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.
2018 / Umbrella review of meta-analyses
An umbrella review analyzed 13 meta-analyses of observational studies and 16 meta-analyses of RCTs assessing adherence to the Mediterranean diet across 37 health outcomes in over 12,800,000 subjects. Robust evidence (P < 0.001, large sample size, and not considerable heterogeneity) supported an association between greater adherence and reduced risks of overall mortality, cardiovascular diseases, coronary heart disease, myocardial infarction, overall cancer incidence, neurodegenerative diseases, and diabetes. Evidence was only suggestive or weak for most site-specific cancers, inflammatory, and metabolic parameters. No evidence was reported for bladder, endometrial, and ovarian cancers, or LDL-cholesterol levels.
Shows an association, not proof of cause and effect.
Dinu M, Pagliai G, Casini A, et al. Eur J Clin Nutr, 2018. PMID 28488692
2018 / Meta-analysis of randomized controlled trials
A meta-analysis of 19 randomized controlled trials with 797 subjects assessed Chlorella supplementation on cardiovascular risk markers. Chlorella significantly reduced total cholesterol by -9.09 mg/dL (95% CI -12.91 to -5.26), LDL cholesterol by -8.32 mg/dL (95% CI -12.22 to -4.42), systolic blood pressure by -4.51 mmHg (95% CI -6.53 to -2.48), diastolic blood pressure by -1.64 mmHg (95% CI -3.28 to -0.01), and fasting blood glucose by -4.23 mg/dL (95% CI -8.29 to -0.17). Triglycerides, HDL cholesterol, and BMI did not change significantly.
Fallah AA, Sarmast E, Habibian Dehkordi S, et al. Clin Nutr, 2018. PMID 29037431
2018 / Systematic review
A systematic review evaluated 113 controlled human intervention trials testing 17 putative superfoods on risk markers for metabolic syndrome. Studied foods included garlic (21 trials), flaxseed (22 trials), cocoa (16 trials), blueberries (8 trials), and strawberries (7 trials). The authors found limited overall evidence for beneficial impacts on waist circumference, BMI, blood pressure, blood glucose, or lipids, citing inconsistent findings across trials and a small number of studies for many evaluated foods.
van den Driessche JJ, Plat J, Mensink RP Food Funct, 2018. PMID 29557436
2018 / Systematic review and meta-analysis of controlled clinical trials
A systematic review and meta-analysis of 8 controlled clinical trials with 842 participants evaluated the effects of yoga compared to physical exercise in type 2 diabetes mellitus. Pooled analysis revealed significant reductions favoring yoga for FBG (15.16 mg/dl), PPBG (28.66 mg/dl), HbA1c (0.39%), and BMI (0.71 kg/m2). No significant differences were observed for lipid parameters, waist circumference, waist-to-hip ratio, or blood pressure. The authors noted considerable heterogeneity among individual studies.
Jayawardena R, Ranasinghe P, Chathuranga T, et al. Diabetes Metab Syndr, 2018. PMID 29685823
2018 / Dose-response meta-analysis of prospective cohort studies
A dose-response meta-analysis of six prospective cohort studies (4,545,230 person-years; 17,758 diabetes cases) examined potato consumption and type 2 diabetes risk. The pooled HR per 1 serving/day increase was reported as 1.20 (95% CI 1.13 to 1.127, P < 0.001) in both men and women. Reported HRs were 1.44 (95% CI 1.28 to 1.63) for 2 servings/day and 1.74 (95% CI 1.45 to 2.09) for 3 servings/day. A non-linear association was observed (X2 = 17.5, P for linearity < 0.001).
Shows an association, not proof of cause and effect.
Bidel Z, Teymoori F, Davari SJ, et al. Clin Nutr ESPEN, 2018. PMID 30144898
2018 / Systematic review and meta-analysis of controlled feeding trials
In controlled feeding trials of at least one week, small amounts of fructose and tagatose decreased HbA1c (mean difference -0.38% and -0.20%, respectively) and reduced blood glucose concentrations without altering fasting insulin levels. Small quantities of allulose showed a borderline decrease in fasting glucose (p = 0.05) without changing HbA1c or insulin, while sorbose had no eligible trials. Evidence certainty was rated low for fructose and allulose, and moderate for tagatose.
Noronha JC, Braunstein CR, Blanco Mejia S, et al. Nutrients, 2018. PMID 30463314
2018 / Systematic review of controlled trials
This systematic review of 11 controlled trials (n=433; mean sample age 54.8 years) evaluated plant-based diet interventions targeting diabetes in adults. Plant-based diets were associated with significant improvement in emotional well-being, physical well-being, depression, quality of life, general health, HbA1c levels, weight, total cholesterol, and low-density lipoprotein cholesterol compared with diabetic associations' official guidelines and other comparator diets. The authors concluded that plant-based diets can significantly improve psychological health, quality of life, HbA1c levels, weight, and diabetes management.
Toumpanakis A, Turnbull T, Alba-Barba I BMJ Open Diabetes Res Care, 2018. PMID 30487971
2018 / Meta-analysis of observational studies
A meta-analysis of 40 observational studies assessed cardiometabolic risk factors in 12,619 vegans and 179,630 omnivores. Compared to controls, vegans had lower body mass index (-1.72 kg/m2, CI -2.30 to -1.16), waist circumference (-2.35 cm, CI -3.93 to -0.76), LDL cholesterol (-0.49 mmol/L, CI -0.62 to -0.36), triglycerides (-0.14 mmol/L, CI -0.24 to -0.05), fasting blood glucose (-0.23 mmol/L, CI -0.35 to -0.10), systolic blood pressure (-2.56 mmHg, CI -4.66 to -0.45), and diastolic blood pressure (-1.33 mmHg, CI -2.67 to -0.02; p<0.0001 for all). In several large studies from Taiwan, however, a vegan diet was not associated with favourable cardio-metabolic risk factors compared to control diets.
Shows an association, not proof of cause and effect.
Benatar JR, Stewart RAH PLoS One, 2018. PMID 30571724
2018 / Meta-analysis of prospective cohort studies
A meta-analysis evaluated the association between potato consumption and type 2 diabetes (T2D) risk across prospective cohort studies. The pooled relative risk (RR) comparing the highest versus lowest category of potato consumption was 1.077 (95% CI: 1.005, 1.155), with a 3.5% increased risk (RR=1.035, 95% CI: 1.004-1.067) per additional three servings per week. For French fries, the highest versus lowest intake had a pooled RR of 1.362 (95% CI: 1.004, 1.850), and risk increased by 18.7% (RR=1.187, 95% CI: 1.067-1.321) for every additional three servings per week.
Shows an association, not proof of cause and effect.
Zhang Y, You D, Lu N, et al. Iran J Public Health, 2018. PMID 30581777
2017 / Systematic review and meta-analysis of cohort studies
A systematic review identified 19 cohort studies on rice intake, chronic conditions, and mortality. In a meta-analysis of 12 cohorts, comparing extreme categories of rice intake showed a non-significant trend for total chronic diseases (RR 1.11, 95% CI: 0.96 to 1.29), with an association present in women (RR 1.40, 95% CI: 1.13 to 1.73) but not men (RR 0.95, 95% CI: 0.72 to 1.24). Across five cohorts, high intake was not associated with mortality overall (RR 0.97, 95% CI: 0.88 to 1.06), but linked to lower mortality in men (RR 0.87, 95% CI: 0.81 to 0.94).
Shows an association, not proof of cause and effect.
Saneei P, Larijani B, Esmaillzadeh A Public Health Nutr, 2017. PMID 27577106
2017 / Systematic review
An updated systematic review examined 31 human studies published since 1994 on weight cycling and metabolic risk factors. A history of weight cycling was correlated with increased body fat and central adiposity in 58% (11/19) of publications, and 50% (4/8) found it increased the likelihood of future weight gain. However, 76% (13/17) of studies showed no detrimental effect on type 2 diabetes risk. The authors concluded evidence is inconclusive regarding body composition or predisposition to future obesity, and suggests little detrimental metabolic effect.
Shows an association, not proof of cause and effect.
Mackie GM, Samocha-Bonet D, Tam CS Obes Res Clin Pract, 2017. PMID 27773644
2017 / Systematic review and meta-analysis of controlled trials
A systematic review and meta-analysis of controlled trials was conducted to evaluate the effect of vinegar intake with a meal on postprandial glucose and insulin responses. Pooled analysis revealed a significant reduction in glucose AUC (standard mean difference = -0.60, 95% CI -1.08 to -0.11, p = 0.01) and insulin AUC (standard mean difference = -1.30, 95% CI -1.98 to -0.62, p < 0.001) in participants consuming vinegar compared with controls. The authors concluded vinegar can effectively reduce postprandial glucose and insulin levels.
Shishehbor F, Mansoori A, Shirani F Diabetes Res Clin Pract, 2017. PMID 28292654
2017 / Meta-analysis of observational studies
A meta-analysis evaluated 14 observational studies from 13 articles assessing vegetarian eating patterns and diabetes risk. Vegetarians had a lower pooled odds of diabetes compared with non-vegetarians (OR 0.726, 95% CI: 0.608 to 0.867). Inverse associations were observed in the Western Pacific (OR 0.514, 95% CI: 0.304 to 0.871) and Europe or North America (OR 0.756, 95% CI: 0.589 to 0.971), but not significantly in Southeast Asia.
Shows an association, not proof of cause and effect.
Lee Y, Park K Nutrients, 2017. PMID 28613258
2017 / Umbrella review of meta-analyses of observational studies and RCTs
An umbrella review evaluated evidence from meta-analyses of observational studies and randomized controlled trials on coffee and caffeine consumption. Across 112 observational meta-analyses covering 59 unique outcomes, coffee was associated with a probable decreased risk of breast, colorectal, colon, endometrial, and prostate cancers, cardiovascular disease and mortality, Parkinson's disease, and type-2 diabetes. Across 20 observational meta-analyses covering 14 unique outcomes, caffeine was associated with a probable decreased risk of Parkinson's disease and type-2 diabetes, but an increased risk of pregnancy loss. Among 9 RCT meta-analyses examining 12 acute outcomes, coffee was associated with elevated serum lipids (affected by significant heterogeneity) and caffeine with elevated blood pressure.
Shows an association, not proof of cause and effect.
Grosso G, Godos J, Galvano F, et al. Annu Rev Nutr, 2017. PMID 28826374
2017 / Meta-analysis of prospective cohorts
A meta-analysis of 11 prospective cohorts evaluated protein consumption across 483,174 participants and 52,637 incident cases. Relative risks for type 2 diabetes were 1.12 (95% CI: 1.08-1.17) for total protein and 1.14 (95% CI: 1.09-1.19) for animal protein. Plant protein intake was inversely associated with risk in women (RR 0.92; 95% CI: 0.86-0.98), while red and processed meats were linked to higher risk.
Shows an association, not proof of cause and effect.
Tian S, Xu Q, Jiang R, et al. Nutrients, 2017. PMID 28878172
2017 / Meta-analysis of randomized controlled trials
A meta-analysis of 16 randomized controlled trials with 538 adult participants evaluated pomegranate supplementation. Pomegranate intake did not significantly alter fasting blood glucose (WMD -0.6 mg/dL, 95% CI: -2.79 to 1.58), fasting blood insulin (WMD 0.29 μIU/mL, 95% CI: -1.16 to 1.75), HOMA-IR, or HbA1c (WMD -0.11%, 95% CI: -0.39 to -0.18). The authors concluded pomegranate is not recommended for glycemic management.
Huang H, Liao D, Chen G, et al. Nutr J, 2017. PMID 28985741
2016 / Meta-analysis of randomized controlled trials
A meta-analysis of 11 randomized controlled trials evaluated the link between niacin therapy and new-onset diabetes in 26,340 non-diabetic participants over a weighted mean follow-up of 3.6 years. In total, 1,371 participants developed diabetes (725 of 13,121 assigned to niacin vs 646 of 13,219 controls). Niacin was associated with a significantly higher risk of new-onset diabetes (RR 1.34, 95% CI 1.21 to 1.49; I2 = 0.0%), equating to one extra diabetes case per 43 participants treated for 5 years. Results remained consistent regardless of background statin therapy (p = 0.88) or combination with laropiprant (p = 0.52).
Goldie C, Taylor AJ, Nguyen P, et al. Heart, 2016. PMID 26370223
2016 / Systematic review and meta-analysis of cohort studies and reviews
A systematic review and meta-analysis evaluated the risk of incident diabetes linked to sleep disturbances compared with traditional risk factors across 36 studies with 1,061,555 participants. Pooled relative risks for sleeping ≤5 h, 6 h, and ≥9 h/d were 1.48 (95% CI: 1.25, 1.76), 1.18 (1.10, 1.26), and 1.36 (1.12, 1.65), respectively. Poor sleep quality (RR 1.40 [1.21, 1.63]), obstructive sleep apnea (RR 2.02 [1.57, 2.61]), and shift work (RR 1.40 [1.18, 1.66]) were also associated with diabetes, while traditional risk factors of overweight, family history, and physical inactivity had pooled RRs of 2.99 (2.42, 3.72), 2.33 (1.79, 2.79), and 1.20 (1.11, 1.32), respectively.
Shows an association, not proof of cause and effect.
Anothaisintawee T, Reutrakul S, Van Cauter E, et al. Sleep Med Rev, 2016. PMID 26687279
2016 / Meta-analysis
A meta-analysis of 86 studies encompassing 8,515,431 individuals across 22 countries assessed nonalcoholic fatty liver disease epidemiology. Global prevalence was 25.24% (95% CI: 22.10-28.65), with frequent comorbidities including obesity (51.34%) and type 2 diabetes (22.51%). Annual liver-specific mortality was 0.77 per 1,000 person-years in nonalcoholic fatty liver disease and 15.44 per 1,000 in nonalcoholic steatohepatitis.
Younossi ZM, Koenig AB, Abdelatif D, et al. Hepatology, 2016. PMID 26707365
2016 / Meta-analysis of randomized controlled trials
A meta-analysis of four randomized controlled trials evaluated the effects of oat β-glucan (OBG) in 350 patients with type 2 diabetes mellitus. Compared to controls, OBG intake of 2.5 to 3.5 g/day for 3 to 8 weeks significantly lowered fasting plasma glucose by -0.52 mmol/L (95% CI: -0.94, -0.10; p = 0.01) and HbA1c by -0.21% (95% CI: -0.40, -0.02; p = 0.03). However, OBG did not significantly alter fasting plasma insulin concentrations. The authors noted that data regarding the long-term effects of OBG intake on glycaemic control and insulin sensitivity were scarce.
Shen XL, Zhao T, Zhou Y, et al. Nutrients, 2016. PMID 26771637
2016 / Meta-analysis with trial sequential analysis of RCTs
A meta-analysis with trial sequential analysis of 22 RCTs (1,251 subjects) evaluated the effects of berry consumption in healthy individuals and patients with CVD. Berry consumption significantly lowered LDL-cholesterol (WMD -0.21 mmol/L; 95% CI -0.34 to -0.07; P = 0.003), systolic blood pressure (WMD -2.72 mmHg; 95% CI -5.32 to -0.12; P = 0.04), fasting glucose (WMD -0.10 mmol/L; 95% CI -0.17 to -0.03; P = 0.004), and BMI (WMD -0.36 kg/m2; 95% CI -0.54 to -0.18; P < 0.00001). It also significantly reduced HbA1c (WMD -0.20%; P = 0.04) and TNF-α (WMD -0.99 pg/mL; P = 0.04), while other markers showed no significant changes.
Huang H, Chen G, Liao D, et al. Sci Rep, 2016. PMID 27006201
2016 / Systematic review and meta-analysis
A systematic review and meta-analysis of 26 papers evaluated the link between metformin use and vitamin B12 deficiency in type 2 diabetes. Ten of 17 observational studies reported statistically significantly lower vitamin B12 levels in patients taking metformin compared to those not taking it. A meta-analysis of four trials found a statistically significant mean B12 reducing effect of metformin of 57 pmol/L [WMD (fixed) = -0.57 (95% CI: -35 to -79 pmol/L)] after 6 weeks to 3 months of use.
Chapman LE, Darling AL, Brown JE Diabetes Metab, 2016. PMID 27130885
2016 / Dose-response meta-analysis of prospective studies
A systematic review and dose-response meta-analysis evaluated 45 prospective studies examining whole grain consumption and disease or mortality risks. A 90 g/day increase in whole grain intake was associated with a lower relative risk of coronary heart disease (RR 0.81, 95% CI 0.75 to 0.87), cardiovascular disease (RR 0.78, 95% CI 0.73 to 0.85), total cancer mortality (RR 0.85, 95% CI 0.80 to 0.91), and all-cause mortality (RR 0.83, 95% CI 0.77 to 0.90). Risk reductions were also seen for respiratory, infectious, and all non-cardiovascular non-cancer mortality, with risk reductions seen up to 210-225 g/day for most outcomes. Little evidence of an association was found for refined grains, white rice, total rice, or total grains.
Shows an association, not proof of cause and effect.
Aune D, Keum N, Giovannucci E, et al. BMJ, 2016. PMID 27301975
2016 / Systematic review and meta-analysis of prospective cohorts
A meta-analysis of 15 country-specific prospective cohorts with 636,151 participants evaluated butter intake over 6.5 million person-years. Each daily 14-gram serving was linked to a slight rise in all-cause mortality (RR = 1.01, 95% CI: 1.00 to 1.03) and a small decrease in diabetes risk (RR = 0.96, 95% CI: 0.93 to 0.99). No significant associations were observed for cardiovascular disease, coronary heart disease, or stroke.
Shows an association, not proof of cause and effect.
NANO counterpoint →
Pimpin L, Wu JH, Haskelberg H, et al. PLoS One, 2016. PMID 27355649