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.
Patterns in how people eat and what happens to them. Shows an association, not proof.
2016 / Cross-sectional survey analysis
Using 2007-2010 NHANES data, researchers assessed the prevalence of iron deficiency (ID), anemia, and iron deficiency anemia (IDA) in children aged 1-5 years. Overall prevalence was 7.1% (5.5, 8.7) for ID, 3.9% (2.0, 4.3) for anemia, and 1.1% (0.6, 1.7) for IDA. The prevalence of both ID and anemia was significantly higher among children aged 1-2 years (p < 0.05), and 50% of anemic children aged 1-2 years were iron deficient. The findings suggest little change in these indicators over the past decade.
Shows an association, not proof of cause and effect.
Gupta PM, Perrine CG, Mei Z, et al. Nutrients, 2016. PMID 27249004
2016 / Ex vivo observational tissue study
Researchers obtained subcutaneous and visceral adipose tissue biopsies from extremely obese participants during gastric-bypass surgery to test for circadian rhythms in insulin signaling via AKT phosphorylation over 24 hours in vitro. Subcutaneous adipose tissue showed robust circadian rhythms in insulin signaling (P < 0.00001), peaking around noon and being 54% higher than at midnight (P = 0.009). The amplitude was positively correlated with sleep duration (r = 0.53; P = 0.023) and negatively with bedtime (r = -0.54; P = 0.020). No circadian rhythms were detected in visceral adipose tissue (P = 0.643).
Shows an association, not proof of cause and effect.
Carrasco-Benso MP, Rivero-Gutierrez B, Lopez-Minguez J, et al. FASEB J, 2016. PMID 27256623
2016 / Cross-sectional observational study
A cross-sectional study compared sperm characteristics among 474 males categorized as lacto-ovo vegetarians (N=26), vegans (N=5), or non-vegetarians (N=443). Lacto-ovo vegetarians had lower sperm concentration (50.7±7.4 M/mL vs 69.6±3.2 M/mL in non-vegetarians). Total motility was lower in lacto-ovo vegetarians (33.2±3.8%) and vegans (51.8±13.4%) compared to non-vegetarians (58.2±1.0%), and vegans had the lowest hyperactive motility (0.8±0.7% vs 5.2±1.2% in lacto-ovo and 4.8±0.3% in non-vegetarians). Strict morphology, rapid progression, and chromatin integrity did not differ between groups, and authors noted the results should be cautiously interpreted.
Shows an association, not proof of cause and effect.
Orzylowska EM, Jacobson JD, Bareh GM, et al. Eur J Obstet Gynecol Reprod Biol, 2016. PMID 27280539
2016 / Longitudinal cohort study
Data from 1,277 participants who attempted to quit smoking in the Ontario Tobacco Survey were evaluated using four different sets of assumptions to estimate attempts needed to successfully quit. The estimated average number of quit attempts before successful cessation ranged from 6.1 under prior research assumptions, to 19.6 using a constant rate approach, 29.6 using the method with expected lowest bias, and 142 including previous recall history. The authors conclude that previous metrics may underestimate attempts, and for many smokers it may take 30 or more attempts to succeed.
Shows an association, not proof of cause and effect.
Chaiton M, Diemert L, Cohen JE, et al. BMJ Open, 2016. PMID 27288378
2016 / Systematic analysis / comparative risk assessment
Using Global Burden of Disease Study 2013 data, researchers estimated the population-attributable fraction of stroke-related DALYs associated with 17 modifiable environmental, behavioural, and metabolic risk factors across 188 countries. Globally, 90.5% (95% UI 88.5-92.2) of the stroke burden was attributable to the analyzed modifiable risks. Behavioural factors (smoking, poor diet, and low physical activity) accounted for 74.2% (95% UI 70.7-76.7) of stroke DALYs, metabolic factors accounted for 72.4%, and environmental factors accounted for 33.4%.
Shows an association, not proof of cause and effect.
Feigin VL, Roth GA, Naghavi M, et al. Lancet Neurol, 2016. PMID 27291521
2016 / Prospective cohort studies
Three prospective US cohorts comprising 200,727 women and men were followed for over 4 million person-years, documenting 16,162 incident T2D cases. Plant-based diet indices (overall PDI, healthful hPDI, and unhealthful uPDI) were assessed from food frequency questionnaires collected every 2-4 years. Both PDI and hPDI were inversely associated with T2D, with BMI-adjusted hazard ratios for extreme deciles of 0.80 (95% CI: 0.74-0.87) and 0.66 (95% CI: 0.61-0.72), respectively. Conversely, uPDI was positively associated with T2D after BMI adjustment (HR 1.16, 95% CI: 1.08-1.25). Limitations noted include self-reported dietary measurement error and the potential for residual or unmeasured confounding.
Shows an association, not proof of cause and effect.
Satija A, Bhupathiraju SN, Rimm EB, et al. PLoS Med, 2016. PMID 27299701
2016 / Repeated cross-sectional study
Using 24-hour dietary recalls from 33,932 US adults across 7 NHANES cycles (1999-2012), researchers assessed trends in diet quality and components. The primary AHA diet score improved from 19.0 to 21.2 points (11.6% improvement), and adults with poor diets declined from 55.9% to 45.6%. Improvements were driven by increased intake of whole grains (0.43 servings/d) and nuts or seeds (0.25 servings/d), and decreased sugar-sweetened beverages (0.49 servings/d), while total fruits/vegetables, processed meat, saturated fat, and sodium showed no significant trends. Disparities persisted or worsened: poor diet prevalence declined among non-Hispanic white adults (53.9% to 42.8%), but similar improvements were not observed for non-Hispanic black or Mexican American adults.
Shows an association, not proof of cause and effect.
Rehm CD, Peñalvo JL, Afshin A, et al. JAMA, 2016. PMID 27327801
2016 / Repeated cross-sectional study
Data from 63,890 participants in the National Health Nutrition and Examination Survey (1999-2014) were examined to evaluate blood lead trends. Mean blood lead levels fell from 1.65 μg/dL (95% CI, 1.62-1.68) in 1999-2000 to 0.84 μg/dL (95% CI, 0.82-0.86) in 2013-2014 (P < .001), with significant reductions across age, gender, ethnicity, and pregnancy status. The percentage of children with blood lead levels ≥5 μg/dL dropped from 9.9% in 1999-2000 to 0.5% in 2013-2014 (P < .05). For children aged 1 to 5 years during 2011-2014, the estimated 97.5 percentile of blood lead was 3.48 μg/dL.
Shows an association, not proof of cause and effect.
Tsoi MF, Cheung CL, Cheung TT, et al. Am J Med, 2016. PMID 27341956
2016 / Two-sample Mendelian randomization study
Researchers conducted a two-sample Mendelian randomization study using summary statistics from the GIANT consortium (n = 322,105) and the IMSGC (14,498 cases and 24,091 controls) using 70 BMI-associated SNPs. A 1 standard deviation increase in genetically determined BMI was associated with a 41% increase in odds of multiple sclerosis (OR 1.41, 95% CI 1.20-1.66, p = 2.7 x 10^-5, I2 = 0%). Sensitivity analyses showed no evidence of pleiotropic effects, though the authors note that residual pleiotropy remains difficult to exclude entirely.
Shows an association, not proof of cause and effect.
Mokry LE, Ross S, Timpson NJ, et al. PLoS Med, 2016. PMID 27351487
2016 / Cross-sectional analysis
Researchers analyzed NHANES 2001-2006 data from 10,308 US adults to examine associations between an individual-level subsidy score (percentage of total calories from subsidized commodities) and cardiometabolic markers. Overall, 56.2% of calories consumed came from major subsidized commodities. Comparing the highest to the lowest quartile of subsidy score, adults had higher probabilities of BMI >= 30 (PR 1.37; 95% CI, 1.23-1.52), waist-to-height ratio >= 0.60 (PR 1.41; 95% CI, 1.25-1.59), CRP >= 0.32 mg/dL (PR 1.34; 95% CI, 1.19-1.51), elevated non-HDL cholesterol (PR 1.14; 95% CI, 1.05-1.25), and dysglycemia (PR 1.21; 95% CI, 1.04-1.40). No statistically significant association was observed with blood pressure.
Shows an association, not proof of cause and effect.
Siegel KR, McKeever Bullard K, Imperatore G, et al. JAMA Intern Med, 2016. PMID 27379488
2016 / Prospective cohort study
In the HEIJO-KYO cohort, 1110 elderly participants were assessed at baseline and 766 at a median follow-up of 21 months to examine associations between objectively measured ambient light exposure and obesity parameters. Multivariable models adjusted for caloric intake, physical activity, and sleep/wake parameters showed nighttime or evening exposure to higher light intensity was significantly linked to subsequent %WHtR gain. Morning exposure for a longer time ≥500 lux or nighttime exposure for a longer time <3 lux was significantly associated with subsequent %WHtR loss, with consistent trends for BMI. Increased nighttime light exposure (mean ≥3 vs <3 lux) was estimated to correspond to a 10.2% WHtR gain and a 10.0% BMI increase over 10 years.
Shows an association, not proof of cause and effect.
Obayashi K, Saeki K, Kurumatani N J Clin Endocrinol Metab, 2016. PMID 27383113
2016 / Longitudinal cohort study
Researchers analyzed longitudinal food diaries of 12,385 randomly sampled Australian adults across 2007, 2009, and 2013 from the Household, Income, and Labour Dynamics in Australia Survey. After adjusting for changing incomes and personal circumstances, increased fruit and vegetable consumption predicted increased happiness, life satisfaction, and well-being. Gains were up to 0.24 life-satisfaction points for an increase of 8 portions a day, an effect size comparable to moving from unemployment to employment, with improvements occurring within 24 months.
Shows an association, not proof of cause and effect.
Mujcic R, J Oswald A Am J Public Health, 2016. PMID 27400354
2016 / Cross-sectional analysis of NHANES data
Using NHANES 2009-2012 data, researchers evaluated the link between inadequate hydration (urine osmolality >= 800 mOsm/kg) and weight status in 9,528 US adults aged 18 to 64 years. In the sample, 32.6% were inadequately hydrated. After adjusting for confounders, adults with inadequate hydration had significantly higher BMIs (1.32 kg/m2; 95% CI, 0.85-1.79; P < .001) and higher odds of obesity (OR = 1.59; 95% CI, 1.35-1.88; P < .001) compared with hydrated adults.
Shows an association, not proof of cause and effect.
Chang T, Ravi N, Plegue MA, et al. Ann Fam Med, 2016. PMID 27401419
2016 / Matched case-control study
Researchers conducted a matched case-control study linking the Iowa Department of Motor Vehicles Drivers' License Database with cancer registry data to evaluate the relationship between BMI and invasive penile cancer in men from 1985 to 2010. Obesity was significantly associated with an increased risk of penile cancer. Specifically, for every five-unit increase in BMI, the risk of invasive penile cancer increased by 53% (OR 1.53, 95% CI 1.29-1.81, p < 0.0001).
Shows an association, not proof of cause and effect.
Barnes KT, McDowell BD, Button A, et al. BMC Urol, 2016. PMID 27411982
2016 / Prospective cohort study
Researchers followed 82,643 women from the NHS and NHSII for 10 years, assessing flavonoid intake via validated food-frequency questionnaires. Over follow-up, 10,752 incident depression cases occurred. Inverse associations with depression risk were observed for flavonol (HR: 0.93; 95% CI: 0.88, 0.99), flavone (HR: 0.92; 95% CI: 0.86, 0.98), and flavanone (HR: 0.90; 95% CI: 0.85, 0.96) intakes when comparing quintile 5 with quintile 1. Consuming ≥2 servings of citrus fruit or juices per day was associated with lower risk compared with <1 serving per week (HR: 0.82; 95% CI: 0.74, 0.91).
Shows an association, not proof of cause and effect.
Chang SC, Cassidy A, Willett WC, et al. Am J Clin Nutr, 2016. PMID 27413131
2016 / Retrospective poison center registry analysis
Researchers analyzed 430 single-substance human exposure calls regarding edible marijuana products reported to the US National Poison Data System from January 2013 to December 2015. Calls increased each year, with the highest numbers in Colorado (n=166) and Washington (n=96), and 91% occurring in states with decriminalized medical/recreational marijuana. The most common age groups were ≤5 years (n=109) and 13–19 years (n=78). Frequent clinical effects included drowsiness/lethargy (43%), tachycardia (31%), agitation/irritability (14%), and confusion (14%). Common therapies included intravenous fluids (20%), dilute/irrigate/wash (11%), and benzodiazepines (11%). Three patients (ages 4, 10, and 57 years) required intubation, while no deaths were reported.
Shows an association, not proof of cause and effect.
Cao D, Srisuma S, Bronstein AC, et al. Clin Toxicol (Phila), 2016. PMID 27418198
2016 / Cross-sectional study
Researchers evaluated the association between drinking reverse osmosis (RO) processed water and vitamin B12 deficiency in 250 participants from Western India. Overall, 70 (28%) participants were vitamin B12 deficient. Vitamin B12 deficiency was present in 40 (50.6%) of 79 participants using RO water compared to 30 (17.5%) of 171 using other sources, and logistic regression showed an independent association between RO water use and deficiency. Male gender, milk intake under 100 ml per day, and duration of RO water intake were significant in univariate analysis but not in logistic regression.
Shows an association, not proof of cause and effect.
Gupta ES, Sheth SP, Ganjiwale JD J Clin Diagn Res, 2016. PMID 27437269
2016 / Cross-sectional analysis of a population-based case-control study
Researchers examined whether seasoning variety (number of 11 seasonings) and quality when preparing beans and white rice were associated with staple food intake in 1,025 Costa Rican adults. In urban areas only, seasoning variety was positively associated with daily servings of beans (β = .02, P = .01; 1.31 and 1.23 servings/day in tertiles 2 and 3 vs 1.02 servings/day in tertile 1, P < .05) and rice (β = .04, P = .005). No differences in ingredient quality across intakes of beans or rice were observed, and joint associations between variety and quality were not significant.
Shows an association, not proof of cause and effect.
Vadiveloo MK, Campos H, Mattei J Nutr Res, 2016. PMID 27440532
2016 / Observational cross-sectional study
Researchers examined how levels of dietary ergothioneine change with age and cognitive status in an elderly population. Whole blood ergothioneine levels declined significantly beyond 60 years of age. Furthermore, a subset of participants with mild cognitive impairment had significantly lower plasma ergothioneine levels compared with age-matched subjects.
Shows an association, not proof of cause and effect.
Cheah IK, Feng L, Tang RMY, et al. Biochem Biophys Res Commun, 2016. PMID 27444382
2016 / Retrospective cohort study
Researchers analyzed hospital admissions and regional poison center cases for pediatric marijuana exposure in Colorado from 2009 to 2015. Mean hospital visits rose from 1.2 to 2.3 per 100,000 population comparing the 2 years before and after legalization (P = .02), with infused edibles accounting for 48% of known products at the hospital and 52% of RPC exposures. Annual pediatric RPC cases rose over 5-fold from 9 in 2009 to 47 in 2015, averaging a 34% annual increase in Colorado versus 19% in the rest of the US (P < .001).
Shows an association, not proof of cause and effect.
Wang GS, Le Lait MC, Deakyne SJ, et al. JAMA Pediatr, 2016. PMID 27454910
2016 / Cross-sectional comparative study
Researchers analyzed serum IgG1, IgG2, IgG3, and IgG4 antibodies specific for wheat, milk, ovalbumin, E. coli, B. fragilis, and S. cerevisiae in patients with Crohn's disease (CD), ulcerative colitis (UC), acute gastroenteritis/colitis, and non-inflammatory controls. Anti-food IgG subclass levels were not significantly different between IBD patients and controls and did not correlate with food intolerance. In contrast, ASCA of all IgG subclasses and anti-B. fragilis IgG1 were elevated in CD patients, and complicated CD was associated with increased ASCA IgG1/IgG3/IgG4, anti-B. fragilis IgG1, and anti-E. coli IgG1.
Shows an association, not proof of cause and effect.
Jansen A, Mandić AD, Bennek E, et al. Scand J Gastroenterol, 2016. PMID 27455092
2016 / Prospective cohort study
Researchers analyzed 131,342 participants from the Nurses' Health Study and Health Professionals Follow-up Study using validated food frequency questionnaires. Animal protein intake was associated with higher cardiovascular mortality (HR, 1.08 per 10% energy increment; 95% CI, 1.01-1.16), while plant protein was associated with lower all-cause mortality (HR, 0.90 per 3% energy increment; 95% CI, 0.86-0.95) and cardiovascular mortality (HR, 0.88; 95% CI, 0.80-0.97). These associations were confined to participants with at least one unhealthy lifestyle factor. Substituting 3% of energy from plant protein for processed red meat, unprocessed red meat, or egg was linked to lower all-cause mortality (HRs of 0.66, 0.88, and 0.81, respectively).
Shows an association, not proof of cause and effect.
Song M, Fung TT, Hu FB, et al. JAMA Intern Med, 2016. PMID 27479196
2016 / Prospective cohort study
Researchers followed 43,880 healthy men in the Health Professionals Follow-Up Study for 24 years to examine associations between habitual anthocyanin and flavanone intake and CVD events. Over follow-up, 4,046 MI and 1,572 stroke cases were confirmed. Higher anthocyanin intake was inversely associated with nonfatal MI (multivariate HR: 0.87; 95% CI: 0.75, 1.00; P = 0.04; P-trend = 0.098), which was stronger in normotensive participants (HR: 0.81; 95% CI: 0.69, 0.96; P-interaction = 0.03), but was not associated with total or fatal MI or stroke. Higher flavanone intake was not associated with MI or total stroke risk, but was linked to lower risk of ischemic stroke (HR: 0.78; 95% CI: 0.62, 0.97; P = 0.03, P-trend = 0.059), with the greatest magnitude in participants aged ≥65 years (P-interaction = 0.04).
Shows an association, not proof of cause and effect.
Cassidy A, Bertoia M, Chiuve S, et al. Am J Clin Nutr, 2016. PMID 27488237
2016 / Cross-sectional survey analysis
Using NHANES 2003-2006 24-h recalls and food-frequency questionnaires, researchers estimated usual coffee consumption among US adults aged ≥20 years, and evaluated trends from 2003 to 2012. An estimated 75% of US adults (154 million) reported drinking coffee, and 49% reported drinking it daily. Among coffee drinkers, mean usual intake was 14.1 ± 0.5 fluid ounces/d (417 ± 15 mL/d). Mean usual intakes were higher in men vs women, older ages vs ages 20 to <30 y, non-Hispanic whites vs non-Hispanic blacks or Hispanic/other races, smokers vs never smokers, and daily alcohol drinkers vs nonconsumers (all P < 0.05). Population mean intake remained stable from 2003 to 2012 (P-trend = 0.09).
Shows an association, not proof of cause and effect.
Loftfield E, Freedman ND, Dodd KW, et al. J Nutr, 2016. PMID 27489008