Section 06
Everything the rest of this site cites lives here: 39 peer-reviewed studies, official reports, and position papers. Each entry links out to the source and back to every article that uses it, so you can audit any claim in both directions.
A meta-analysis covering about 38,700 farms in 119 countries. The largest dataset comparing the environmental footprint of foods; the source of the finding that meat and dairy provide 18% of calories while using 83% of farmland.
Models global food-system emissions and attributes 57% of them to animal-based foods versus 29% for plant-based foods.
The standard global accounting of water footprints for animal products: roughly 15,400 litres per kilogram of beef, most of it water embedded in feed.
Quantifies the calories and protein lost when crops are routed through animals: replacing animal foods with plant equivalents could feed an additional 350 million people in the US alone.
The WHO cancer agency's classification of processed meat as a Group 1 carcinogen and red meat as Group 2A. The classification grades strength of evidence, not size of risk; the absolute risk increase is modest.
A meta-analysis of 49 resistance-training trials (1,863 adults). Extra protein increased muscle and strength gains up to about 1.62 g/kg/day, beyond which no further benefit appeared, setting a practical ceiling for how much protein lifters need.
A 12-week trial: 19 habitual vegans (using soy protein) and 19 omnivores did matched resistance training at about 1.6 g/kg/day protein. Both gained the same muscle and strength, so an exclusively plant-based diet was not different from a mixed diet for building muscle when protein was matched.
A comparison of 27 vegan and 27 omnivore untrained women after a resistance workout. The vegans were significantly less sore 48 hours later at several muscle sites and showed higher post-exercise grip strength. A small, single-session study, so suggestive rather than conclusive.
In nearly 3,000 mostly meat-eating adults, about 39% had low-normal B12. Supplements and fortified cereal and milk protected B12 status; meat intake did not. Evidence that eating meat does not guarantee good B12 and that fortified foods and supplements are the reliable source.
Among healthy young Germans, vegans who supplemented B12 (about 90% of them, median 250 mcg/day) had B12 status comparable to omnivores, while lacto-ovo-vegetarians fared worse. Supplementing vegans match meat-eaters on B12.
Follows about 209,000 adults and finds plant-based diets built on whole foods are linked to lower coronary heart disease risk, while plant-based diets heavy in refined foods are not. Quality matters, not just the label.
Dose-response meta-analysis across food groups: higher intakes of whole grains, vegetables, fruit, nuts, and legumes track with lower all-cause mortality; higher red and processed meat intake tracks with higher mortality.
The Global Burden of Disease analysis of diet and death worldwide. The biggest dietary killers it identifies are diets low in whole grains, fruits, nuts, and vegetables, alongside high sodium.
EPIC-Oxford cohort of about 48,000 people: vegetarians and vegans had lower ischaemic heart disease risk than meat eaters, alongside a smaller increase in haemorrhagic stroke. An honest picture with findings in both directions.
Cohort of about 73,000 Seventh-day Adventists: vegetarian dietary patterns were associated with 12% lower all-cause mortality over roughly six years of follow-up.
Commissioned by the WHO: people eating the most fiber (mostly from whole plant foods) had 15 to 30% lower all-cause mortality and lower incidence of heart disease, stroke, type 2 diabetes, and colorectal cancer.
A review of the scientific study of animal sentience: where the evidence stands for mammals, birds, and fish, and how the field has moved from whether animals feel to what they feel.
Raised 1957, 1978, and 2005 broiler chicken breeds under identical conditions: the 2005 bird grew to over four times the weight in the same time. The clearest documentation of what selective breeding has done to the animals themselves.
The standard estimate of how many individual wild fish are caught each year: roughly 1.1 to 2.2 trillion, an order of magnitude more individuals than all farmed land animals combined.
Pooled 17 randomized double-blind placebo-controlled trials. Soy intake lowered systolic blood pressure by about 1.7 mmHg and diastolic by about 1.3 mmHg on average, with larger effects in younger adults.
Re-analyzed the 46 trials the FDA used to weigh its soy heart-health claim. A median 25 g of soy protein a day lowered LDL, the harmful cholesterol, by about 3 to 4%, the basis for the claim that soy protein may reduce heart-disease risk.
Pooled 41 clinical studies covering more than 1,700 men and found that neither soy foods nor isoflavones changed testosterone or estrogen levels at any dose. The clearest evidence against the myth that soy feminizes men.
Across 12 studies of more than 37,000 women with breast cancer, soy and isoflavone intake was not harmful and was associated with lower recurrence, with a small survival benefit for pre-diagnosis intake. Supports that soy is safe for breast-cancer survivors.
A comprehensive review of the human evidence on soy. Concludes that soy foods do not harm thyroid function in people with adequate iodine, do not raise breast-cancer risk, and do not affect male hormones, while modestly helping cholesterol.
A randomized crossover trial in 600 adults with milk intolerance. Milk with only A2 beta-casein caused significantly less bloating, pain, and loose stool than ordinary A1/A2 milk, and the benefit held even in people who could digest lactose, pointing to dairy protein, not just lactose, as a trigger.
Switching to an animal-based diet changed the gut bacteria within a single day: it increased bile-tolerant microbes including Bilophila wadsworthia, which is linked to inflammation and inflammatory bowel disease, and reduced the Firmicutes that ferment plant fiber. A plant-based diet shifted the community the other way.
The original potential renal acid load (PRAL) framework. Protein-rich animal foods like meat, fish, hard cheese, and eggs are acid-forming in the body, while fruits and vegetables are alkaline-forming. The basis for the idea that an animal-heavy diet raises the body's dietary acid load.
About half of people starting a daily serving of beans reported more gas in the first week, but it was temporary: 70% or more of those who noticed it felt it fade by the second or third week as their gut adjusted. Evidence that fiber bloat is an adjustment phase, not a permanent verdict.
Pooled 14 trials in 906 people with IBS. Fiber helped overall, but the benefit was limited to soluble fiber such as psyllium; insoluble wheat bran did not help and can worsen bloating. The reason the type of fiber matters when you start adding more plants.
A randomized crossover trial in 30 people with IBS. Cutting fermentable carbs (FODMAPs) roughly halved gut-symptom scores compared with a typical diet. The evidence base for using a low-FODMAP elimination-and-reintroduction approach to isolate which foods are triggering symptoms.
A systematic review of reflux lifestyle advice. Weight loss, raising the head of the bed, and leaving 2 to 3 hours between the last meal and lying down had real supporting evidence. Common trigger foods like chocolate, coffee, and fat relax the esophageal valve in lab studies, but the proof that avoiding them improves symptoms is weaker.
In 184 people with throat-reflux symptoms, a 90 to 95% plant-based Mediterranean diet with alkaline water matched proton pump inhibitor drugs: 62.6% improved on the diet versus 54.1% on the medication. Among the strongest evidence that a plant-forward diet can rival reflux medication.
The IPCC's assessment of land use and climate, including the finding that dietary change toward plant-based foods is a major mitigation opportunity.
The most-cited estimate of AI water use. Its abstract reports that training GPT-3 in US data centers directly evaporated about 700,000 litres of freshwater; the paper also estimates roughly 500 ml consumed per 10 to 50 medium-length responses. A preprint, so treat the numbers as estimates; we use it because public AI water data is scarce.
The largest US dietetics body's position: appropriately planned vegetarian and vegan diets are healthful, nutritionally adequate, and appropriate for all life stages, including pregnancy, infancy, and athletes.
The 300-study review that led the UK to legally recognize octopuses, crabs, and lobsters as sentient. A model of how sentience evidence is weighed.
The NIH reference on B12: requirements, sources, deficiency symptoms, and supplementation. The practical basis for the rule that every vegan should supplement B12.
The US government's own guideline for how claims like "free range" and "humanely raised" get onto packages: companies substantiate them with paperwork submitted to USDA, not through on-farm inspections. Reading it is the fastest way to see how little most labels guarantee.
The NIH reference on lactose intolerance: about 65% of people worldwide lose much of their ability to digest lactose after infancy, from roughly 5% of people of Northern European descent to 70 to 100% in East Asian populations.
