Food and health — a field with unusually noisy evidence.
10 topics · 11 curated works
No prior grounding assumed.
Is Everything We Eat Associated with Cancer Risk?
Jonathan D. Schoenfeld & John P. A. Ioannidis · 2013
Picks fifty common ingredients from a cookbook and finds most have been linked to either raising or lowering cancer risk in some published study,…
+3 more at this level
Assumes you know the vocabulary.
Fats and Fatty Acids in Human Nutrition: Report of an Expert Consultation
Food and Agriculture Organization & World Health Organization · 2010
Reviews the evidence on dietary fat and disease risk and shifts the recommendation from a single total-fat ceiling to the balance between fat types,…
+2 more at this level
Primary sources and full treatments.
A Treatise of the Scurvy
James Lind · 1753
Reports a controlled comparison of six proposed scurvy treatments given to sailors in matched pairs, finding citrus fruit cured the disease — often…
+3 more at this level
11 works
Jonathan D. Schoenfeld & John P. A. Ioannidis
Picks fifty common ingredients from a cookbook and finds most have been linked to either raising or lowering cancer risk in some published study, most with weak evidence, arguing this pattern should make readers doubt any single nutrition headline.
World Health Organization
Sets criteria for nutrient-profiling and warning-label schemes on packaged food, arguing that simple, interpretive front-of-pack labels change purchasing behaviour more than a nutrition facts panel does.
link checked 17 Sept 2026U.S. Department of Agriculture & U.S. Department of Health and Human Services
Recommends whole dietary patterns rather than isolated nutrients as the unit that predicts health outcomes, and extends federal guidance to infants and toddlers for the first time.
link checked 17 Sept 2026Global Nutrition Report Independent Expert Group
Synthesises global data on all forms of malnutrition — undernutrition, micronutrient deficiency and obesity coexisting within the same countries and often the same households — and argues that inequity, not just resource scarcity, is what is driving continued poor nutrition outcomes worldwide.
link checked 17 Sept 2026Food and Agriculture Organization & World Health Organization
Reviews the evidence on dietary fat and disease risk and shifts the recommendation from a single total-fat ceiling to the balance between fat types, especially replacing saturated and trans fats with unsaturated ones.
link checked 17 Sept 2026Institute of Medicine
Reviews the evidence on front-of-pack labelling schemes and recommends a single, standardised interpretive system built around calories, saturated fat, sodium and added sugar, arguing that the proliferation of competing industry-designed labels was confusing rather than informing shoppers.
link checked 17 Sept 2026Carlos A. Monteiro et al.
Sets out the NOVA classification, which groups food by degree and purpose of industrial processing rather than by nutrient content, arguing processing itself predicts health outcomes independent of nutrients.
link checked 17 Sept 2026James Lind
Reports a controlled comparison of six proposed scurvy treatments given to sailors in matched pairs, finding citrus fruit cured the disease — often cited as an early controlled clinical trial.
link checked 17 Sept 2026Stanley J. Dudrick, Douglas W. Wilmore, Harry M. Vars & Jonathan E. Rhoads
Reports sustaining normal growth in beagle puppies, and then in an infant, entirely through nutrients infused into a central vein, establishing that the gut could be bypassed altogether when it cannot be used to feed a patient.
Kevin D. Hall et al.
Uses a metabolic ward trial to show that, at matched calorie deficits, cutting fat produces more body-fat loss than cutting carbohydrate — the opposite of what the carbohydrate-insulin model predicts.
link checked 17 Sept 2026John P. A. Ioannidis
Argues that most diet-disease associations from observational nutrition studies are too small and too confounded by self-reported data to support the dietary advice built on them.
This subject genuinely sits in more than one domain. These fields approach the same ground with different methods.