4,771 peer-reviewed studies, each with a short note written from its published abstract: what kind of study it was, who took part, what was measured and what the authors reported. Observational studies show associations, not proof.
People assigned to a diet, food or program and measured.
2021 / Randomized crossover trial
In a randomized crossover trial, healthy adults consumed 100 grams of bread paired with 250 milliliters of water, black tea, or lemon juice. Lemon juice reduced peak blood glucose concentration by 30% on average (p < 0.01) and delayed this peak by over 35 minutes (78 versus 41 minutes with water, p < 0.0001). Black tea had no effect on glycemic response, and neither beverage altered subsequent ad libitum energy intake.
Freitas D, Boué F, Benallaoua M, et al. Eur J Nutr, 2021. PMID 32201919
2020 / Randomized crossover trial
Fifty healthy adults completed a randomized crossover trial comparing four weeks of daily non-fried potato side dishes against refined grains. Fasting plasma glucose did not differ between treatments (-0.05 mmol/l, 95% CI -0.14 to 0.04, P = 0.15), and no differences emerged for cardiometabolic endpoints. Potato consumption resulted in higher Healthy Eating Index-2015 scores (difference 3.5, 95% CI 0.6 to 6.4), potassium (547 mg, 95% CI 331 to 764), and dietary fiber (2.4 g, 95% CI 0.6 to 4.2).
Johnston EA, Petersen KS, Kris-Etherton PM Br J Nutr, 2020. PMID 31964428
2020 / Controlled feeding studies
Volunteers consumed either a 60% kcal fat meal (n=98) or a 36% kcal fat meal (n=62, placebo group). At least 87.5% of plasma LPS measurements in 32 tested participants were below the assay sensitivity (0.002 EU/mL). Although polymyxin B suppressed IL-1β (P = 0.035) and IL-6 (P = 0.0487) production in 3 h postprandial blood after 24 h incubation in the 60% fat study, the LPS present did not increase cytokine concentrations above fasting levels. In vitro lipoprotein lipase treatment significantly increased FFA and cytokine concentrations (P < 0.05) in both studies.
Mo Z, Huang S, Burnett DJ, et al. J Nutr, 2020. PMID 32040591
2020 / Randomized controlled trial (secondary analysis)
In a secondary analysis of a factorial randomized trial involving 3,365 adults followed for 22.3 years, smoking was associated with increased risk of gastric cancer incidence (OR, 1.72; 95% CI, 1.003-2.93) and mortality (HR, 2.01; 95% CI, 1.01-3.98). Garlic supplementation was associated with reduced gastric cancer mortality exclusively among individuals who never drank alcohol (HR, 0.33; 95% CI, 0.15-0.75; interaction P = .03). No significant interactions occurred between vitamin supplementation and lifestyle behaviors.
Guo Y, Li ZX, Zhang JY, et al. JAMA Netw Open, 2020. PMID 32589229
2019 / Crossover trial
Researchers evaluated postprandial glucose and insulin responses to an oral glucose tolerance test (OGTT at 0800h vs 2000h, n=10) and a low GI meal consumed at 0800h, 2000h, and 0000h (n=9). In the OGTT, glucose iAUC was higher in the evening than morning (p=0.007). Following the low GI meal, glucose iAUC was higher at evening and midnight compared to morning (p=0.008, p=0.021), with no significant difference between evening and midnight (p=0.594). Insulin iAUC was also significantly higher at evening and midnight compared to morning (p=0.008 for both).
Leung GKW, Huggins CE, Bonham MP Clin Nutr, 2019. PMID 29248250
2019 / Three experimental studies (laboratory and field)
Three studies examined the effect of a justification cue (license condition) versus no cue (control condition) on caloric consumption. In Studies 1 (N = 85) and 2 (N = 95), participants could eat freely from unhealthy snacks, while in Study 3 (N = 110), participants chose a snack at a take-out lunch place. Across all studies, caloric consumption was higher in the license condition compared to the control condition, with participants eating more unhealthy snacks in Studies 1 and 2 and choosing higher-calorie snacks in Study 3.
Prinsen S, Evers C, de Ridder DTD Psychol Health, 2019. PMID 30422003
2019 / Controlled trial
In a controlled trial, 49 strict vegan participants took oral creatine monohydrate (5 g daily, n = 31) or placebo (n = 18) for three weeks. Creatine intake normalized homocysteine levels in all participants and increased blood folic acid concentrations. Supplementation also increased baseline skin capillary density and endothelium-dependent capillary recruitment, while enhancing microvascular vasodilation specifically in participants with elevated baseline homocysteine.
Van Bavel D, de Moraes R, Tibirica E Fundam Clin Pharmacol, 2019. PMID 30506745
2019 / Clinical trial
Researchers evaluated isotope-labeled l-carnitine and gamma-butyrobetaine metabolism in 40 omnivores and 32 vegans or vegetarians. Following oral d3-l-carnitine, omnivores generated over 20-fold more d3-TMAO than vegans or vegetarians (P = 0.001), despite comparable fasting plasma carnitine levels. Gut bacteria transformed carnitine into gamma-butyrobetaine in both dietary groups, but conversion of gamma-butyrobetaine to TMA was induced by chronic carnitine intake and omnivorous habits.
Koeth RA, Lam-Galvez BR, Kirsop J, et al. J Clin Invest, 2019. PMID 30530985
2019 / Randomized controlled trial
In an 8-week randomized trial of 40 adolescent boys with active nonalcoholic fatty liver disease, restricting free sugars to under 3% of daily calories reduced hepatic steatosis from 25% to 17%, compared with 21% to 20% in the usual diet group (adjusted mean difference -6.23%; 95% CI, -9.45% to -3.02%; P < .001). Alanine aminotransferase levels decreased more with the low-sugar diet (adjusted ratio of geometric means 0.65; 95% CI, 0.53 to 0.81; P < .001). Seven secondary outcomes were null.
Schwimmer JB, Ugalde-Nicalo P, Welsh JA, et al. JAMA, 2019. PMID 30667502
2019 / Cross-sectional comparison and short-term dietary intervention
Researchers studied dietary methionine restriction in mice and evaluated circulating FGF21 levels in humans following different diets. In humans, plasma FGF21 levels were markedly higher in vegans compared with omnivores, and circulating FGF21 increased significantly in omnivores after 4 days on a vegetarian diet. In NZO mice, a 9-week methionine-restricted diet elevated plasma FGF21, increased insulin sensitivity, raised energy expenditure, and prevented hyperglycemia without altering food intake or fat mass.
Castaño-Martinez T, Schumacher F, Schumacher S, et al. FASEB J, 2019. PMID 30841758
2019 / Randomized controlled crossover trial
Twenty-four overweight men completed a randomized crossover trial evaluating extruded sorghum versus extruded wheat during two 8-week periods with a 4-week washout. Sorghum intake reduced body fat percentage and raised daily intakes of carbohydrate and dietary fiber compared with wheat. Serum glutathione peroxidase concentration increased with sorghum, but this change did not differ significantly from wheat.
Anunciação PC, Cardoso LM, Alfenas RCG, et al. Food Res Int, 2019. PMID 30884705
2019 / Randomized crossover trial
A randomized crossover study in 30 healthy adults tested the glycemic and insulinemic responses to steel-cut oats, old fashioned large-flake oats, instant oats, Cheerios cereal, and Cream of Rice. Mean glucose peak rise was significantly lower after steel-cut oats (1.93 mmol/l) than after Cream of Rice (2.61 mmol/l), Cheerios (2.49 mmol/l), and instant oats (2.47 mmol/l). Steel-cut oats also produced significantly lower glucose incremental AUC than Cream of Rice and Cheerios, as well as lower insulin incremental AUC than instant oats.
Wolever TMS, Johnson J, Jenkins AL, et al. Br J Nutr, 2019. PMID 31068229
2019 / Inpatient randomized crossover trial
Twenty weight-stable adults were admitted to an inpatient clinical center and randomized to consume ad libitum ultra-processed or unprocessed diets for 2 weeks each in a crossover design, with meals matched for presented calories, energy density, macronutrients, sugar, sodium, and fiber. Energy intake was higher during the ultra-processed diet (508 ± 106 kcal/day; p = 0.0001), driven by increased carbohydrate (280 ± 54 kcal/day; p < 0.0001) and fat (230 ± 53 kcal/day; p = 0.0004), but not protein (-2 ± 12 kcal/day; p = 0.85). Participants gained 0.9 ± 0.3 kg (p = 0.009) during the ultra-processed diet and lost 0.9 ± 0.3 kg (p = 0.007) during the unprocessed diet.
Hall KD, Ayuketah A, Brychta R, et al. Cell Metab, 2019. PMID 31105044
2019 / Single-arm pilot intervention trial
Seventeen women aged 20 to 45 with Hashimoto's thyroiditis participated in a 10-week online health coaching program implementing the Autoimmune Protocol elimination diet. Among the 16 women who completed questionnaires, statistically significant improvements were observed across all eight SF-36 subscales, and the MSQ symptom burden decreased from an average of 92 (SD 25) to 29 (SD 20). High-sensitivity C-reactive protein (n = 14) decreased significantly by 29% (p = 0.0219), from 1.63 mg/L (SD 1.72) to 1.15 mg/L (SD 1.31). No statistically significant changes were observed in thyroid function tests or thyroid antibodies.
Abbott RD, Sadowski A, Alt AG Cureus, 2019. PMID 31275780
2019 / Randomized crossover trial
In a randomized crossover study, 24 healthy adult males consumed a high-fat breakfast with or without 50 g of freeze-dried mango pulp. At 1 hour post-meal, mango consumption resulted in significantly lower glucose (95.8 ± 4.4 mg/dL vs. 104.8 ± 5.4 mg/dL; P = .002) and higher HDL-C (58.4 ± 2.7 mg/dL vs. 55.2 ± 2.3 mg/dL; P = .01). However, no significant meal × time interactions were observed for triglycerides (P = .88), inflammatory markers (C-reactive protein, P = .17; interleukin-6, P = .30), or antioxidant enzymes (superoxide dismutase, P = .77; glutathione peroxidase, P = .36; catalase, P = .32).
O'Hara C, Ojo B, Emerson SR, et al. Nutr Metab Insights, 2019. PMID 31452602
2019 / Randomized crossover trial
In a randomized crossover trial of 30 females with elevated fasting glucose and insulin, consuming 250 g of baked then chilled potatoes reduced glucose at 15 and 30 minutes by 4.8% and 9.2%, insulin by 25.8% and 22.6%, and GIP by 41.1% and 37.6% compared to boiled potatoes. Insulin and GIP areas under the curve were significantly decreased. Glucose area under the curve, GLP-1, peptide YY, and satiety did not differ.
Patterson MA, Fong JN, Maiya M, et al. Nutrients, 2019. PMID 31484331
2019 / Randomized crossover trial
In a randomized crossover trial of 11 healthy subjects adhering to 2-week low-salt and high-salt diets, salt increased monocyte expression of the chemokine receptor CCR2. A high-salt diet also increased plasma MCP-1, transendothelial monocyte migration, and skin macrophage density. In addition, macrophages displayed an elevated proinflammatory phenotype, including increased in vitro cytokine secretion of IL-6, TNF, and IL-10, as well as increased skin macrophage HLA-DR expression.
Wenstedt EF, Verberk SG, Kroon J, et al. JCI Insight, 2019. PMID 31672939
2019 / Randomized controlled trial
In a randomized trial, 70 healthy Japanese adults with a BMI between 23 and 30 consumed 9 grams daily of either quercetin-rich onion powder or placebo for 12 weeks. Visceral fat area did not differ significantly between the groups overall, though it was lower with onion powder in participants with low HDL cholesterol. Alanine aminotransferase levels were also significantly reduced in the onion powder arm compared to placebo.
Nishimura M, Muro T, Kobori M, et al. Nutrients, 2019. PMID 31905615
2018 / Randomized crossover trial
In a randomized crossover trial with 13 healthy adults, adding two portions of cooked broccoli to mashed potato led to lower early postprandial blood glucose and insulin surges than potato alone. The broccoli-potato combination reduced peak glucose, incremental peak, and 30-minute incremental area under the curve for both markers. However, adding isolated broccoli fibre or cellulose did not alter post-meal curves compared with potato eaten by itself.
Ballance S, Knutsen SH, Fosvold ØW, et al. Eur J Nutr, 2018. PMID 27655525
2018 / Controlled trial
Fifteen healthy females were evaluated for mood, cognition, and whole-brain MR spectroscopic imaging before and after 72 hours of fasting, and compared with 15 age-matched non-fasting females. After fasting, mood scores significantly increased. Glx decreased across all nine brain regions examined, tCho decreased in eight, NAA in four, and tCr in one, with Glx showing the greatest change and the frontal lobes being the most affected brain region.
Ding XQ, Maudsley AA, Schweiger U, et al. J Cereb Blood Flow Metab, 2018. PMID 28273723
2018 / 2-year prospective randomized trial
In a 2-year randomized trial, 70 healthy postmenopausal overweight or obese women were assigned to a Paleolithic-type diet (PD, n=35) or a diet based on Nordic Nutrition Recommendations (NNR, n=35). At baseline, median 24-h urinary iodine concentration (24-UIC, 71.0 μg/l) and excretion (24-UIE, 134.0 μg/d) were similar between groups. After 6 months, 24-UIC fell to 36.0 μg/l (P=0.001) and 24-UIE dropped to 77.0 μg/d (P=0.001) in the PD group, while levels remained unaltered in the NNR group. Free thyroxin, TSH, and free triiodothyronine (FT3) were generally similar between groups, though FT3 was lower in the PD group at 6 months.
Manousou S, Stål M, Larsson C, et al. Eur J Clin Nutr, 2018. PMID 28901333
2018 / Randomized crossover trial
During a three-day randomized crossover trial in 21 healthy participants, an isocaloric vegan diet lowered triglycerides, insulin, bile acids, and insulin resistance assessed by HOMA-IR (p < 0.05). It also raised magnesium levels and altered branched-chain amino acid metabolism compared with an omnivorous regimen. Fiber and polyunsaturated fats correlated inversely with triglycerides, insulin, and HOMA-IR.
Draper CF, Vassallo I, Di Cara A, et al. Mol Nutr Food Res, 2018. PMID 29087622
2018 / Randomized crossover trial
A double-blind, randomized crossover trial evaluated 13 trained male runners completing a 1-mile race 60 minutes after consuming 0.09 g/kg caffeinated coffee, decaffeinated coffee, or placebo. Running time was 1.3% faster with coffee (04:35.37) than decaffeinated coffee (04:39.14; P = .018; 95% CI, -0.11 to -0.01) and 1.9% faster than placebo (04:41.00; P = .006; 95% CI, -0.15 to -0.03). Differences between decaffeinated coffee and placebo were trivial.
Clarke ND, Richardson DL, Thie J, et al. Int J Sports Physiol Perform, 2018. PMID 29140142
2018 / Secondary analysis of a randomized controlled trial
Thirty-four obese participants completed 12 months of alternate-day fasting and were categorized by achieving at least 5% weight loss or under 5% weight loss. The high-loss group experienced increased fullness (p = 0.04) and decreased hunger (p = 0.03). They also raised dietary protein from 15% to 20% of energy intake (p = 0.04) and adhered more closely to fast-day caloric targets.
Kroeger CM, Trepanowski JF, Klempel MC, et al. Nutr Health, 2018. PMID 29353535