8,456 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.
2005 / Interventional crossover trial
Ten obese women completed two 2-week isocaloric diet periods containing either 16% (low-fat) or 56% (high-fat) of energy as fat. Baseline liver fat averaged 10 +/- 7%; it decreased by 20 +/- 9% during the low-fat diet and increased by 35 +/- 21% during the high-fat diet (P = 0.042 for change between diets). Fasting serum insulin decreased from 70 +/- 41 pmol/liter at baseline to 60 +/- 24 pmol/liter after the low-fat diet and increased to 81 +/- 44 pmol/liter after the high-fat diet (P = 0.005 for change between diets). Serum lipids, free fatty acids, and intraabdominal and subcutaneous fat masses remained unchanged.
Westerbacka J, Lammi K, Häkkinen AM, et al. J Clin Endocrinol Metab, 2005. PMID 15741262
2005 / Repeated-measures clinical studies
In two repeated-measures studies, adults consumed potato test meals after overnight fasting to evaluate glycemic response. In Study 1 (10 adults), precooked Russet potatoes elicited a lower glucose area under the curve than day-cooked potatoes (P < .05), whereas precooking had no effect on white potatoes. In Study 2 (12 adults), potato glycemic index varied significantly by variety and preparation (P = .003), ranging from 56 for cold boiled red potatoes to 89 for hot boiled red potatoes.
Fernandes G, Velangi A, Wolever TM J Am Diet Assoc, 2005. PMID 15800557
2005 / Randomized controlled trial
A randomized pilot trial assigned 242 newborn infants with HLA susceptibility and a first-degree relative with type 1 diabetes to receive casein hydrolysate or conventional cow's milk formula until 6 to 8 months of age. Over a mean 4.7-year follow-up, cumulative autoantibody emergence was somewhat lower with hydrolysed formula. Adjusted life-table analysis indicated significant protection against islet cell antibodies (P = 0.02) and at least one autoantibody (P = 0.03).
Akerblom HK, Virtanen SM, Ilonen J, et al. Diabetologia, 2005. PMID 15838685
2005 / Interventional trial with pre- and post-biopsies
Ten insulin-sensitive males were fed an isoenergetic high-fat diet (HFD) for 3 days, with muscle biopsies analyzed via oligonucleotide microarray. The HFD differentially regulated 297 genes, including decreasing six genes involved in oxidative phosphorylation (NDUFB3, NDUFB5, NDUFS1, NDUFV1, SDHB, and SLC25A12). Furthermore, PGC1alpha and PGC1beta mRNA levels decreased by -20% (P < 0.01) and -25% (P < 0.01), respectively. In mice fed a HFD for 3 weeks, similar OXPHOS and PGC1 mRNAs decreased by approximately 90%, and cytochrome C and PGC1alpha protein by approximately 40%.
Sparks LM, Xie H, Koza RA, et al. Diabetes, 2005. PMID 15983191
2005 / Randomized crossover trial
In a randomized crossover trial of 12 healthy adults, adding vinegar (18, 23, or 28 mmol acetic acid) to white wheat bread lowered postprandial blood glucose and insulin responses in an inverse dose-response pattern. Satiety ratings increased linearly with higher acetic acid doses. The highest vinegar dose significantly lowered 90-minute incremental areas for glycemic and insulinemic indices, though 120-minute values did not differ from reference bread.
Ostman E, Granfeldt Y, Persson L, et al. Eur J Clin Nutr, 2005. PMID 16015276
2005 / Randomized crossover trial
Thirteen healthy volunteers consumed four potato-based breakfast meals in random order to evaluate the effects of cold storage and vinegar addition. Chilling boiled potatoes for 24 hours increased resistant starch from 3.3% to 5.2% and lowered the insulin index by 28%. Adding vinaigrette to cold potatoes decreased the glycemic and insulinemic indices by 43% and 31%, respectively, relative to freshly boiled potatoes (glycemic index 96 vs 168; insulin index 128 vs 185).
Leeman M, Ostman E, Björck I Eur J Clin Nutr, 2005. PMID 16034360
2005 / Randomized controlled trial
In an outpatient setting, 64 overweight, postmenopausal women were randomly assigned to a low-fat, vegan diet or an NCEP-based control diet without energy limits for 14 weeks while maintaining exercise. The intervention group's body weight decreased by 5.8 +/- 3.2 kg compared with 3.8 +/- 2.8 kg in the control group (P = .012). An index of insulin sensitivity increased from 4.6 +/- 2.9 to 5.7 +/- 3.9 (P = .017) in the intervention group, but the between-group difference was not significant (P = .17).
Barnard ND, Scialli AR, Turner-McGrievy G, et al. Am J Med, 2005. PMID 16164885
2005 / Randomized crossover trial
Eleven healthy adults consumed a high-glycemic load meal or a lower-glycemic load meal alongside vinegar, peanut products, or control treatments in a crossover protocol. Vinegar and peanuts reduced 60-minute postprandial glucose responses by approximately 55%, reaching statistical significance for the high-glycemic load option. Later energy intake for the day dropped by roughly 200 to 275 kcal after the high-glycemic load meal, though this difference was not statistically significant (P = .111).
Johnston CS, Buller AJ J Am Diet Assoc, 2005. PMID 16321601
2004 / Randomized double-blind placebo-controlled trial
In a randomized double-blind placebo-controlled trial across Europe, Canada, and the USA, 552 relatives of individuals with type 1 diabetes received oral modified release nicotinamide (1.2 g/m2) or placebo for 5 years. Over the trial, 159 participants developed diabetes (82 taking nicotinamide and 77 on placebo). The unadjusted hazard ratio was 1.07 (95% CI 0.78-1.45; p=0.69), and the adjusted hazard ratio was 1.01 (95% CI 0.73-1.38; p=0.97). Nicotinamide also did not affect growth in children, first-phase insulin secretion, or rates of serious adverse events.
Gale EA, Bingley PJ, Emmett CL, et al. Lancet, 2004. PMID 15043959
2004 / Randomized crossover trial
In a randomized crossover trial of nine healthy lean women, participants consumed their normal diet either regularly across six daily meals or irregularly with three to nine daily meals for 14 days each. While fasting glucose and insulin were unchanged, peak insulin and insulin area under the curve after a test meal were higher with irregular eating (P < 0.01). Fasting total cholesterol (P < 0.01) and LDL cholesterol (P < 0.05) were also significantly higher following the irregular meal pattern.
Farshchi HR, Taylor MA, Macdonald IA Eur J Clin Nutr, 2004. PMID 15220950
2004 / Randomized crossover trial
Fourteen men with type 2 diabetes and nephropathy received 0.5 g/(kg*d) of protein as isolated soy protein or casein for 8 weeks each in a crossover study. Soy protein intake significantly decreased urinary albumin excretion by 9.5% (P < 0.0001), lowered total-to-HDL (-0.45) and LDL-to-HDL (-0.20) ratios, and raised HDL cholesterol by 4.3% (P = 0.0040). Albumin excretion correlated negatively with plasma isoflavone levels.
Teixeira SR, Tappenden KA, Carson L, et al. J Nutr, 2004. PMID 15284369
2004 / Randomized controlled trial
In a randomized trial of 180 adults with metabolic syndrome, 90 participants followed a Mediterranean-style diet and 90 followed a prudent diet for 2 years. The Mediterranean diet group experienced greater weight loss (-4.0 kg versus -1.2 kg; P < .001) and significant reductions in hs-CRP, IL-6, and insulin resistance. Endothelial function scores improved significantly, and fewer patients retained metabolic syndrome features at 2 years.
NANO counterpoint →
Esposito K, Marfella R, Ciotola M, et al. JAMA, 2004. PMID 15383514
2004 / Interventional crossover trial
Five obese and five normal-weight women were evaluated after receiving either a 100 g oral butter load or a 100 ml water load, followed 120 minutes later by a 180-minute hyperglycemic clamp. Insulin secretion rate increased significantly after butter versus water in controls (1305.6 [SE 124.1] vs 616.1 [SE 52.5] pmol/min; P<0.01) and in obese subjects (1975.0 [SE 44.1] vs 1417.5 [se 56.0] pmol/min; P<0.05). In controls, but not in obese participants, butter impaired insulin sensitivity (0.60 x 10(-2) [SE 0.11 x 10(-2)] vs 2.26 x 10(-2) [SE 0.32 x 10(-2)] ml/min per kg FFM per [pmol/l]; P<0.01) and reduced insulin clearance (0.89 [SE 0.22] vs 1.70 [SE 0.15] litres/min; P<0.01).
Manco M, Bertuzzi A, Salinari S, et al. Br J Nutr, 2004. PMID 15613251
2003 / Randomized crossover trial
Twelve overweight patients with type 2 diabetes consumed three test meals in random order: a control meal with 50 g carbohydrate, the control meal plus 100 g butter, or the control meal plus 80 g olive oil. No significant differences were seen in glucose, insulin, or fatty acid responses between the two fat-rich meals. However, plasma and chylomicron triacylglycerol responses were highest after the butter meal, HDL-cholesterol concentrations decreased significantly only after butter, and GLP-1 responses were highest after the olive oil meal.
Thomsen C, Storm H, Holst JJ, et al. Am J Clin Nutr, 2003. PMID 12600850
2003 / Clinical trial
Blood glucose was measured for 120 minutes following test meals in 58 healthy adults to evaluate glycemic index values. White rice correlated strongly with glucose area under the curve (r = 0.853, P < 0.0001). Combining the grain with acidic dressings, milk items, or legume dishes lowered the glycemic value by 20% to 40%, regardless of consumption order.
NANO counterpoint →
Sugiyama M, Tang AC, Wakaki Y, et al. Eur J Clin Nutr, 2003. PMID 12792658
2003 / Randomized controlled trial protocol
The Look AHEAD randomized controlled trial protocol was designed to assess the long-term effects (up to 11.5 years) of an intensive weight loss program delivered over 4 years in overweight and obese individuals with type 2 diabetes. Approximately 5,000 participants aged 45-74 years with BMI >=25 kg/m2 are randomized to an intensive lifestyle intervention (decreased caloric intake and increased physical activity) or a diabetes support and education control. The primary outcome is time to incidence of a major CVD event, powered to detect an 18% difference between groups.
Ryan DH, Espeland MA, Foster GD, et al. Control Clin Trials, 2003. PMID 14500058
2003 / Clinical trial
Researchers compared three isoenergetic meals (high-fat; high-carbohydrate, low-fiber; and high-carbohydrate, high-fiber with whole wheat) in 30 newly diagnosed type 2 diabetes patients and 30 matched controls. The high-fat meal increased IL-18 and decreased adiponectin (P < 0.05) in both cohorts. In contrast, the high-fiber meal decreased serum IL-18 (P < 0.05) in both groups, whereas IL-8 did not shift significantly after any meal.
Esposito K, Nappo F, Giugliano F, et al. Am J Clin Nutr, 2003. PMID 14668275
2003 / Controlled crossover study
Ten male subjects were studied under three conditions: hyperosmolality (fluid restriction and hypertonic NaCl), hypo-osmolality (desmopressin, liberal water drinking, and hypotonic saline), and iso-osmolality (ad libitum water intake). Plasma osmolality increased from 285+/-1 to 296+/-1 mosm/kg during hyperosmolality and decreased from 286+/-1 to 265+/-1 mosm/kg during hypo-osmolality (both P<0.001). Total body leucine flux and whole-body leucine oxidation rate decreased significantly during hypo-osmolality. Glucose metabolic clearance rate during hyperinsulinaemic-euglycemic clamping increased significantly less during hypo-osmolality than iso-osmolality, while glycerol turnover increased significantly under hypo-osmolar conditions.
Keller U, Szinnai G, Bilz S, et al. Eur J Clin Nutr, 2003. PMID 14681716
2002 / Randomized crossover trial
Twenty healthy adults consumed four grain products (whole-kernel rye bread, whole-meal rye bread with oat beta-glucan, dark durum wheat pasta, and white wheat bread) providing 50 g available carbohydrate. Glucose responses and gastric emptying rates did not differ between the rye breads or pasta and white wheat bread. However, postprandial insulin, GIP, and GLP-1 responses (except GLP-1 for the beta-glucan rye bread) were lower after consuming the rye breads and pasta compared with white wheat bread.
Juntunen KS, Niskanen LK, Liukkonen KH, et al. Am J Clin Nutr, 2002. PMID 11815315
2002 / Randomized controlled trial
A total of 3,234 nondiabetic adults with elevated fasting and post-load glucose concentrations were randomly assigned to placebo, metformin, or a lifestyle-modification program for an average of 2.8 years. The lifestyle intervention reduced diabetes incidence by 58% (95% CI, 48 to 66%), whereas metformin reduced incidence by 31% (95% CI, 17 to 43%), relative to placebo. Lifestyle modification proved significantly more effective than metformin.
Knowler WC, Barrett-Connor E, Fowler SE, et al. N Engl J Med, 2002. PMID 11832527
2002 / Randomized controlled trial
Patients with high-ferritin type 2 diabetes were randomized to blood letting (three 500-ml phlebotomies at 2-week intervals) or observation, with outcomes evaluated at baseline, 4 months, and 12 months. Blood HbA1c decreased significantly only in the blood-letting group at 4 months (mean difference, -0.61; 95% CI, -0.17 to -1.048; P = 0.01). Changes in insulin sensitivity differed significantly between groups at 4 months (80.6 +/- 43.2% vs. -8.6 +/- 9.9%, P = 0.049) and at 12 months (55.5 +/- 24.8% vs. -26.8 +/- 9.9%, P = 0.005). AUC C-peptide decreased by -10.2 +/- 6.3% after blood letting and increased by 10.4 +/- 6.4% in controls at 4 months (P = 0.032), returning to baseline levels in both groups at 12 months.
Fernández-Real JM, Peñarroja G, Castro A, et al. Diabetes, 2002. PMID 11916918
2002 / Randomized crossover trial
Eleven overweight or obese adults with hyperinsulinemia followed whole-grain or refined-grain diets for six weeks in a randomized crossover trial. Fasting insulin was 10% lower following the whole-grain phase compared to refined grains (difference: -15 +/- 5.5 pmol/L; P = 0.03). In addition, whole grains led to a higher glucose infusion rate during the final 30 minutes of a hyperinsulinemic clamp (difference of 0.07 x 10(-4), 95% CI: 0.003 x 10(-4) to 0.144 x 10(-4)).
Pereira MA, Jacobs DR, Pins JJ, et al. Am J Clin Nutr, 2002. PMID 11976158
2002 / Randomized controlled trial
An 82-patient randomized controlled trial evaluated the effects of a low-protein diet (prescribed 0.6 g/kg/day) compared with a usual-protein diet over four years in type 1 diabetic patients with progressive diabetic nephropathy. Actual intake was 0.89 g/kg/day in the low-protein group versus 1.02 g/kg/day in the usual-protein group (P = 0.005). Mean declines in GFR were 3.8 mL/min/year in the low-protein group and 3.9 mL/min/year in the usual-protein group. ESRD or death occurred in 10% of patients on the low-protein diet compared with 27% on the usual-protein diet (P = 0.042; adjusted relative risk 0.23, 95% CI 0.07 to 0.72).
Hansen HP, Tauber-Lassen E, Jensen BR, et al. Kidney Int, 2002. PMID 12081581
2002 / Controlled dietary intervention trial
In 24 individuals with diabetes participating in either a two-week crossover (n=11) or six-week trial (n=13), a diet low in advanced glycation end products reduced serum AGEs by 30% at two weeks (P = 0.02) and 40% at six weeks (P = 0.02). Conversely, a high-AGE diet increased serum AGEs by 64.5% at two weeks (P = 0.02). At six weeks, the high-AGE diet increased peripheral blood mononuclear cell TNF-alpha by 86.3% (P = 0.006) and C-reactive protein by 35% (P = 0.014).
Vlassara H, Cai W, Crandall J, et al. Proc Natl Acad Sci U S A, 2002. PMID 12429856