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.
2014 / Systematic review and meta-analysis
A meta-analysis of 25 studies involving 17,332 cancer cases assessed associations between circulating 25(OH)D levels at or near diagnosis and patient outcomes. Comparing the highest to lowest quartiles, pooled hazard ratios for overall survival were 0.55 (95% CI = 0.33-0.91) for colorectal cancer, 0.63 (95% CI = 0.51-0.77) for breast cancer, and 0.48 (95% CI = 0.36-0.64) for lymphoma. Higher 25(OH)D levels were also significantly associated with reduced cancer-specific mortality for colorectal cancer (P = .005) and lymphoma (P < .001), as well as improved disease-free survival for breast cancer (P < .001) or lymphoma (P < .05). A 10-nmol/L increment in circulating 25(OH)D was associated with a hazard ratio of 0.96 (95% CI = 0.95-0.97) for overall survival.
Shows an association, not proof of cause and effect.
Li M, Chen P, Li J, et al. J Clin Endocrinol Metab, 2014. PMID 24780061
2014 / Meta-analysis of randomized controlled trials
A meta-analysis of 124 randomized controlled studies evaluated plant sterols and stanols across 201 strata, with average intake of 2.1 g/d (range 0.2-9.0 g/d). Daily intakes of 0.6 to 3.3 g/d reduced LDL cholesterol by an average of 6% to 12%. Both sterols and stanols exhibited comparable dose-response patterns, with reductions continuing to rise up to approximately 3 g/d.
Ras RT, Geleijnse JM, Trautwein EA Br J Nutr, 2014. PMID 24780090
2014 / Pooled analysis of prospective cohorts
In a pooled analysis of three prospective US cohorts totaling 205,357 participants, dietary intake of n-3 and n-6 polyunsaturated fatty acids was assessed via food-frequency questionnaires. Multivariable pooled relative risks comparing the highest to lowest quartiles ranged from 1.08 to 1.46 for n-3 PUFAs (P-trend = 0.11-0.52) and 0.68 to 1.19 for n-6 PUFAs (P-trend = 0.09-0.54). The authors observed no evidence that intake of fish or n-3 fatty acids lowered suicide mortality.
Shows an association, not proof of cause and effect.
Tsai AC, Lucas M, Okereke OI, et al. Am J Epidemiol, 2014. PMID 24812159
2014 / Systematic review
A critical review reanalyzed dose-response data from international and Japanese studies on sugar consumption and dental caries in children aged 6 years or older and adults. Caries developed in both resistant and susceptible teeth when sugar intakes reached 2% to 3% of energy, given exposure exceeding 3 years. The review observed a progressive linear rise in caries across life despite increased enamel resistance after tooth eruption.
Sheiham A, James WP Public Health Nutr, 2014. PMID 24892213
2014 / Meta-analysis of randomized controlled trials
A meta-analysis evaluated randomized controlled trials of vitamin D supplementation spanning 2 to 7 years. Supplementation at doses of 400 to 1,100 IU daily did not affect total cancer incidence across four trials with 4,333 cases (summary RR 1.00, 95% CI, 0.94 to 1.06). However, doses of 400 to 833 IU daily significantly reduced cancer mortality across three trials with 1,190 deaths (summary RR 0.88, 95% CI, 0.78 to 0.98).
Keum N, Giovannucci E Br J Cancer, 2014. PMID 24918818
2014 / Meta-analysis of prospective cohorts
A meta-analysis of seven prospective cohort studies evaluated dietary protein intake and stroke risk among 254,489 individuals. Comparing the highest to lowest categories of protein intake yielded a pooled relative risk of stroke of 0.80 (95% CI, 0.66 to 0.99). In dose-response evaluation, every 20-g/d increase in protein intake was linked to a 26% lower risk, with animal protein intake showing an RR of 0.71 (95% CI, 0.50 to 0.99).
Shows an association, not proof of cause and effect.
NANO counterpoint →
Zhang Z, Xu G, Yang F, et al. Neurology, 2014. PMID 24920855
2014 / Meta-analysis
A meta-analysis evaluated 32 studies investigating serum 25-hydroxyvitamin D categories and all-cause mortality. Participants in the lowest category (0-9 ng/mL) had higher mortality compared to the highest category (> 30 ng/mL), with a hazard ratio of 1.9 (95% CI: 1.6, 2.2; P < .001). Blood levels at or below 30 ng/mL showed a significant relationship with increased mortality compared to values exceeding that threshold (P < .01).
Garland CF, Kim JJ, Mohr SB, et al. Am J Public Health, 2014. PMID 24922127
2014 / Meta-analysis of prospective cohort studies
In a meta-analysis of nine prospective studies with 13,284 participants, highest versus lowest plasma homocysteine categories were linked to higher risk of ischemic stroke (RR 1.69; 95% CI: 1.29-2.20) and recurrent stroke (RR 1.76; 95% CI: 1.37-2.24). In contrast, no distinct link was found for hemorrhagic stroke (RR 1.65; 95% CI: 0.61-4.45).
Shows an association, not proof of cause and effect.
He Y, Li Y, Chen Y, et al. Nutr Metab Cardiovasc Dis, 2014. PMID 24984821
2014 / Prospective cohort study
In a prospective study of 1,598 patients with stage I to III colorectal cancer, postoperative plasma 25-hydroxyvitamin D concentrations were measured alongside vitamin D receptor genotypes. Comparing the highest to lowest tertile, higher vitamin D levels were associated with lower colorectal cancer mortality (adjusted HR 0.68, 95% CI, 0.50 to 0.90) and all-cause mortality (adjusted HR 0.70, 95% CI, 0.55 to 0.89). Significant interactions were also observed between 25-hydroxyvitamin D levels and vitamin D receptor genetic variants.
Shows an association, not proof of cause and effect.
Zgaga L, Theodoratou E, Farrington SM, et al. J Clin Oncol, 2014. PMID 25002714
2014 / Meta-analysis of randomized controlled trials
A meta-analysis analyzed 30 randomized controlled trials including 5,615 individuals with a mean age of 61.1 years. Vitamin D supplementation had a small positive effect on overall muscle strength (SMD 0.17; P = .02), particularly in participants with baseline 25-hydroxyvitamin D below 30 nmol/L and those aged 65 or older. No significant improvements occurred in muscle mass (SMD 0.058; P = .52) or muscle power (SMD 0.057; P = .657).
Beaudart C, Buckinx F, Rabenda V, et al. J Clin Endocrinol Metab, 2014. PMID 25033068
2014 / Systematic review and meta-analysis of randomized controlled trials
In a meta-analysis of 35 randomized trials with 43,407 participants, vitamin D3 supplementation did not significantly affect insulin resistance (MD -0.04; 95% CI -0.30 to 0.22), beta-cell function (MD 1.64; 95% CI -25.94 to 29.22), or hemoglobin A1c (MD -0.05%; 95% CI -0.12 to 0.03). In four trials assessing progression to diabetes, risk did not differ from control (OR 1.02; 95% CI 0.94 to 1.10). Adverse events were rare.
Seida JC, Mitri J, Colmers IN, et al. J Clin Endocrinol Metab, 2014. PMID 25062463
2014 / Systematic review and meta-analysis of randomized controlled trials
A systematic review evaluated 14 randomized controlled trials with 906 adults experiencing irritable bowel syndrome. Fiber supplementation significantly decreased the risk of persistent symptoms compared to control therapies (RR=0.86; 95% CI 0.80-0.94; NNT=10; 95% CI 6-33). Benefit occurred specifically with soluble fiber (RR=0.83; 95% CI 0.73-0.94; NNT=7), whereas bran showed no significant effect (RR=0.90; 95% CI 0.79-1.03).
Moayyedi P, Quigley EM, Lacy BE, et al. Am J Gastroenterol, 2014. PMID 25070054
2014 / Systematic review
A systematic review evaluated the relationship between nutritional status and periodontal disease. Findings indicated potential adjunct benefits for zinc, lycopene, vitamin E, and vitamin B complex. Nevertheless, existing evidence was inadequate to establish a connection between host nutritional status and periodontal inflammation.
Kulkarni V, Bhatavadekar NB, Uttamani JR J Calif Dent Assoc, 2014. PMID 25087348
2014 / Systematic review
A systematic review analyzed 43 reports from 38 studies published between 2009 and 2012 evaluating fiscal food interventions. Two randomized controlled trials demonstrated that price changes shifted purchasing in grocery store and away-from-home settings. Modeled studies showed larger shifts when taxes targeted noncore foods or beverages with accessible untaxed alternatives, such as soft drinks.
Thow AM, Downs S, Jan S Nutr Rev, 2014. PMID 25091552
2014 / Individual participant meta-analysis
An individual participant meta-analysis examined 14 circulating fatty acids among 5,098 case patients and 6,649 control patients across seven prospective studies over an average 5.1-year follow-up. Stearic acid showed an inverse relationship with total disease (OR Q5 vs Q1 = 0.88, 95% CI: 0.78 to 1.00, P trend = .043). Modest positive associations occurred for eicosapentaenoic acid (OR = 1.14, 95% CI: 1.01 to 1.29) and docosapentaenoic acid (OR = 1.16, 95% CI: 1.02 to 1.33), with between-study heterogeneity.
Crowe FL, Appleby PN, Travis RC, et al. J Natl Cancer Inst, 2014. PMID 25210201
2014 / Systematic review of randomized controlled trials
A systematic review identified only one eligible trial evaluating garlic supplementation for the common cold. That study randomized 146 participants to daily garlic or placebo for 12 weeks, finding fewer colds in the garlic arm (24 versus 65, P < 0.001) and fewer sick days (111 versus 366). Days to recovery were comparable (4.63 versus 5.63 days), leaving insufficient evidence to draw firm conclusions.
Lissiman E, Bhasale AL, Cohen M Cochrane Database Syst Rev, 2014. PMID 25386977
2014 / Systematic review and meta-analysis of randomized controlled trials
A systematic review and meta-analysis of 18 RCTs evaluated the effect of oral vitamin E supplementation on all-cause mortality in 142,219 apparently healthy participants with at least 6 months of follow-up. Fixed effect and random effects analyses showed that vitamin E supplementation was not associated with all-cause mortality (RR 1.01, 95% CI 0.97-1.05, p = 0.65). Subgroup analyses by vitamin E type (natural or synthetic), dose or duration, study design, quality, or pre-specified mortality outcome also found no association.
Curtis AJ, Bullen M, Piccenna L, et al. Cardiovasc Drugs Ther, 2014. PMID 25398301
2013 / Meta-analysis of prospective cohorts
A meta-analysis of six prospective cohort studies containing 314,864 individuals and 8,920 stroke cases assessed dietary fiber consumption. Comparing the highest against the lowest intake categories showed an inverse association with stroke risk (RR 0.87, 95% CI: 0.77-0.99). Dose-response evaluation linked each 10-g daily increase in fiber intake to an estimated 12% lower risk of stroke (RR 0.88, 95% CI: 0.79-0.97).
Shows an association, not proof of cause and effect.
Chen GC, Lv DB, Pang Z, et al. Eur J Clin Nutr, 2013. PMID 23073261
2013 / Meta-analysis of randomized controlled trials
A meta-analysis analyzed 21 randomized trials with 91,074 adults followed for an average of 43 months. Supplementation showed no effect on all-cause mortality (RR 0.98, 95% CI 0.94 to 1.02), vascular mortality (RR 1.01, 95% CI 0.93 to 1.09), or cancer mortality (RR 0.96, 95% CI 0.88 to 1.04). A non-significant trend toward reduced all-cause mortality occurred across primary prevention trials (RR 0.94, 95% CI 0.89 to 1.00).
Macpherson H, Pipingas A, Pase MP Am J Clin Nutr, 2013. PMID 23255568
2013 / Systematic review and meta-analysis
A meta-analysis of seventeen studies examined homocysteine and vitamin B12 levels among omnivores, vegetarians, and vegans. Across 3,230 participants, an inverse relationship was confirmed between circulating homocysteine and serum vitamin B12. Only two studies showed that vegan biomarker levels did not differ from those of omnivores, highlighting a tendency toward elevated homocysteine among individuals avoiding animal products.
NANO counterpoint →
Obersby D, Chappell DC, Dunnett A, et al. Br J Nutr, 2013. PMID 23298782
2013 / Systematic review and meta-analysis of randomized controlled trials
Researchers conducted meta-analyses of 54 randomized controlled trials evaluating dietary and psychological treatments for ADHD. When assessments were made by raters closest to the therapeutic setting, all dietary (SMDs = 0.21–0.48) and psychological (SMDs = 0.40–0.64) treatments had statistically significant effects. Under the best probably blinded assessments, effects remained significant only for free fatty acid supplementation (SMD = 0.16) and artificial food color exclusion (SMD = 0.42), while other treatments were attenuated to nonsignificance. The authors noted that artificial food color exclusion was often tested in individuals selected for food sensitivities, and the clinical significance of free fatty acid supplementation remains to be determined.
Sonuga-Barke EJ, Brandeis D, Cortese S, et al. Am J Psychiatry, 2013. PMID 23360949
2013 / Systematic review and meta-analysis of randomized controlled trials
A meta-analysis of 11 randomized trials including 5,272 participants assessed maternal omega-3 supplementation during pregnancy or lactation. Offspring showed no overall differences in cognitive, language, or motor scores, though two to five-year-olds scored higher in two high-bias trials (mean difference: 3.92; 95% CI: 0.77, 7.08; n = 156). Additionally, six of eight visual assessments across five trials reported no significant differences between groups.
Gould JF, Smithers LG, Makrides M Am J Clin Nutr, 2013. PMID 23364006
2013 / Meta-analysis of prospective cohorts
In a meta-analysis of 24 prospective cohort studies comprising 7.5 million person-years of follow-up, dietary glycemic load was positively associated with type 2 diabetes risk (RR 1.45, 95% CI: 1.31, 1.61 per 100-g daily increment per 2,000 kcal). The association was statistically significant only at intakes exceeding 95 g per 2,000 kcal. Sex, dietary instrument validity, and ethnicity together accounted for 97% of the observed heterogeneity across studies.
Shows an association, not proof of cause and effect.
Livesey G, Taylor R, Livesey H, et al. Am J Clin Nutr, 2013. PMID 23364021
2013 / Meta-analysis of prospective cohorts
A meta-analysis evaluated the long-term health outcomes linked to low-carbohydrate diets across multiple observational cohorts. High low-carbohydrate scores were associated with increased all-cause mortality (RR 1.31, 95% CI 1.07-1.59 across 4 cohorts with 272,216 participants). In contrast, risks for cardiovascular mortality (RR 1.10, 95% CI 0.98-1.24) and cardiovascular incidence (RR 0.98, 95% CI 0.78-1.24) were not statistically significant.
Shows an association, not proof of cause and effect.
Noto H, Goto A, Tsujimoto T, et al. PLoS One, 2013. PMID 23372809