beverages contribute to weight gain and type 2 diabetes	18
The sugar industry funded and shaped an undisclosed 1967 NEJM review	18
Linoleic acid is essential; its absence causes a named deficiency syndrome	19
Modern linoleic acid intake is historically unprecedented (adipose tissue LA rose over 50 yrs)	19
Replacing saturated fat with PUFA lowers LDL-C	19
Approved GM foods currently on the market are safe to eat	20
Basic food hygiene reduces foodborne illness (handwashing, separation, temperature, refrigeration)	20
Eating more fruit and vegetables is beneficial	20
Certain groups should not fast without medical supervision	21
Ultra-processed diets cause greater ad libitum energy intake than nutrient-matched minimally-processed diets	22
Carbohydrate availability determines sustained high-intensity endurance performance	23
Carbohydrate intake during prolonged exercise improves performance	23
Fat adaptation (raised fat oxidation) occurs with sustained low-carbohydrate diets	23
Programmed over-drinking during endurance events is dangerous	23
Sustained low energy availability impairs health and performance (RED-S)	23
Third-party certification is necessary for drug-tested athletes (strict liability)	23
Adaptive thermogenesis occurs after weight loss	24
Appetite adaptation after weight loss persists (ghrelin up, leptin/PYY/CCK down, at 12 months)	24
BMI heritability is approximately 40-70%	24
Bariatric surgery produces durable loss and improved long-term outcomes	24
GLP-1 receptor agonists produce substantial weight loss	24
Higher protein and resistance training preserve lean mass during a deficit	24
Modest weight loss produces disproportionate health benefit (DPP: ~7% loss, ~58% risk reduction)	24
Most people who lose weight regain most of it	24
Several common medication classes cause substantial weight gain	24
The population-level rise in body weight is environmental	24
Weight loss requires an energy deficit	24
B12 deficiency is common in older adults	25
Breastfeeding reduces GI and respiratory infections in infancy	25
Early allergen introduction reduces peanut allergy (LEAP)	25
Folic acid preconception reduces neural tube defects	25
Formula is a safe and adequate alternative supporting normal growth	25
Iodine and iron requirements rise substantially in pregnancy	25
Lactation energy requirements exceed any trimester of pregnancy	25
Older adults require more protein per kg than the general adult RDA	25
Peak bone mass is largely accrued by the late teens to early twenties	25
Recurrent chest infections in an older person warrant a swallowing assessment	25
Resistance training is the most effective intervention against sarcopenia	25
Unintended weight loss in older adults is a red flag requiring investigation	25
Vitamin D supplementation for breastfed infants	25
Alcohol is a Group 1 carcinogen and excess adiposity raises risk of 12+ cancers	26
LDL-C causally contributes to atherosclerotic cardiovascular disease	26
Physical activity improves insulin sensitivity independently of weight	26
Potassium-enriched salt substitutes reduce stroke and cardiovascular events (SSaSS)	26
Reducing sodium lowers blood pressure, dose-dependently	26
The DASH dietary pattern lowers blood pressure	26
Type 2 diabetes can go into remission through weight loss (DiRECT), dose-dependent	26
Weight loss improves fatty liver histology, dose-dependently	26
A low-FODMAP diet reduces symptoms in a substantial proportion of people with IBS	27
Butyrate is the primary energy source for colonocytes; fermentable fibre is the substrate	27
Exclusive enteral nutrition induces remission in paediatric Crohn's disease	27
FMT cures recurrent Clostridioides difficile infection	27
Human milk oligosaccharides are indigestible by the infant and feed bifidobacteria	27
Specific probiotic strains prevent necrotizing enterocolitis in preterm infants	27
Antihistamines do not treat anaphylaxis	28
Coeliac disease is autoimmune and requires strict lifelong gluten avoidance	28
Coeliac testing is only valid while the person is eating gluten	28
Early allergen introduction reduces the risk of developing food allergy	28
Intramuscular adrenaline is first-line for anaphylaxis; delay is associated with death	28
Lactase non-persistence is the global norm and is dose-dependent	28
Milk, egg, wheat and soy allergies are often outgrown; peanut, tree nut, fish and shellfish usually persist	28
Oral immunotherapy achieves desensitization, not cure	28
Skin prick and specific IgE tests measure sensitization, not allergy	28
Bed rest causes rapid measurable loss of muscle mass and function	29
Early full-dose feeding in critical illness is not superior to starting low and advancing	29
Home enteral nutrition supports independent living	29
Hospital malnutrition is common and associated with worse outcomes	29
Nasogastric tube position must be confirmed before use (pH and/or radiography)	29
Preoperative nutritional optimization in malnourished patients reduces complications	29
Prolonged preoperative fasting is unnecessary (clear fluids ~2h, light meal ~6h)	29
Refeeding a starved person too rapidly causes electrolyte collapse and death	29
Warfarin requires consistent rather than minimal vitamin K intake	29
"Gluten-free" is legally defined at 20 ppm or below	30
"Use by" is a safety date; "best before" is a quality date	30
Declared allergens are mandatory, emphasized and enforceable	30
Declared nutrition values may be calculated rather than measured, with tolerances around 20%	30
Food may be frozen up to the use-by date, resetting the clock	30
Ingredients are listed in descending order by weight	30
QUID: a named, pictured or emphasized ingredient must have its percentage declared (EU/UK)	30
kcal per 100 g is energy density and is mandatory on European labels	30
Household food waste is concentrated in fresh produce, bread and prepared leftovers, and is driven by over-buying, poor storage and date-label confusion	31
What is available in the home is a major determinant of what gets eaten	31
Food insecurity and obesity co-occur rather than being opposites	32
Food insecurity is associated with poorer diet quality, worse mental health, and higher rates of obesity and type 2 diabetes	32
Higher prices reduce consumption of the costlier item, more so in lower-income households	32
Low-income households pay more for equivalent goods (the poverty premium)	32
The cheapest foods per calorie and the cheapest per nutrient are different foods	32
"Wanting" (incentive salience) and "liking" (hedonic pleasure) are separable and can diverge	33
Homeostatic and hedonic feeding are dissociable systems with different substrates and different levers	33
People adjust their intake toward that of their eating companions without noticing (modelling)	33
Pressuring a child to eat a food reduces liking for it, persistently	33
Restricting a specific palatable food in children increases wanting and unsupervised consumption	33
Restriction increases the salience of and preoccupation with food	33
Sensory-specific satiety increases intake when variety is available	33
What is available in the immediate environment substantially determines what is eaten	33
Autistic people are over-represented in anorexia nervosa and ARFID	34
Duration of untreated illness predicts outcome and is a modifiable factor	34
Eating disorders are psychiatric illnesses with substantial heritability and among the highest mortality of any psychiatric category	34
Eating disorders occur across sexes, ages, ethnicities, body sizes and socioeconomic groups	34
Malnutrition itself produces rigidity, preoccupation, low mood and social withdrawal that resemble personality	34
Medical complications of restriction follow from the behaviour and the rate of change, not from a position on a weight chart	34
Most people with eating disorders are not underweight	34
Refeeding syndrome is a real, potentially fatal risk requiring medical monitoring	34
Individual postprandial glucose responses to identical foods differ substantially between people	35
Lactase persistence and alcohol-flushing genotypes are real and clinically meaningful	35
Personalization based on diagnosed allergy, coeliac disease, intolerance, disease, medication interaction, life stage, athletic demand or documented deficiency is evidence-based	35
The glycaemic index is a population average that conceals large individual variation	35
Those responses are reasonably stable within a person and partly predictable from measurable features	35
Eliminating industrial trans fat reduces population intake and cardiovascular risk	36
Food marketing influences children's food preferences, requests and consumption	36
Food waste is a major driver of food's environmental impact	36
Roughly a third of food produced is never eaten, upstream in lower-income and downstream in higher-income countries	36
Ruminant meat has substantially higher environmental impact per unit protein or calorie than plant sources	36
Sugar-sweetened beverage taxes reduce purchased sugar, substantially via manufacturer reformulation	36
The environmentally optimal and nutritionally optimal dietary patterns converge substantially	36
Within-category producer variation is large, but the highest-impact plant producers still beat the lowest-impact beef producers	36
A boring, stable recommendation is therefore a certain one	37
A plant-forward, minimally-processed dietary pattern with adequate protein is supported across independent lines of evidence	37
Adherence dominates composition in free-living dietary outcomes	37
Most supplements provide no benefit to people without a documented deficiency	37
Population dietary guidance across countries agrees substantially on the core pattern	37
Sugar-sweetened beverages are the clearest single dietary target	37
There is no protective dose of alcohol	37
Intention predicts behaviour much less well than people expect, and situation explains more of the gap than motivation	38

🟢 Probably true — 97 claims

Claim	Ch
A fiber supplement replaces high-fiber foods	11
A daily multivitamin is good insurance	13
Calcium supplements prevent fractures in healthy adults	14
Electrolyte replacement beyond ~60–90 min hard exercise / heat / salty sweaters	15
Caffeine improves exercise performance	16
Chicken soup helps when you have a cold	17
Ginger helps with nausea	17
Rinsing rice (and cooking in excess water) is worth doing	17
Sugar contributes to obesity (via energy intake)	18
The added-vs-natural sugar distinction is meaningful	18
Repeatedly heated commercial frying oil produces harmful oxidation products	19
Replacing saturated fat with PUFA reduces cardiovascular events	19
Two portions of oily fish per week (preformed EPA/DHA)	19
Organic milk and meat have higher omega-3 proportions	20
Organic produce has lower pesticide residues	20
Hunger is entrained to habitual meal times and comes in waves	21
IF helps some people eat less (adherence case)	21
IF produces weight loss comparable to continuous energy restriction	21
Late-night eating is worse for glycemic control than the same food earlier	21
Energy density and eating rate mediate the intake effect	22
Higher UPF intake is associated with adverse health outcomes	22
Ultra-processed food is cheaper per calorie than minimally processed food	22
Caffeine, nitrate, beta-alanine and sodium bicarbonate improve performance in defined contexts	23
Drink-to-thirst is adequate for most athletes in most conditions	23
Multiple transportable carbohydrates permit higher oxidation rates and reduce GI distress	23
Protein distribution across the day (~0.3 g/kg, 3-5 meals) beats the same total in fewer meals	23
The gut is trainable for carbohydrate tolerance during exercise	23
Physiological adaptation rather than willpower explains most weight regain	24
Semaglutide reduces major adverse cardiovascular events in a defined high-risk group (SELECT)	24
Weight stigma worsens the outcomes it purports to address	24
Calcium plus vitamin D in frail and institutionalized older adults	25
Oily fish twice weekly in pregnancy (with species advice)	25
Parental weight talk is associated with disordered eating and weight gain in adolescents	25
Pressure and restriction in child feeding backfire (in opposite directions)	25
Repeated neutral exposure increases acceptance of new foods in children	25
ApoB is a better risk marker than LDL-C where the two are discordant	26
Coffee is associated with less liver disease progression	26
Population-level sodium reduction reduces cardiovascular events	26
The Mediterranean pattern reduces cardiovascular events in high-risk people (PREDIMED)	26
Fermented foods increase microbiota diversity and reduce inflammatory markers	27
Peppermint oil, gut-directed hypnotherapy and CBT improve IBS symptoms	27
Prebiotics change microbiota composition as described	27
S. boulardii or L. rhamnosus GG reduce antibiotic-associated diarrhoea	27
Cofactors (exercise, alcohol, NSAIDs, infection) can convert a tolerated exposure into anaphylaxis	28
Oral allergy syndrome proteins are heat-labile, so cooked forms are usually tolerated	28
Sulphites can provoke bronchoconstriction in people with asthma	28
Early postoperative oral intake is safe after most abdominal surgery	29
Oral nutritional supplements benefit malnourished older adults	29
Prehabilitation before major surgery improves recovery	29
Preoperative carbohydrate loading (ERAS)	29
Front-of-pack warning labels change purchasing behaviour	30
Serving sizes are used to make per-serving figures look smaller	30
Long-term diet-trial dropout follows discrete failures rather than gradual decay	31
Online grocery ordering reduces impulse purchasing by removing the aisle	31
Shopping without a list produces more unplanned purchases, skewed toward Group 4 foods	31
Cost is a major barrier to diet quality independent of nutritional knowledge	32
Food purchasing and intake vary systematically across the pay/benefit cycle	32
Income supplementation and cash transfers improve food security and diet quality	32
Waste as a percentage of food purchased is lower, not higher, in lower-income households	32
Alcohol acutely stimulates appetite and impairs restraint, in addition to supplying unregistered energy	33
Chronic stress and low mood increase intake of highly palatable food specifically	33
Eating slowly and without distraction reduces intake at that meal	33
Implementation intentions ("when X, I will Y") improve follow-through	33
Intuitive eating is associated with better psychological outcomes and less disordered eating	33
Meals eaten with others are larger and longer than meals eaten alone	33
Mindful eating reduces binge-eating episodes	33
Restricting a specific food increases craving for that specific food	33
Short sleep increases intake, especially of highly palatable food	33
CBT-E is the best-supported approach for adults across eating disorder diagnoses	34
Early nutritional restoration improves psychological symptoms independent of therapy	34
Family-based treatment is the best-supported treatment for adolescents with anorexia nervosa	34
Food insecurity is a risk factor for binge eating and eating-disorder symptoms	34
Insulin omission as a weight-control behaviour occurs in type 1 diabetes and carries serious risk	34
LGBTQ+ people, and gay and bisexual men particularly, have elevated rates	34
Men, older adults, people in larger bodies and minoritized groups experience longer diagnostic delay	34
Pro-eating-disorder online content is harmful to those exposed	34
An algorithm can design a diet that lowers an individual's postprandial glucose	35
Matching a diet to preference, culture, budget and schedule is the largest available personalization lever	35
Personalized-nutrition outputs converge substantially on conventional dietary advice	35
Well-designed n-of-1 experiments can answer short-latency, high-signal individual questions	35
Air-freighted produce is a genuine high-emission exception	36
Front-of-pack warning labels change purchasing	36
Interventions that change the default outperform interventions requiring a decision	36
Mandatory reformulation targets reduce population intake of the targeted nutrient	36
Retail listing and shelf position are substantially determined by purchased placement rather than demand alone	36
School food standards and public procurement standards improve intake in the populations reached	36
Ultra-processed food is abundant primarily because it solves supply-side business problems	36
Claims that ADJUST the established dietary pattern are more often correct than claims that REPLACE it	37
Novelty predicts evidential weakness in nutrition claims	37
Only six categories of dietary requirement are genuinely non-optional for a healthy adult	37
Recommendations built on dietary patterns are more robust to mechanistic error than recommendations built on mechanisms	37
Refuted nutrition claims recur because refutation is distributed differently from the claim	37
The individual-personalization and planetary-environmental analyses are at least partly independent	37
Implementation intentions bridge the intention–behaviour gap better than exhortation	38
Interpreting a lapse as a personal failing predicts abandonment; interpreting it as predictable does not	38
Media-literacy and inoculation interventions produce real short-term effects	38
People who appear to have more self-control report exercising it less often	38

🟡 Unclear / it depends — 91 claims

Claim	Ch
Nutrition science is bought and paid for by industry	1
There's a study for everything, so you can prove anything about food	1
Processed meat is Group 1 — same as smoking, no safe amount	2
"Leaky gut syndrome" causes chronic illness; this protocol fixes it	3
95% of your serotonin is in your gut, so gut health = mental health	3
A calorie is a calorie	4
Green tea / capsaicin / caffeine / cold exposure usefully boost metabolism	5
Blood sugar "spikes" harm metabolically healthy people	7
Erythritol causes heart attacks and strokes	7
Glycemic index tells you which carbohydrates are healthy	7
You need a protein supplement to hit your target	8
Fish oil supplements prevent heart disease	9
Saturated fat causes heart disease; limit to <10% of calories	9
Almost everyone is vitamin D deficient and should supplement	13
Vitamin C prevents colds	13
Almost everyone is magnesium deficient; supplements fix sleep/anxiety/cramps	14
Everyone should reduce sodium below 2,300 mg	14
Zinc lozenges shorten colds	14
Mild dehydration significantly impairs cognition	15
Non-stick cookware leaches toxic chemicals into food	17
Plastic containers leach hormone-disrupting chemicals	17
Eating sugar directly causes type 2 diabetes	18
Fructose has harmful effects independent of calories	18
Non-sugar sweeteners are harmful / are a free substitution	18
Soda taxes improve health outcomes	18
Sugar is addictive	18
Refining loses antioxidants and generates small amounts of trans fat	19
The recovered trials (Sydney, Minnesota) refute the diet-heart hypothesis	19
Glyphosate causes cancer	20
Organic farming is better for the environment	20
Organic food is more nutritious	20
Fasting reverses type 2 diabetes	21
IF causes greater lean mass loss than continuous restriction	21
Meal timing has metabolic effects independent of calories	21
Hyper-palatability and weakened satiety signalling mediate the effect	22
The NOVA classification is reliable and reproducible between coders	22
UPF harms health independently of energy intake and nutrient profile	22
Carbohydrate periodization ("train low") improves competitive performance	23
Low-carbohydrate diets improve ultra-endurance performance at low relative intensity	23
GLP-1 receptor agonists are safe and effective over decades	24
Weight cycling is harmful	24
Weight-neutral approaches are the right clinical target	24
Which model of weight regulation is correct (set point / settling point / dual intervention)	24
Baby-led weaning versus purees	25
Breastfeeding improves long-term outcomes (IQ, obesity, chronic disease)	25
Phytoestrogens or soy for vasomotor symptoms	25
The obesity paradox in older adults reflects biology rather than reverse causation	25
"Anti-inflammatory diet" describes a distinct mechanism	26
Cherries for gout	26
Protein restriction meaningfully slows CKD progression	26
Raising HDL-C is a useful treatment target	26
Very low sodium targets benefit normotensive people	26
FMT for indications other than C. difficile	27
Intestinal permeability is a real measurable phenomenon in defined diseases	27
Microbial diversity is a useful individual target	27
Prebiotic supplements improve clinical outcomes in healthy people	27
Probiotics for IBS, C. difficile prevention, and SIBO management	27
Artificial colours with sodium benzoate increase hyperactivity in children	28
Histamine intolerance as a defined diagnosable entity	28
Non-coeliac gluten sensitivity is a gluten-mediated entity	28
Higher protein delivery improves outcomes in critical illness	29
Immunonutrition in surgical patients	29
Nutritional screening programmes change outcomes (rather than identifying risk)	29
Menu calorie labelling changes consumer ordering	30
Nutri-Score captures the health value of a food adequately	30
Reading detailed nutrition panels changes what people eat	30
Above a fairly low floor, dietary variety is a nutritional requirement rather than a value	31
Cooking-skills interventions substantially improve diet quality	31
Meal planning improves diet quality and reduces overweight prevalence	31
"Food deserts" — access alone — explain most of the diet gap	32
Cognitive load from scarcity explains low-income food decision-making	32
Fast food is cheaper than cooking, per meal at the point of purchase	32
Acute stress increases food intake	33
Everyone has reliable internal hunger and fullness cues to return to	33
Intuitive eating produces weight loss	33
Smaller plates and packages reduce intake	33
Dieting causes eating disorders	34
Medication is a primary treatment for anorexia nervosa	34
App-based questionnaire "personalization" is useful	35
Caffeine-metabolism genotype should determine caffeine intake	35
Continuous glucose monitoring benefits people without diabetes	35
Lower postprandial glucose improves health outcomes in people without diabetes	35
Buying local substantially reduces a diet's greenhouse gas footprint	36
Organic farming has lower environmental impact	36
Removing agricultural subsidies would substantially narrow the price gap between processed and whole foods	36
Restricting broadcast food marketing to children produces measurable population dietary change	36
Subsidising fruit and vegetables improves intake	36
Optimal macronutrient proportions can be specified for a general population	37
The ideal amount of dairy is known	37
Critical-appraisal and scientific-reasoning training improves claim evaluation	38
Education that transfers a FILTER outperforms education that transfers a LIST	38

🟠 Probably false — 67 claims

Claim	Ch
Experts keep reversing themselves, so nutrition can't be trusted	1
Breakfast is the most important meal; skipping it causes weight gain	2
Observational nutrition studies are worthless; only RCTs count	2
You can only absorb 20–30 g of protein per meal	3
Calories don't matter — it's insulin (carbohydrate-insulin model, strong form)	4
The scale is the best measure of whether your diet is working	4
Each pound of muscle burns ~50 extra kcal/day	5
Carbohydrate is essential — your brain runs on glucose	6
Ketosis is "fat-burning mode," so keto burns more body fat	6
Train in the "fat-burning zone" to lose fat	6
Carbohydrates make you fat	7
You must eat protein within 30 minutes of training	8
Coconut oil is a health food because of its MCTs	9
Dietary cholesterol raises blood cholesterol; limit eggs	9
Saturated fat has been exonerated; the diet-heart hypothesis was fraud	9
Paleo reflects what our ancestors evolved to eat	10
Fiber causes bloating, so we're not meant to eat it	11
Alcohol calories don't count — ethanol isn't stored as fat	12
Moderate drinking protects your heart (the J-curve)	12
Soil depletion means you need mineral supplements	14
By the time you're thirsty you're already dehydrated	15
Coffee and tea dehydrate you and don't count toward fluid intake	15
"It's natural, so it's safe"	16
Microwaving destroys nutrients	17
...and thereby single-handedly redirected nutrition science	18
HFCS is uniquely harmful compared with sucrose	18
Sugar is metabolically like alcohol and should be regulated as such	18
Refined seed oils are unstable and should not be heated	19
Seed oils are toxic and a primary driver of modern chronic disease	19
The omega-6:omega-3 ratio drives chronic inflammation and disease	19
Conventional produce residues pose a meaningful health risk at measured levels	20
The Dirty Dozen is a useful guide to what to buy organic	20
Fasting "resets" or "boosts" metabolism	21
IF is metabolically superior to calorie restriction at matched intake	21
Our ancestors fasted deliberately, so we should	21
All Group 4 foods are equally harmful	22
BCAAs and glutamine benefit athletes eating adequate total protein	23
Low-carbohydrate diets improve moderate-to-high-intensity endurance performance	23
Recreational athletes need elite fuelling protocols	23
The 30-60 minute post-exercise "anabolic window"	23
"You only need to cut 50 calories a day"	24
The Biggest Loser findings generalize to typical weight loss	24
Weight loss is futile and should not be attempted	24
"Eating for two" in pregnancy	25
Menopause causes the weight gain (as opposed to the fat redistribution)	25
Dietary modification substitutes for urate-lowering therapy	26
Probiotics improve "general gut health", immunity, weight or mood in healthy people	27
Probiotics shorten acute gastroenteritis in children	27
Gluten-free replacement products are nutritionally equivalent to conventional versions	28
Improving hospital food alone fixes hospital malnutrition	29
Restrictive therapeutic diets in poorly-eating inpatients are appropriate	29
Meal planning advice as usually written transfers to resource-constrained households	31
Households on low incomes eat as they do primarily because they lack cooking knowledge	32
Nutrition education alone substantially improves the diets of food-insecure households	32
Applying restriction or "eat less" to cue-driven eating helps	33
Habits form in about 21 days	33
Willpower is the useful unit of analysis for eating behaviour	33
Recovery means never thinking about food again	34
ACTN3 genotype should be used to prescribe training or diet type	35
Broad direct-to-consumer micronutrient panels in asymptomatic people are useful	35
Consumer microbiome reports are a sound basis for dietary advice	35
MTHFR variants require special folate supplementation in people eating adequate folate	35
"Individual choices don't matter, only systems do"	36
Consumer education campaigns alone meaningfully change population diet	36
Packaging is a major driver of food's environmental footprint	36
Nutrition science knows essentially nothing because advice keeps changing	37
A general-audience book can meaningfully change population dietary intake	38

❌ Not supported — 69 claims

Claim	Ch
Don't combine protein and starch in the same meal	3
Your body accumulates toxins; you need a cleanse/detox	3
You have to count calories to lose weight	4
Six small meals a day stoke your metabolism	5
Skipping breakfast slows your metabolism	5
Exogenous ketone supplements make you burn body fat	6
L-carnitine increases fat burning / aids weight loss	6
Whole grains are no better than white bread (phytates, lectins)	7
High protein damages your kidneys (healthy people)	8
High protein leaches calcium from bones	8
Plant proteins are incomplete — you must combine at each meal	8
A DNA test can identify your best diet	10
Fiber is just roughage that scrubs your intestines	11
Red wine is protective because of resveratrol	12
High-dose antioxidant vitamins prevent cancer and heart disease	13
Alkaline / structured / oxygenated water	15
You need eight 8-ounce glasses of water a day	15
Proprietary blends deliver clinically-studied ingredients	16
Celery juice heals chronic illness	17
Detoxes and cleanses remove accumulated toxins (category restatement of Ch 3)	17
Negative-calorie foods exist	17
Superfoods have special health properties	17
Supplements "boost" the immune system	17
The alkaline diet changes blood pH and prevents disease	17
The blood type diet matches food to ABO type	17
Sugar feeds cancer	18
Hexane residue in refined oil is a meaningful health risk	19
"Clean eating" is a definable healthy dietary approach	20
Organic food is grown without pesticides	20
Fasting cures or prevents cancer	21
"Processing" per se is the problem	22
Restrictive diets, detoxes or fasting in pregnancy	25
Diet treats or cures cancer (sugar-starvation, alkaline, fasting, juicing)	26
Lp(a) responds meaningfully to diet	26
"Leaky gut syndrome" as a diagnosis for non-specific symptoms	27
Avoiding nuts, seeds and popcorn in diverticular disease	27
Consumer microbiome tests guide diet or assess health risk	27
Hair analysis, applied kinesiology, VEGA and cytotoxic testing diagnose food reactions	28
IgG food antibody panels diagnose food intolerance	28
MSG causes a characteristic syndrome	28
Skin signs are always present in anaphylaxis	28
Artificial nutrition at end of life prolongs life or improves comfort	29
PEG feeding in advanced dementia improves survival, aspiration, ulcers or comfort	29
"May contain" indicates a meaningful, risk-graded level of allergen	30
"Natural", "artisan", "wholesome" and "clean" have regulated definitions	30
"No added sugar" means a product is low in sugar	30
"Reduced" and "light" indicate a low level rather than a comparison	30
A declaration of "0 g" always means none is present	30
Buying in bulk is a matter of planning rather than available capital	32
Cravings signal a specific nutrient deficiency (e.g. chocolate for magnesium)	33
Amenorrhoea in an athlete is a normal sign of training hard enough	34
Eating disorders are a phase people grow out of	34
Eating disorders are about vanity or attention-seeking	34
Eating disorders only affect young women	34
Families, and mothers particularly, cause eating disorders	34
Someone must want to recover, or hit rock bottom, before treatment can work	34
You can identify who has an eating disorder by looking at them	34
Blood-type diets work through blood type	35
Direct-to-consumer genetic diet tests give reproducible advice from the same sample	35
Genotype testing can identify who should eat low-carbohydrate versus low-fat	35
Hair mineral analysis is valid for nutritional assessment	35
Insulin secretion predicts which diet an individual will do better on	35
Live blood analysis is valid	35
A single environmental ranking of foods is method-independent	36
Reformulation of an ultra-processed product addresses the finding of Ch 22	36
A single trial, mechanism paper or confident book should overturn a triangulated dietary finding	37
The science is settled and the details are known	37
A book can be written so as to be safe for every reader	38
Nutrition knowledge alone reliably produces dietary behaviour change	38

⚗️ Untested — 11 claims

Claim	Ch
Fasting activates autophagy, slowing ageing	6
Humans thrive on an all-meat carnivore diet	10
16 hours of fasting triggers autophagy and thereby prevents disease	21
Fasting extends human lifespan	21
Emulsifiers cause disease in humans at dietary exposures	22
Ketone esters improve endurance performance	23
A pre-decided fallback meal prevents abandonment of a plan after one failure	31
Batching components rather than complete meals improves adherence to a plan	31
Palatability decay is the main reason batch cooking is abandoned	31
Orthorexia is a distinct disorder	34
Eating to a personalized glucose curve improves health outcomes in well people	35

Part 2 — By chapter

⚠️ Use this half when you want to revisit what a particular chapter concluded, or to check a chapter's confidence against its argument.

Chapter 1 — 3 claims

🟡 2 · 🟠 1

Claim	Verdict
Nutrition science is bought and paid for by industry	🟡 Unclear / it depends
Experts keep reversing themselves, so nutrition can't be trusted	🟠 Probably false
There's a study for everything, so you can prove anything about food	🟡 Unclear / it depends

Chapter 2 — 3 claims

🟡 1 · 🟠 2

Claim	Verdict
Breakfast is the most important meal; skipping it causes weight gain	🟠 Probably false
Processed meat is Group 1 — same as smoking, no safe amount	🟡 Unclear / it depends
Observational nutrition studies are worthless; only RCTs count	🟠 Probably false

Chapter 3 — 5 claims

🟡 2 · 🟠 1 · ❌ 2

Claim	Verdict
You can only absorb 20–30 g of protein per meal	🟠 Probably false
Don't combine protein and starch in the same meal	❌ Not supported
Your body accumulates toxins; you need a cleanse/detox	❌ Not supported
"Leaky gut syndrome" causes chronic illness; this protocol fixes it	🟡 Unclear / it depends
95% of your serotonin is in your gut, so gut health = mental health	🟡 Unclear / it depends

Chapter 4 — 4 claims

🟡 1 · 🟠 2 · ❌ 1

Claim	Verdict
Calories don't matter — it's insulin (carbohydrate-insulin model, strong form)	🟠 Probably false
A calorie is a calorie	🟡 Unclear / it depends
You have to count calories to lose weight	❌ Not supported
The scale is the best measure of whether your diet is working	🟠 Probably false

Chapter 5 — 4 claims

🟡 1 · 🟠 1 · ❌ 2

Claim	Verdict
Each pound of muscle burns ~50 extra kcal/day	🟠 Probably false
Six small meals a day stoke your metabolism	❌ Not supported
Green tea / capsaicin / caffeine / cold exposure usefully boost metabolism	🟡 Unclear / it depends
Skipping breakfast slows your metabolism	❌ Not supported

Chapter 6 — 6 claims

🟠 3 · ❌ 2 · ⚗ 1

Claim	Verdict
L-carnitine increases fat burning / aids weight loss	❌ Not supported
Train in the "fat-burning zone" to lose fat	🟠 Probably false
Carbohydrate is essential — your brain runs on glucose	🟠 Probably false
Ketosis is "fat-burning mode," so keto burns more body fat	🟠 Probably false
Fasting activates autophagy, slowing ageing	⚗️ Untested (in humans)
Exogenous ketone supplements make you burn body fat	❌ Not supported

Chapter 7 — 5 claims

🟡 3 · 🟠 1 · ❌ 1

Claim	Verdict
Glycemic index tells you which carbohydrates are healthy	🟡 Unclear / it depends
Blood sugar "spikes" harm metabolically healthy people	🟡 Unclear / it depends
Carbohydrates make you fat	🟠 Probably false
Whole grains are no better than white bread (phytates, lectins)	❌ Not supported
Erythritol causes heart attacks and strokes	🟡 Unclear / it depends

Chapter 8 — 5 claims

🟡 1 · 🟠 1 · ❌ 3

Claim	Verdict
Plant proteins are incomplete — you must combine at each meal	❌ Not supported
You must eat protein within 30 minutes of training	🟠 Probably false
High protein damages your kidneys (healthy people)	❌ Not supported
High protein leaches calcium from bones	❌ Not supported
You need a protein supplement to hit your target	🟡 Unclear / it depends

Chapter 9 — 5 claims

🟡 2 · 🟠 3

Claim	Verdict
Saturated fat causes heart disease; limit to <10% of calories	🟡 Unclear / it depends
Saturated fat has been exonerated; the diet-heart hypothesis was fraud	🟠 Probably false
Dietary cholesterol raises blood cholesterol; limit eggs	🟠 Probably false
Fish oil supplements prevent heart disease	🟡 Unclear / it depends
Coconut oil is a health food because of its MCTs	🟠 Probably false

Chapter 10 — 3 claims

🟠 1 · ❌ 1 · ⚗ 1

Claim	Verdict
Paleo reflects what our ancestors evolved to eat	🟠 Probably false
Humans thrive on an all-meat carnivore diet	⚗️ Untested
A DNA test can identify your best diet	❌ Not supported

Chapter 11 — 3 claims

🟢 1 · 🟠 1 · ❌ 1

Claim	Verdict
Fiber is just roughage that scrubs your intestines	❌ Not supported
A fiber supplement replaces high-fiber foods	🟡 Unclear / it depends (psyllium 🟢 for LDL/glycemia/stool)
Fiber causes bloating, so we're not meant to eat it	🟠 Probably false

Chapter 12 — 3 claims

🟠 2 · ❌ 1

Claim	Verdict
Alcohol calories don't count — ethanol isn't stored as fat	🟠 Probably false
Moderate drinking protects your heart (the J-curve)	🟠 Probably false
Red wine is protective because of resveratrol	❌ Not supported

Chapter 13 — 7 claims

✅ 3 · 🟢 1 · 🟡 2 · ❌ 1

Claim	Verdict
Folic acid before conception prevents neural tube defects	✅ Well supported
Vegans need B12 supplementation	✅ Well supported
Vitamin D supplementation in the genuinely deficient	✅ Well supported
High-dose antioxidant vitamins prevent cancer and heart disease	❌ Not supported
Vitamin C prevents colds	❌ Not supported (🟡 in extreme physical stress)
Almost everyone is vitamin D deficient and should supplement	🟡 Unclear / it depends
A daily multivitamin is good insurance	🟠 Probably false (🟢 with genuinely poor intake)

Chapter 14 — 7 claims

✅ 2 · 🟢 1 · 🟡 3 · 🟠 1

Claim	Verdict
Iron supplementation helps the iron-deficient, including without anemia	✅ Well supported
Iodine adequacy in pregnancy matters for neurodevelopment	✅ Well supported
Calcium supplements prevent fractures in healthy adults	🟠 Probably false (🟢 frail/institutionalized + vit D)
Almost everyone is magnesium deficient; supplements fix sleep/anxiety/cramps	🟡 Unclear / it depends
Zinc lozenges shorten colds	🟡 Unclear / it depends
Everyone should reduce sodium below 2,300 mg	🟡 Unclear / it depends (live disagreement)
Soil depletion means you need mineral supplements	🟠 Probably false (real only for selenium/iodine)

Chapter 15 — 7 claims

✅ 1 · 🟢 1 · 🟡 1 · 🟠 2 · ❌ 2

Claim	Verdict
Oral rehydration solution for diarrhoeal illness	✅ Well supported
Electrolyte replacement beyond ~60–90 min hard exercise / heat / salty sweaters	🟢 Probably true
You need eight 8-ounce glasses of water a day	❌ Not supported
Alkaline / structured / oxygenated water	❌ Not supported
By the time you're thirsty you're already dehydrated	🟠 Probably false
Coffee and tea dehydrate you and don't count toward fluid intake	🟠 Probably false
Mild dehydration significantly impairs cognition	🟡 Unclear / it depends

Chapter 16 — 4 claims

✅ 1 · 🟢 1 · 🟠 1 · ❌ 1

Claim	Verdict
Creatine monohydrate improves strength, power and lean mass	✅ Well supported
Caffeine improves exercise performance	🟢 Probably true
Proprietary blends deliver clinically-studied ingredients	❌ Not supported
"It's natural, so it's safe"	🟠 Probably false

Chapter 17 — 14 claims

✅ 1 · 🟢 3 · 🟡 2 · 🟠 1 · ❌ 7

Claim	Verdict
Superfoods have special health properties	❌ Not supported
The alkaline diet changes blood pH and prevents disease	❌ Not supported
Celery juice heals chronic illness	❌ Not supported
Supplements "boost" the immune system	❌ Not supported
Negative-calorie foods exist	❌ Not supported
The blood type diet matches food to ABO type	❌ Not supported
Detoxes and cleanses remove accumulated toxins (category restatement of Ch 3)	❌ Not supported
Microwaving destroys nutrients	🟠 Probably false
Non-stick cookware leaches toxic chemicals into food	🟡 Unclear / it depends
Plastic containers leach hormone-disrupting chemicals	🟡 Unclear / it depends
Cooking tomatoes increases lycopene bioavailability	✅ Well supported
Ginger helps with nausea	🟢 Probably true
Rinsing rice (and cooking in excess water) is worth doing	🟢 Probably true
Chicken soup helps when you have a cold	🟢 Probably true

Chapter 18 — 14 claims

✅ 3 · 🟢 2 · 🟡 5 · 🟠 3 · ❌ 1

Claim	Verdict
Sugar-sweetened beverages contribute to weight gain and type 2 diabetes	✅ Well supported
The added-vs-natural sugar distinction is meaningful	🟢 Probably true
Sugar contributes to obesity (via energy intake)	🟢 Probably true
HFCS is uniquely harmful compared with sucrose	🟠 Probably false
Sugar is metabolically like alcohol and should be regulated as such	🟠 Probably false
Fructose has harmful effects independent of calories	🟡 Unclear / it depends
Eating sugar directly causes type 2 diabetes	🟡 Unclear / it depends
Sugar is addictive	🟡 Unclear / it depends
Non-sugar sweeteners are harmful / are a free substitution	🟡 Unclear / it depends
Sugar feeds cancer	❌ Not supported
The sugar industry funded and shaped an undisclosed 1967 NEJM review	✅ Well supported (documented)
...and thereby single-handedly redirected nutrition science	🟠 Probably false
Soda taxes reduce purchases of taxed drinks	✅ Well supported
Soda taxes improve health outcomes	🟡 Unclear / not yet established

Chapter 19 — 12 claims

✅ 3 · 🟢 3 · 🟡 2 · 🟠 3 · ❌ 1

Claim	Verdict
Modern linoleic acid intake is historically unprecedented (adipose tissue LA rose over 50 yrs)	✅ Well supported (documented)
Linoleic acid is essential; its absence causes a named deficiency syndrome	✅ Well supported
Replacing saturated fat with PUFA lowers LDL-C	✅ Well supported
Repeatedly heated commercial frying oil produces harmful oxidation products	🟢 Probably true
Replacing saturated fat with PUFA reduces cardiovascular events	🟢 Probably true (moderate certainty)
Two portions of oily fish per week (preformed EPA/DHA)	🟢 Probably true (as food)
The recovered trials (Sydney, Minnesota) refute the diet-heart hypothesis	🟡 Unclear / it depends
Refining loses antioxidants and generates small amounts of trans fat	🟡 Unclear / it depends
The omega-6:omega-3 ratio drives chronic inflammation and disease	🟠 Probably false
Seed oils are toxic and a primary driver of modern chronic disease	🟠 Probably false
Refined seed oils are unstable and should not be heated	🟠 Probably false (smoke point; stability caveat)
Hexane residue in refined oil is a meaningful health risk	❌ Not supported

Chapter 20 — 12 claims

✅ 3 · 🟢 2 · 🟡 3 · 🟠 2 · ❌ 2

Claim	Verdict
Eating more fruit and vegetables is beneficial	✅ Well supported
Approved GM foods currently on the market are safe to eat	✅ Well supported (NASEM 2016 consensus)
Basic food hygiene reduces foodborne illness (handwashing, separation, temperature, refrigeration)	✅ Well supported
Organic produce has lower pesticide residues	🟢 Probably true (trivially)
Organic milk and meat have higher omega-3 proportions	🟢 Probably true (small absolute amounts)
Organic food is more nutritious	🟡 Unclear / it depends
Organic farming is better for the environment	🟡 Unclear / it depends (yield gap is the crux)
Glyphosate causes cancer	🟡 Unclear / it depends (IARC hazard vs regulatory risk; occupational vs dietary)
Conventional produce residues pose a meaningful health risk at measured levels	🟠 Probably false
The Dirty Dozen is a useful guide to what to buy organic	🟠 Probably false
Organic food is grown without pesticides	❌ Not supported
"Clean eating" is a definable healthy dietary approach	❌ Not supported

Chapter 21 — 14 claims

✅ 1 · 🟢 4 · 🟡 3 · 🟠 3 · ❌ 1 · ⚗ 2

Claim	Verdict
Certain groups should not fast without medical supervision	✅ Well supported
IF helps some people eat less (adherence case)	🟢 Probably true
IF produces weight loss comparable to continuous energy restriction	🟢 Probably true
Hunger is entrained to habitual meal times and comes in waves	🟢 Probably true
Late-night eating is worse for glycemic control than the same food earlier	🟢 Probably true
Meal timing has metabolic effects independent of calories	🟡 Unclear / it depends (early windows more promising)
IF causes greater lean mass loss than continuous restriction	🟡 Unclear / it depends
Fasting reverses type 2 diabetes	🟡 Unclear / it depends (weight loss is the agent, not the timing)
IF is metabolically superior to calorie restriction at matched intake	🟠 Probably false
Fasting "resets" or "boosts" metabolism	🟠 Probably false
Our ancestors fasted deliberately, so we should	🟠 Probably false
Fasting cures or prevents cancer	❌ Not supported
16 hours of fasting triggers autophagy and thereby prevents disease	⚗️ Untested (no human dose-response)
Fasting extends human lifespan	⚗️ Untested (no human data)

Chapter 22 — 10 claims

✅ 1 · 🟢 3 · 🟡 3 · 🟠 1 · ❌ 1 · ⚗ 1

Claim	Verdict
Ultra-processed diets cause greater ad libitum energy intake than nutrient-matched minimally-processed diets	✅ Well supported (Hall ward trial)
Higher UPF intake is associated with adverse health outcomes	🟢 Probably true
Energy density and eating rate mediate the intake effect	🟢 Probably true
Ultra-processed food is cheaper per calorie than minimally processed food	🟢 Probably true
UPF harms health independently of energy intake and nutrient profile	🟡 Unclear / it depends
Hyper-palatability and weakened satiety signalling mediate the effect	🟡 Unclear / it depends
The NOVA classification is reliable and reproducible between coders	🟡 Unclear / it depends
All Group 4 foods are equally harmful	🟠 Probably false
Emulsifiers cause disease in humans at dietary exposures	⚗️ Untested (preliminary human microbiome data only)
"Processing" per se is the problem	❌ Not supported

Chapter 23 — 18 claims

✅ 6 · 🟢 5 · 🟡 2 · 🟠 4 · ⚗ 1

Claim	Verdict
Sustained low energy availability impairs health and performance (RED-S)	✅ Well supported
Carbohydrate availability determines sustained high-intensity endurance performance	✅ Well supported
Carbohydrate intake during prolonged exercise improves performance	✅ Well supported
Programmed over-drinking during endurance events is dangerous	✅ Well supported
Third-party certification is necessary for drug-tested athletes (strict liability)	✅ Well supported
Fat adaptation (raised fat oxidation) occurs with sustained low-carbohydrate diets	✅ Well supported
Multiple transportable carbohydrates permit higher oxidation rates and reduce GI distress	🟢 Probably true
Caffeine, nitrate, beta-alanine and sodium bicarbonate improve performance in defined contexts	🟢 Probably true
Protein distribution across the day (~0.3 g/kg, 3-5 meals) beats the same total in fewer meals	🟢 Probably true
Drink-to-thirst is adequate for most athletes in most conditions	🟢 Probably true
The gut is trainable for carbohydrate tolerance during exercise	🟢 Probably true
Carbohydrate periodization ("train low") improves competitive performance	🟡 Unclear / it depends
Low-carbohydrate diets improve ultra-endurance performance at low relative intensity	🟡 Unclear / it depends
The 30-60 minute post-exercise "anabolic window"	🟠 Probably false
Low-carbohydrate diets improve moderate-to-high-intensity endurance performance	🟠 Probably false
Recreational athletes need elite fuelling protocols	🟠 Probably false
BCAAs and glutamine benefit athletes eating adequate total protein	🟠 Probably false
Ketone esters improve endurance performance	⚗️ Untested

Chapter 24 — 21 claims

✅ 11 · 🟢 3 · 🟡 4 · 🟠 3

Claim	Verdict
Weight loss requires an energy deficit	✅ Well supported
Most people who lose weight regain most of it	✅ Well supported
Adaptive thermogenesis occurs after weight loss	✅ Well supported
Appetite adaptation after weight loss persists (ghrelin up, leptin/PYY/CCK down, at 12 months)	✅ Well supported
Modest weight loss produces disproportionate health benefit (DPP: ~7% loss, ~58% risk reduction)	✅ Well supported
GLP-1 receptor agonists produce substantial weight loss	✅ Well supported
Bariatric surgery produces durable loss and improved long-term outcomes	✅ Well supported
BMI heritability is approximately 40-70%	✅ Well supported
The population-level rise in body weight is environmental	✅ Well supported
Several common medication classes cause substantial weight gain	✅ Well supported
Higher protein and resistance training preserve lean mass during a deficit	✅ Well supported
Semaglutide reduces major adverse cardiovascular events in a defined high-risk group (SELECT)	🟢 Probably true
Physiological adaptation rather than willpower explains most weight regain	🟢 Probably true
Weight stigma worsens the outcomes it purports to address	🟢 Probably true
Which model of weight regulation is correct (set point / settling point / dual intervention)	🟡 Unclear / it depends
Weight-neutral approaches are the right clinical target	🟡 Unclear / it depends
Weight cycling is harmful	🟡 Unclear / it depends
GLP-1 receptor agonists are safe and effective over decades	🟡 Unclear / not yet established
Weight loss is futile and should not be attempted	🟠 Probably false
The Biggest Loser findings generalize to typical weight loss	🟠 Probably false
"You only need to cut 50 calories a day"	🟠 Probably false

Chapter 25 — 25 claims

✅ 13 · 🟢 5 · 🟡 4 · 🟠 2 · ❌ 1

Claim	Verdict
Folic acid preconception reduces neural tube defects	✅ Well supported
Iodine and iron requirements rise substantially in pregnancy	✅ Well supported
Early allergen introduction reduces peanut allergy (LEAP)	✅ Well supported
Breastfeeding reduces GI and respiratory infections in infancy	✅ Well supported
Formula is a safe and adequate alternative supporting normal growth	✅ Well supported
Older adults require more protein per kg than the general adult RDA	✅ Well supported
Resistance training is the most effective intervention against sarcopenia	✅ Well supported
B12 deficiency is common in older adults	✅ Well supported
Unintended weight loss in older adults is a red flag requiring investigation	✅ Well supported
Peak bone mass is largely accrued by the late teens to early twenties	✅ Well supported
Lactation energy requirements exceed any trimester of pregnancy	✅ Well supported
Recurrent chest infections in an older person warrant a swallowing assessment	✅ Well supported
Vitamin D supplementation for breastfed infants	✅ Well supported
Oily fish twice weekly in pregnancy (with species advice)	🟢 Probably true
Pressure and restriction in child feeding backfire (in opposite directions)	🟢 Probably true
Parental weight talk is associated with disordered eating and weight gain in adolescents	🟢 Probably true
Repeated neutral exposure increases acceptance of new foods in children	🟢 Probably true
Calcium plus vitamin D in frail and institutionalized older adults	🟢 Probably true
Baby-led weaning versus purees	🟡 Unclear / it depends
Phytoestrogens or soy for vasomotor symptoms	🟡 Unclear / it depends
Breastfeeding improves long-term outcomes (IQ, obesity, chronic disease)	🟡 Unclear / heavily confounded
The obesity paradox in older adults reflects biology rather than reverse causation	🟡 Unclear / it depends
"Eating for two" in pregnancy	🟠 Probably false
Menopause causes the weight gain (as opposed to the fat redistribution)	🟠 Probably false
Restrictive diets, detoxes or fasting in pregnancy	❌ Not supported

Chapter 26 — 20 claims

✅ 8 · 🟢 4 · 🟡 5 · 🟠 1 · ❌ 2

Claim	Verdict
LDL-C causally contributes to atherosclerotic cardiovascular disease	✅ Well supported
Type 2 diabetes can go into remission through weight loss (DiRECT), dose-dependent	✅ Well supported
The DASH dietary pattern lowers blood pressure	✅ Well supported
Reducing sodium lowers blood pressure, dose-dependently	✅ Well supported
Potassium-enriched salt substitutes reduce stroke and cardiovascular events (SSaSS)	✅ Well supported
Weight loss improves fatty liver histology, dose-dependently	✅ Well supported
Alcohol is a Group 1 carcinogen and excess adiposity raises risk of 12+ cancers	✅ Well supported
Physical activity improves insulin sensitivity independently of weight	✅ Well supported
ApoB is a better risk marker than LDL-C where the two are discordant	🟢 Probably true
Population-level sodium reduction reduces cardiovascular events	🟢 Probably true
Coffee is associated with less liver disease progression	🟢 Probably true
The Mediterranean pattern reduces cardiovascular events in high-risk people (PREDIMED)	🟢 Probably true
Raising HDL-C is a useful treatment target	🟡 Unclear / it depends
Very low sodium targets benefit normotensive people	🟡 Unclear / it depends
Protein restriction meaningfully slows CKD progression	🟡 Unclear / it depends
"Anti-inflammatory diet" describes a distinct mechanism	🟡 Unclear / it depends
Cherries for gout	🟡 Unclear / it depends
Dietary modification substitutes for urate-lowering therapy	🟠 Probably false
Lp(a) responds meaningfully to diet	❌ Not supported
Diet treats or cures cancer (sugar-starvation, alkaline, fasting, juicing)	❌ Not supported

Chapter 27 — 20 claims

✅ 6 · 🟢 4 · 🟡 5 · 🟠 2 · ❌ 3

Claim	Verdict
Butyrate is the primary energy source for colonocytes; fermentable fibre is the substrate	✅ Well supported
FMT cures recurrent Clostridioides difficile infection	✅ Well supported
Exclusive enteral nutrition induces remission in paediatric Crohn's disease	✅ Well supported
A low-FODMAP diet reduces symptoms in a substantial proportion of people with IBS	✅ Well supported
Specific probiotic strains prevent necrotizing enterocolitis in preterm infants	✅ Well supported
Human milk oligosaccharides are indigestible by the infant and feed bifidobacteria	✅ Well supported
Fermented foods increase microbiota diversity and reduce inflammatory markers	🟢 Probably true
S. boulardii or L. rhamnosus GG reduce antibiotic-associated diarrhoea	🟢 Probably true
Prebiotics change microbiota composition as described	🟢 Probably true
Peppermint oil, gut-directed hypnotherapy and CBT improve IBS symptoms	🟢 Probably true
Microbial diversity is a useful individual target	🟡 Unclear / it depends
Prebiotic supplements improve clinical outcomes in healthy people	🟡 Unclear / it depends
Probiotics for IBS, C. difficile prevention, and SIBO management	🟡 Unclear / it depends
FMT for indications other than C. difficile	🟡 Unclear / it depends
Intestinal permeability is a real measurable phenomenon in defined diseases	🟡 Unclear (direction of causation unestablished)
Probiotics shorten acute gastroenteritis in children	🟠 Probably false
Probiotics improve "general gut health", immunity, weight or mood in healthy people	🟠 Probably false
Consumer microbiome tests guide diet or assess health risk	❌ Not supported
"Leaky gut syndrome" as a diagnosis for non-specific symptoms	❌ Not supported
Avoiding nuts, seeds and popcorn in diverticular disease	❌ Not supported

Chapter 28 — 20 claims

✅ 9 · 🟢 3 · 🟡 3 · 🟠 1 · ❌ 4

Claim	Verdict
Intramuscular adrenaline is first-line for anaphylaxis; delay is associated with death	✅ Well supported
Antihistamines do not treat anaphylaxis	✅ Well supported
Skin prick and specific IgE tests measure sensitization, not allergy	✅ Well supported
Coeliac disease is autoimmune and requires strict lifelong gluten avoidance	✅ Well supported
Coeliac testing is only valid while the person is eating gluten	✅ Well supported
Lactase non-persistence is the global norm and is dose-dependent	✅ Well supported
Early allergen introduction reduces the risk of developing food allergy	✅ Well supported
Oral immunotherapy achieves desensitization, not cure	✅ Well supported
Milk, egg, wheat and soy allergies are often outgrown; peanut, tree nut, fish and shellfish usually persist	✅ Well supported
Oral allergy syndrome proteins are heat-labile, so cooked forms are usually tolerated	🟢 Probably true
Sulphites can provoke bronchoconstriction in people with asthma	🟢 Probably true
Cofactors (exercise, alcohol, NSAIDs, infection) can convert a tolerated exposure into anaphylaxis	🟢 Probably true
Artificial colours with sodium benzoate increase hyperactivity in children	🟡 Unclear / it depends
Non-coeliac gluten sensitivity is a gluten-mediated entity	🟡 Unclear / it depends
Histamine intolerance as a defined diagnosable entity	🟡 Unclear / it depends
Gluten-free replacement products are nutritionally equivalent to conventional versions	🟠 Probably false
IgG food antibody panels diagnose food intolerance	❌ Not supported
Hair analysis, applied kinesiology, VEGA and cytotoxic testing diagnose food reactions	❌ Not supported
MSG causes a characteristic syndrome	❌ Not supported
Skin signs are always present in anaphylaxis	❌ Not supported

Chapter 29 — 20 claims

✅ 9 · 🟢 4 · 🟡 3 · 🟠 2 · ❌ 2

Claim	Verdict
Hospital malnutrition is common and associated with worse outcomes	✅ Well supported
Refeeding a starved person too rapidly causes electrolyte collapse and death	✅ Well supported
Nasogastric tube position must be confirmed before use (pH and/or radiography)	✅ Well supported
Prolonged preoperative fasting is unnecessary (clear fluids ~2h, light meal ~6h)	✅ Well supported
Preoperative nutritional optimization in malnourished patients reduces complications	✅ Well supported
Early full-dose feeding in critical illness is not superior to starting low and advancing	✅ Well supported
Bed rest causes rapid measurable loss of muscle mass and function	✅ Well supported
Warfarin requires consistent rather than minimal vitamin K intake	✅ Well supported
Home enteral nutrition supports independent living	✅ Well supported
Oral nutritional supplements benefit malnourished older adults	🟢 Probably true
Early postoperative oral intake is safe after most abdominal surgery	🟢 Probably true
Preoperative carbohydrate loading (ERAS)	🟢 Probably true
Prehabilitation before major surgery improves recovery	🟢 Probably true
Higher protein delivery improves outcomes in critical illness	🟡 Unclear / contested
Immunonutrition in surgical patients	🟡 Unclear / it depends
Nutritional screening programmes change outcomes (rather than identifying risk)	🟡 Unclear / it depends
Restrictive therapeutic diets in poorly-eating inpatients are appropriate	🟠 Probably false
Improving hospital food alone fixes hospital malnutrition	🟠 Probably false
PEG feeding in advanced dementia improves survival, aspiration, ulcers or comfort	❌ Not supported
Artificial nutrition at end of life prolongs life or improves comfort	❌ Not supported

Chapter 30 — 18 claims

✅ 8 · 🟢 2 · 🟡 3 · ❌ 5

Claim	Verdict
Ingredients are listed in descending order by weight	✅ Well supported (legal requirement)
QUID: a named, pictured or emphasized ingredient must have its percentage declared (EU/UK)	✅ Well supported (legal requirement)
Declared allergens are mandatory, emphasized and enforceable	✅ Well supported
"Gluten-free" is legally defined at 20 ppm or below	✅ Well supported
"Use by" is a safety date; "best before" is a quality date	✅ Well supported
Food may be frozen up to the use-by date, resetting the clock	✅ Well supported
kcal per 100 g is energy density and is mandatory on European labels	✅ Well supported
Declared nutrition values may be calculated rather than measured, with tolerances around 20%	✅ Well supported
Front-of-pack warning labels change purchasing behaviour	🟢 Probably true
Serving sizes are used to make per-serving figures look smaller	🟢 Probably true
Nutri-Score captures the health value of a food adequately	🟡 Unclear / it depends (does not capture processing)
Reading detailed nutrition panels changes what people eat	🟡 Unclear / weaker than for front-of-pack warnings
Menu calorie labelling changes consumer ordering	🟡 Unclear (supply-side effects may be larger)
"May contain" indicates a meaningful, risk-graded level of allergen	❌ Not supported
"Natural", "artisan", "wholesome" and "clean" have regulated definitions	❌ Not supported
"No added sugar" means a product is low in sugar	❌ Not supported
"Reduced" and "light" indicate a low level rather than a comparison	❌ Not supported
A declaration of "0 g" always means none is present	❌ Not supported

Chapter 31 — 12 claims

✅ 2 · 🟢 3 · 🟡 3 · 🟠 1 · ⚗ 3

Claim	Verdict
Household food waste is concentrated in fresh produce, bread and prepared leftovers, and is driven by over-buying, poor storage and date-label confusion	✅ Well supported (WRAP household waste reporting)
What is available in the home is a major determinant of what gets eaten	✅ Well supported (Ch 22's food-environment evidence, applied)
Shopping without a list produces more unplanned purchases, skewed toward Group 4 foods	🟢 Probably true
Online grocery ordering reduces impulse purchasing by removing the aisle	🟢 Probably true
Long-term diet-trial dropout follows discrete failures rather than gradual decay	🟢 Probably true (Ch 10 §10.4)
Meal planning improves diet quality and reduces overweight prevalence	🟡 Unclear / observational and confounded (NutriNet-Santé; healthy-user structure)
Cooking-skills interventions substantially improve diet quality	🟡 Unclear (skill and confidence move more than intake)
Above a fairly low floor, dietary variety is a nutritional requirement rather than a value	🟡 Unclear / it depends (Part III, Ch 27)
Palatability decay is the main reason batch cooking is abandoned	⚗️ Untested (plausible, widely reported, not formally studied)
Batching components rather than complete meals improves adherence to a plan	⚗️ Untested (this chapter's central practical recommendation)
A pre-decided fallback meal prevents abandonment of a plan after one failure	⚗️ Untested (mechanism from Ch 10; the recommendation I would defend hardest)
Meal planning advice as usually written transfers to resource-constrained households	🟠 Probably false (§31.14; Ch 32)

Chapter 32 — 15 claims

✅ 5 · 🟢 4 · 🟡 3 · 🟠 2 · ❌ 1

Claim	Verdict
Food insecurity is associated with poorer diet quality, worse mental health, and higher rates of obesity and type 2 diabetes	✅ Well supported
Food insecurity and obesity co-occur rather than being opposites	✅ Well supported (mechanism not settled)
Higher prices reduce consumption of the costlier item, more so in lower-income households	✅ Well supported (Ch 18 SSB taxes)
The cheapest foods per calorie and the cheapest per nutrient are different foods	✅ Well supported (Drewnowski energy-cost work)
Low-income households pay more for equivalent goods (the poverty premium)	✅ Well supported (consumer and utility research)
Food purchasing and intake vary systematically across the pay/benefit cycle	🟢 Probably true
Cost is a major barrier to diet quality independent of nutritional knowledge	🟢 Probably true
Income supplementation and cash transfers improve food security and diet quality	🟢 Probably true (feared misallocation effects smaller than predicted)
Waste as a percentage of food purchased is lower, not higher, in lower-income households	🟢 Probably true (WRAP)
"Food deserts" — access alone — explain most of the diet gap	🟡 Unclear / probably not; new supermarkets changed purchasing less than predicted
Fast food is cheaper than cooking, per meal at the point of purchase	🟡 Unclear / it depends (🟠 per calorie; 🟡 once time, fuel, equipment, skill and storage are counted)
Cognitive load from scarcity explains low-income food decision-making	🟡 Unclear (real, but not a substitute for the resource account — §32.11)
Nutrition education alone substantially improves the diets of food-insecure households	🟠 Probably false
Households on low incomes eat as they do primarily because they lack cooking knowledge	🟠 Probably false (§32.11 the competence assumption)
Buying in bulk is a matter of planning rather than available capital	❌ Not supported

Chapter 33 — 25 claims

✅ 8 · 🟢 9 · 🟡 4 · 🟠 3 · ❌ 1

Claim	Verdict
Homeostatic and hedonic feeding are dissociable systems with different substrates and different levers	✅ Well supported (THE THRESHOLD)
Restriction increases the salience of and preoccupation with food	✅ Well supported (Minnesota Starvation Experiment, Keys 1944–45)
"Wanting" (incentive salience) and "liking" (hedonic pleasure) are separable and can diverge	✅ Well supported (Berridge/Robinson)
What is available in the immediate environment substantially determines what is eaten	✅ Well supported (Ch 22, independent route)
Sensory-specific satiety increases intake when variety is available	✅ Well supported
Restricting a specific palatable food in children increases wanting and unsupervised consumption	✅ Well supported (Birch)
Pressuring a child to eat a food reduces liking for it, persistently	✅ Well supported (Birch)
People adjust their intake toward that of their eating companions without noticing (modelling)	✅ Well supported
Meals eaten with others are larger and longer than meals eaten alone	🟢 Probably true (social facilitation; mechanism largely duration)
Short sleep increases intake, especially of highly palatable food	🟢 Probably true
Alcohol acutely stimulates appetite and impairs restraint, in addition to supplying unregistered energy	🟢 Probably true
Implementation intentions ("when X, I will Y") improve follow-through	🟢 Probably true
Restricting a specific food increases craving for that specific food	🟢 Probably true (the argument against forbidden-food lists)
Eating slowly and without distraction reduces intake at that meal	🟢 Probably true (real, modest)
Intuitive eating is associated with better psychological outcomes and less disordered eating	🟢 Probably true (largely cross-sectional)
Mindful eating reduces binge-eating episodes	🟢 Probably true
Chronic stress and low mood increase intake of highly palatable food specifically	🟢 Probably true
Acute stress increases food intake	🟡 Unclear / it depends (increases in some, decreases in others, fairly reliably by individual)
Everyone has reliable internal hunger and fullness cues to return to	🟡 Unclear / it depends (§33.10's substantive doubt)
Intuitive eating produces weight loss	🟡 Unclear (its proponents mostly do not claim this)
Smaller plates and packages reduce intake	🟡 Unclear / real but oversold; poor replication record
Habits form in about 21 days	🟠 Probably false (median nearer two to three months, hugely variable)
Willpower is the useful unit of analysis for eating behaviour	🟠 Probably false (§33.12; argument made independent of ego depletion)
Applying restriction or "eat less" to cue-driven eating helps	🟠 Probably false — counterproductive via §33.6
Cravings signal a specific nutrient deficiency (e.g. chocolate for magnesium)	❌ Not supported (sodium appetite is the one clear exception)

Chapter 34 — 27 claims

✅ 8 · 🟢 8 · 🟡 2 · 🟠 1 · ❌ 7 · ⚗ 1

Claim	Verdict
Eating disorders are psychiatric illnesses with substantial heritability and among the highest mortality of any psychiatric category	✅ Well supported
Most people with eating disorders are not underweight	✅ Well supported (THE chapter's key fact)
Medical complications of restriction follow from the behaviour and the rate of change, not from a position on a weight chart	✅ Well supported
Refeeding syndrome is a real, potentially fatal risk requiring medical monitoring	✅ Well supported (Ch 29 §29.7)
Duration of untreated illness predicts outcome and is a modifiable factor	✅ Well supported
Malnutrition itself produces rigidity, preoccupation, low mood and social withdrawal that resemble personality	✅ Well supported (Minnesota; Ch 33 §33.6)
Eating disorders occur across sexes, ages, ethnicities, body sizes and socioeconomic groups	✅ Well supported
Autistic people are over-represented in anorexia nervosa and ARFID	✅ Well supported
Family-based treatment is the best-supported treatment for adolescents with anorexia nervosa	🟢 Probably true
CBT-E is the best-supported approach for adults across eating disorder diagnoses	🟢 Probably true (adult anorexia evidence base is thin)
Early nutritional restoration improves psychological symptoms independent of therapy	🟢 Probably true
Food insecurity is a risk factor for binge eating and eating-disorder symptoms	🟢 Probably true
LGBTQ+ people, and gay and bisexual men particularly, have elevated rates	🟢 Probably true
Men, older adults, people in larger bodies and minoritized groups experience longer diagnostic delay	🟢 Probably true (consistent across studies)
Pro-eating-disorder online content is harmful to those exposed	🟢 Probably true
Insulin omission as a weight-control behaviour occurs in type 1 diabetes and carries serious risk	🟢 Probably true
Dieting causes eating disorders	🟡 Unclear / it depends — a real risk factor, not a cause; most dieters do not develop one
Medication is a primary treatment for anorexia nervosa	🟡 Unclear / adjunctive at best; has a clearer role in bulimia nervosa and BED
Orthorexia is a distinct disorder	⚗️ Untested as a distinct entity (names something real; measurement contested; often meets OSFED/ARFID)
Recovery means never thinking about food again	🟠 Probably false — and setting it as the standard makes real recovery look like failure
Eating disorders are a phase people grow out of	❌ Not supported — and the belief causes delay
Eating disorders are about vanity or attention-seeking	❌ Not supported
Eating disorders only affect young women	❌ Not supported
You can identify who has an eating disorder by looking at them	❌ Not supported — the mechanism by which people are told they are not ill
Families, and mothers particularly, cause eating disorders	❌ Not supported — and the belief harmed families for decades
Someone must want to recover, or hit rock bottom, before treatment can work	❌ Not supported — ambivalence is a feature; the belief justifies waiting
Amenorrhoea in an athlete is a normal sign of training hard enough	❌ Not supported — it is a symptom

Chapter 35 — 24 claims

✅ 5 · 🟢 4 · 🟡 4 · 🟠 4 · ❌ 6 · ⚗ 1

Claim	Verdict
Individual postprandial glucose responses to identical foods differ substantially between people	✅ Well supported (Zeevi/Segal 2015; replicated)
Those responses are reasonably stable within a person and partly predictable from measurable features	✅ Well supported
The glycaemic index is a population average that conceals large individual variation	✅ Well supported
Lactase persistence and alcohol-flushing genotypes are real and clinically meaningful	✅ Well supported (and self-evident without a test)
Personalization based on diagnosed allergy, coeliac disease, intolerance, disease, medication interaction, life stage, athletic demand or documented deficiency is evidence-based	✅ Well supported
An algorithm can design a diet that lowers an individual's postprandial glucose	🟢 Probably true (demonstrated short-term)
Matching a diet to preference, culture, budget and schedule is the largest available personalization lever	🟢 Probably true (Ch 10 adherence)
Personalized-nutrition outputs converge substantially on conventional dietary advice	🟢 Probably true
Well-designed n-of-1 experiments can answer short-latency, high-signal individual questions	🟢 Probably true
Lower postprandial glucose improves health outcomes in people without diabetes	🟡 Unclear — THE UNSUPPORTED LINK the sector rests on
Continuous glucose monitoring benefits people without diabetes	🟡 Unclear (educational value short-term; no outcome evidence; real interpretation risk)
Caffeine-metabolism genotype should determine caffeine intake	🟡 Unclear (variant real; acting on it changing outcomes is thin)
App-based questionnaire "personalization" is useful	🟡 Unclear — what it personalizes is preference, which is legitimate
MTHFR variants require special folate supplementation in people eating adequate folate	🟠 Probably false
ACTN3 genotype should be used to prescribe training or diet type	🟠 Probably false (population association, negligible individual predictive power)
Consumer microbiome reports are a sound basis for dietary advice	🟠 Probably false (no reference standard; method-dependent; advice is generic)
Broad direct-to-consumer micronutrient panels in asymptomatic people are useful	🟠 Probably false (Ch 13)
Genotype testing can identify who should eat low-carbohydrate versus low-fat	❌ Not supported (DIETFITS pre-specified this and it failed)
Insulin secretion predicts which diet an individual will do better on	❌ Not supported (DIETFITS)
Blood-type diets work through blood type	❌ Not supported (tested directly; benefit tracked dietary quality, not blood type)
Hair mineral analysis is valid for nutritional assessment	❌ Not supported
Live blood analysis is valid	❌ Not supported
Direct-to-consumer genetic diet tests give reproducible advice from the same sample	❌ Not supported
Eating to a personalized glucose curve improves health outcomes in well people	⚗️ Untested (the claim the sector is sold on)

Chapter 36 — 25 claims

✅ 8 · 🟢 7 · 🟡 5 · 🟠 3 · ❌ 2

Claim	Verdict
Ruminant meat has substantially higher environmental impact per unit protein or calorie than plant sources	✅ Well supported (Poore & Nemecek; ordering robust to method)
Within-category producer variation is large, but the highest-impact plant producers still beat the lowest-impact beef producers	✅ Well supported
Food waste is a major driver of food's environmental impact	✅ Well supported
Roughly a third of food produced is never eaten, upstream in lower-income and downstream in higher-income countries	✅ Well supported (UNEP/FAO)
Food marketing influences children's food preferences, requests and consumption	✅ Well supported (not seriously contested)
Eliminating industrial trans fat reduces population intake and cardiovascular risk	✅ Well supported
Sugar-sweetened beverage taxes reduce purchased sugar, substantially via manufacturer reformulation	✅ Well supported (Ch 18)
The environmentally optimal and nutritionally optimal dietary patterns converge substantially	✅ Well supported (THE SECOND CONVERGENCE)
Interventions that change the default outperform interventions requiring a decision	🟢 Probably true (matches Ch 33 §33.4 from an independent direction)
Mandatory reformulation targets reduce population intake of the targeted nutrient	🟢 Probably true (salt programmes)
Front-of-pack warning labels change purchasing	🟢 Probably true (Ch 30 §30.9)
School food standards and public procurement standards improve intake in the populations reached	🟢 Probably true
Air-freighted produce is a genuine high-emission exception	🟢 Probably true
Ultra-processed food is abundant primarily because it solves supply-side business problems	🟢 Probably true (§36.5's supply-side account)
Retail listing and shelf position are substantially determined by purchased placement rather than demand alone	🟢 Probably true
Removing agricultural subsidies would substantially narrow the price gap between processed and whole foods	🟡 Unclear / probably modest — the raw commodity is a small fraction of retail price
Buying local substantially reduces a diet's greenhouse gas footprint	🟡 Unclear / usually overstated — transport is a small share; production dominates
Organic farming has lower environmental impact	🟡 Unclear / it depends on the unit (lower per hectare, often higher per unit of food)
Restricting broadcast food marketing to children produces measurable population dietary change	🟡 Unclear (modest effects; displacement to digital channels)
Subsidising fruit and vegetables improves intake	🟡 Unclear / promising and less tested
Packaging is a major driver of food's environmental footprint	🟠 Probably false as usually framed (small share; waste-preventing packaging often nets positive)
Consumer education campaigns alone meaningfully change population diet	🟠 Probably false — weakest tier, and helps most those who needed it least
"Individual choices don't matter, only systems do"	🟠 Probably false as stated — half right; aggregate demand is what the system responds to
Reformulation of an ultra-processed product addresses the finding of Ch 22	❌ Not supported — a reformulated Group 4 product is still Group 4
A single environmental ranking of foods is method-independent	❌ Not supported — per kg, per calorie and per gram protein rank foods differently

Chapter 37 — 18 claims

✅ 7 · 🟢 6 · 🟡 2 · 🟠 1 · ❌ 2

Claim	Verdict
A plant-forward, minimally-processed dietary pattern with adequate protein is supported across independent lines of evidence	✅ Well supported (THE THRESHOLD's substantive content)
Adherence dominates composition in free-living dietary outcomes	✅ Well supported (Ch 10; load-bearing for this chapter)
Sugar-sweetened beverages are the clearest single dietary target	✅ Well supported (Ch 18)
Most supplements provide no benefit to people without a documented deficiency	✅ Well supported (Ch 13, 16)
There is no protective dose of alcohol	✅ Well supported (Ch 12)
Population dietary guidance across countries agrees substantially on the core pattern	✅ Well supported
Recommendations built on dietary patterns are more robust to mechanistic error than recommendations built on mechanisms	🟢 Probably true (§37.1's fourth reason; argued, not tested as a set)
The individual-personalization and planetary-environmental analyses are at least partly independent	🟢 Probably true (§37.2 concedes routes 1 and 3 may share an epidemiological ancestor)
Claims that ADJUST the established dietary pattern are more often correct than claims that REPLACE it	🟢 Probably true (§37.2b, historical rather than experimental)
Novelty predicts evidential weakness in nutrition claims	🟢 Probably true (structural argument from incentives)
Refuted nutrition claims recur because refutation is distributed differently from the claim	🟢 Probably true (§37.6b)
Only six categories of dietary requirement are genuinely non-optional for a healthy adult	🟢 Probably true (§37.4)
Optimal macronutrient proportions can be specified for a general population	🟡 Unclear — the plate is given as ranges for this reason
The ideal amount of dairy is known	🟡 Unclear (Ch 14, 25)
A boring, stable recommendation is therefore a certain one	🟠 Probably false — §37.11; over 400 verdicts and the 🟡 count grew faster than the ✅ count
Nutrition science knows essentially nothing because advice keeps changing	🟠 Probably false (one of §37.11's two comfortable extremes)
The science is settled and the details are known	❌ Not supported (the other comfortable extreme)
A single trial, mechanism paper or confident book should overturn a triangulated dietary finding	❌ Not supported (§37.8's deliberate asymmetry)

Chapter 38 — 10 claims

✅ 1 · 🟢 4 · 🟡 2 · 🟠 1 · ❌ 2

Claim	Verdict
Intention predicts behaviour much less well than people expect, and situation explains more of the gap than motivation	✅ Well supported (intention–behaviour gap literature)
Implementation intentions bridge the intention–behaviour gap better than exhortation	🟢 Probably true (Ch 33 §33.8)
People who appear to have more self-control report exercising it less often	🟢 Probably true (Wood; Ch 33 §33.12)
Interpreting a lapse as a personal failing predicts abandonment; interpreting it as predictable does not	🟢 Probably true (Ch 33 §33.6 step 5)
Media-literacy and inoculation interventions produce real short-term effects	🟢 Probably true (with faster decay than practitioners would like)
Critical-appraisal and scientific-reasoning training improves claim evaluation	🟡 Unclear — encouraging trials, small literature, transfer not established
Education that transfers a FILTER outperforms education that transfers a LIST	🟡 Unclear — THE CHAPTER'S CENTRAL CLAIM, held at 🟡 and flagged as convenient to the author
A general-audience book can meaningfully change population dietary intake	🟠 Probably false (Ch 36 §36.7; this book sits at the bottom of the ladder)
A book can be written so as to be safe for every reader	❌ Not supported — §38.1's concession; no framing removes the vector risk
Nutrition knowledge alone reliably produces dietary behaviour change	❌ Not supported (the chapter's opening problem)

⚠️ A note on what is NOT in here

⚠️ This index contains claims the book ASSESSED. It does not contain:

⚠️ The reasoning — which is the actual content, and is in the chapters. ⚠️ The practical material in Part VI, most of which Chapter 31 §31.16 conceded is craft rather than science and was therefore not given verdicts. ⚠️ The threshold concepts, which are not claims but reframings — they are listed in Chapter 38 §38.5b. ⚠️ Anything about individuals. A verdict is a population statement; Chapter 37 §37.5 lists the eight things that genuinely differ between people, and none of them is looked up here.

⚠️ And a claim's absence from this index is not a verdict. It means the book did not assess it — which for a field this size is true of far more claims than are listed above.

Where to go next

Chapter 2	⚠️ The evidence ladder underneath every verdict here
Chapter 37 §37.7	⚠️ The six-question check — for claims NOT in this index
Chapter 38 §38.5	⚠️ What this distribution means
Chapter 38 §38.8	⚠️ What to do when one of these turns out wrong
Appendix D	⚠️ The six questions as a worksheet
Appendix H	⚠️ Supplements specifically, graded

⚠️ This is the only object in this book designed to still be useful in 2035.

⚠️ Its use is a lookup, and it takes ten seconds.