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 / Web-based expert opinion survey
A web-based questionnaire with 20 closed and 5 open questions was administered to topic specialists to gather empirical knowledge and opinions on Mycobacterium avium ssp. paratuberculosis. Specialists believed it is likely a risk to human health (44.8%) and most frequently ranked it as a moderate public health issue (40.1%). Contact with ruminants and dairy products were identified as the most likely human exposure routes. Most respondents agreed on enhancing on-farm industry programmes (68.4%) and allocating government resources for monitoring (92%).
Shows an association, not proof of cause and effect.
Waddell LA, Rajić A, Stärk KD, et al. Zoonoses Public Health, 2016. PMID 26272619
2016 / Cross-sectional survey
Using 2014 Summer Styles survey data from 4,269 U.S. adults, researchers assessed the prevalence and modes of marijuana use as well as reasons for use. Overall, 7.2% reported current use and 34.5% reported ever use. Among current users, 10.5% reported medicinal-only use, 53.4% recreational-only use, and 36.1% both. The most common modes among current users were bowls/pipes (49.5%) and joints (49.2%), while edibles/drinks were reported by 16.1% and vaporizers by 7.6%; 92.1% reported combusted-only use.
Shows an association, not proof of cause and effect.
Schauer GL, King BA, Bunnell RE, et al. Am J Prev Med, 2016. PMID 26277652
2016 / Retrospective study
A retrospective study evaluated 23 Chinese patients with subacute combined degeneration (SCD) for clinical presentations, laboratory tests, and neuroimaging. For most patients, sensory symptoms presented earlier than motor symptoms, and signs of disease were more pronounced than symptoms. Six patients had sensory deficit levels mimicking transverse myelopathy, and anemia was not always detected. Notably, seven patients had normal or even elevated serum levels of vitamin B12, and spinal cord lesions on MRI were detected in six patients.
Shows an association, not proof of cause and effect.
Li J, Ren M, Dong A, et al. Int J Neurosci, 2016. PMID 26279228
2016 / Econometric interrupted time series / demand system analysis
Researchers analyzed monthly sales data from 1,293 stores of a major Danish retail chain between January 2010 and October 2012 to evaluate the impact of the 2011 saturated fat tax. The tax led to a 13-16% price increase for high-fat varieties of minced beef and cream, while having an insignificant or small negative effect on low- and medium-fat varieties. Overall, the tax was associated with a 4-6% decrease in saturated fat intake from minced beef and regular cream, with a negligible effect on sour cream. The authors concluded that the tax had statistically significant effects on sales of fat in these products, but reduced beyond-recommendation saturated fat intake to only a limited extent.
Shows an association, not proof of cause and effect.
Jensen JD, Smed S, Aarup L, et al. Public Health Nutr, 2016. PMID 26306542
2016 / Prospective cohort study
Researchers analyzed 35,107 pregnant women delivering singleton boys in the Norwegian Mother and Child Cohort Study to evaluate whether organic food intake was associated with male urogenital anomalies. Seventy-four male newborns were diagnosed with hypospadias (0.2%) and 151 with cryptorchidism (0.4%). Consuming any organic food during pregnancy was associated with lower odds of hypospadias (OR = 0.42; 95% CI: 0.25, 0.70; 21 cases), with the strongest associations for organic vegetables (OR = 0.36; 95% CI: 0.15, 0.85) and organic milk/dairy (OR = 0.43; 95% CI: 0.17, 1.07). No substantial association was seen for cryptorchidism. The authors noted findings were based on small numbers of cases and require replication.
Shows an association, not proof of cause and effect.
Brantsæter AL, Torjusen H, Meltzer HM, et al. Environ Health Perspect, 2016. PMID 26307850
2016 / Cross-sectional study
A cross-sectional study of 630 dental professionals assessed mercury exposure via questionnaires and biomarker analysis of urine, hair, and blood. Geometric mean levels were 1.28 μg/l in urine, 0.60 μg/g in hair, and 3.67 μg/l in blood, which were higher than the general US population. In multivariable linear regression, fish consumption patterns predicted hair and blood mercury levels, with higher levels in Asians than Caucasians. Five species contributed the majority of estimated mercury intake from fish: swordfish, fresh tuna, white canned tuna, whitefish, and king mackerel.
Shows an association, not proof of cause and effect.
Goodrich JM, Chou HN, Gruninger SE, et al. J Expo Sci Environ Epidemiol, 2016. PMID 26329138
2016 / Prospective cohort study
Researchers examined pre-diagnostic dietary intake and survival among 2,339 participants from the Multiethnic Cohort diagnosed with non-Hodgkin lymphoma during follow-up. During 4.5±4.1 years of follow-up, 1,348 total deaths and 903 NHL-specific deaths occurred. Dairy intake was associated with higher all-cause mortality (highest vs lowest tertile: HR=1.14, 95% CI 1.00-1.31, Ptrend=0.03) and NHL-specific mortality (HR=1.16, 95% CI 0.98-1.37). Legume intake above the lowest tertile was related to 13-16% lower all-cause and NHL-specific mortality, while intermediate intake of red meat and fish was associated with lower NHL-specific mortality.
Shows an association, not proof of cause and effect.
Leo QJ, Ollberding NJ, Wilkens LR, et al. Eur J Clin Nutr, 2016. PMID 26330148
2016 / Prospective cohort study
In a prospective study of 412 infants followed at 4, 8, and 12 months, researchers examined associations between maternal pregnancy urinary arsenic and infant infections and symptoms over the first year. Each doubling of maternal urinary arsenic was related to an increase in total infections requiring prescription medication (RR = 1.1; 95% CI: 1.0, 1.2). Urinary arsenic was also related to respiratory symptoms lasting ≥ 2 days (RR = 1.1; 95% CI: 1.0, 1.2) or requiring prescription medication (RR = 1.2; 95% CI: 1.0, 1.5), and to wheezing lasting ≥ 2 days, resulting in a doctor visit, or prescription medication treatment (RR = 1.3, RR = 1.3, and RR = 1.5, respectively). Associations were also observed with diarrhea (RR = 1.4; 95% CI: 1.1, 1.9) and fever resulting in a doctor visit (RR = 1.2; 95% CI: 1.0, 1.5).
Shows an association, not proof of cause and effect.
Farzan SF, Li Z, Korrick SA, et al. Environ Health Perspect, 2016. PMID 26359651
2016 / Cross-sectional analysis
In a cross-sectional analysis of 392 men aged 30-49 years, dietary intake and targeted metabolomic plasma concentrations of amino acids were compared among meat-eaters, fish-eaters, vegetarians, and vegans. Plasma concentrations of 6 out of 21 amino acids varied significantly by diet group, with differences of -13% to +16% between meat-eaters and vegans. Methionine, tryptophan, and tyrosine were highest in fish-eaters and vegetarians, followed by meat-eaters, and lowest in vegans; lysine showed a broadly similar pattern, while alanine was highest in fish-eaters and lowest in meat-eaters, and glycine was highest in vegans and lowest in meat-eaters. Dietary intakes of all 18 amino acids differed, being lowest in vegans (up to 47% lower than in meat-eaters), but plasma concentration differences were less marked and did not necessarily mirror intakes.
Shows an association, not proof of cause and effect.
Schmidt JA, Rinaldi S, Scalbert A, et al. Eur J Clin Nutr, 2016. PMID 26395436
2016 / Cross-sectional survey
In a cross-sectional survey of 153 Italian adults following omnivore, vegetarian, or vegan diets, researchers assessed gut microbiota and metabolome in relation to habitual diet and Mediterranean diet (MD) adherence. The majority of vegan and vegetarian participants and 30% of omnivores had high adherence to the MD. Consumption of vegetable-based diets was significantly associated with higher levels of fecal short-chain fatty acids, Prevotella, and certain fiber-degrading Firmicutes. Conversely, lower adherence to the MD was associated with higher levels of urinary trimethylamine oxide.
Shows an association, not proof of cause and effect.
De Filippis F, Pellegrini N, Vannini L, et al. Gut, 2016. PMID 26416813
2016 / Cross-sectional comparative study
Investigators compared anthropometric values, nutrient intake, Healthy Eating Index, and food variety using 3-day food diaries in children with autism spectrum disorder (ASD) in Valencia, Spain (20 on a gluten-free casein-free [GFCF] diet and 85 on a regular diet). Those on the GFCF diet had lower weight, body mass index, and total energy, pantothenic acid, calcium, phosphorus, and sodium intake, but higher intake of fiber, legumes, and vegetables. The GFCF diet group also had better fat intake quality but needed vitamin D supplementation.
Shows an association, not proof of cause and effect.
Marí-Bauset S, Llopis-González A, Zazpe I, et al. J Autism Dev Disord, 2016. PMID 26428353
2016 / Calibration sub-study / cross-sectional analysis
Researchers examined the association between soya food intake (estimated via six repeated 24 h recalls over 6 months) and serum TSH concentrations in 548 women and 295 men. In men, soya isoflavone and protein intakes were not associated with high TSH (>5 mIU/l). In women, higher soya isoflavone consumption was associated with higher TSH (adjusted OR for highest vs. lowest quintile = 4.17; 95% CI 1.73, 10.06). Soya protein consumption was also associated with increased odds of high TSH in women (highest quintile midpoint 11 g/d vs. lowest quintile midpoint 0 g/d: OR = 2.69; 95% CI 1.34, 5.30).
Shows an association, not proof of cause and effect.
Tonstad S, Jaceldo-Siegl K, Messina M, et al. Public Health Nutr, 2016. PMID 26450571
2016 / Comparative risk assessment / epidemiological modeling study
Researchers updated global estimates of cancer cases and deaths attributable to alcohol consumption for the year 2012 using WHO drinking prevalence data, global cancer statistics, and meta-analysis relative risks. In 2012, alcohol-attributable cancer cases rose to approximately 770,000 globally (5.5% of total cancer cases), including 540,000 men (7.2%) and 230,000 women (3.5%). Attributable cancer deaths rose to approximately 480,000 (5.8% of total cancer deaths), comprising 360,000 men (7.8%) and 120,000 women (3.3%). Proportions were particularly high in the WHO Western Pacific, European, and South-East Asia regions.
Shows an association, not proof of cause and effect.
Praud D, Rota M, Rehm J, et al. Int J Cancer, 2016. PMID 26455822
2016 / Case-control study
A case-control study in China evaluated diet and risk factors among 600 women with uterine fibroids and 600 controls. Multifactor analysis showed that women with uterine fibroids reported less frequent intake of broccoli (OR: 0.552; 95% CI: 0.316-0.964), cabbage (OR: 0.446; 95% CI: 0.211-0.943), Chinese cabbage (OR: 0.311; 95% CI: 0.102-0.946), tomato (OR: 0.453; 95% CI: 0.241-0.853), and apple (OR: 0.416; 95% CI: 0.213-0.814) (all P < 0.05).
Shows an association, not proof of cause and effect.
Shen Y, Wu Y, Lu Q, et al. J Obstet Gynaecol Res, 2016. PMID 26458740
2016 / Pre-post evaluation of program participants
Researchers evaluated 7,172 Seventh-day Adventist (SDA, 21.3%) and non-SDA (78.7%) individuals participating in the Complete Health Improvement Program, a 30-day diet and lifestyle intervention across 241 programs from 2006 to 2012. Changes in BMI, blood pressure, pulse, lipid profile, and fasting plasma glucose were assessed. Reductions were greater among non-SDA participants for BMI, pulse, and blood lipids. Furthermore, the majority of non-SDA participants in the highest risk classifications for blood pressure, lipids, and fasting plasma glucose showed improvements of 20% or more, whereas among SDA participants this occurred only for some lipids.
Shows an association, not proof of cause and effect.
Kent LM, Morton DP, Ward EJ, et al. J Relig Health, 2016. PMID 26472654
2016 / Observational cross-sectional study
An observational, cross-sectional study evaluated 440 non-diabetic patients aged over 40 treated at cardiovascular risk units in Catalonia. Erectile dysfunction (ED) was present in 186 patients (42.3%), including 24.8% mild, 6.8% moderate, and 10.7% severe. Multivariate analysis showed that consumption of nuts > twice a week (OR 0.41, 95% CI 0.25 to 0.67) and vegetables ≥ once a day (OR 0.47, 95% CI 0.28-0.77) were inversely associated with ED. Conversely, obesity (OR 2.49, 95% CI 1.48-4.17), ischaemic disease (OR 2.30, 95% CI 1.22 to 4.33), daily alcohol consumption (OR 1.14, 95% CI 1.04 to 1.26), and age (OR 1.07, 95% CI 1.04-1.10) were directly related to ED.
Shows an association, not proof of cause and effect.
Ramírez R, Pedro-Botet J, García M, et al. Intern Med J, 2016. PMID 26482327
2016 / Prospective cohort study
Researchers followed 193,742 postmenopausal women from 1995 to 2006, identifying 9,305 incident breast cancers, to examine intakes of red meat, processed red meat, nitrite, and heme iron. Total red meat intake was positively associated with regional/distant cancer risk (HR Q5 vs. Q1 = 1.25, 95% CI: 1.03-1.52, p-trend = 0.02). Processed red meat intake was associated with a 27% higher risk of localized breast cancer (HR Q5 vs. Q1 = 1.27, 95% CI: 1.01-1.27, p-trend = 0.03), and nitrite from processed red meat was positively associated with localized cancer (HR Q5 vs. Q1 = 1.23, 95% CI: 1.09-1.39, p-trend < 0.0001). Heme iron intake was positively associated with breast cancer risk overall and across all stages (p-trend = 0.02-0.05), with no heterogeneity by hormone receptor status.
Shows an association, not proof of cause and effect.
Inoue-Choi M, Sinha R, Gierach GL, et al. Int J Cancer, 2016. PMID 26505173
2016 / Prospective longitudinal cohort study
Researchers analyzed data from 24,029 individuals aged >50 years in the Health and Retirement Study to evaluate the association between alcohol consumption and all-cause mortality over 206,966 person-years of follow-up (7,902 deaths). Drinking level was measured at 3 points over 4 years prior to follow-up, and occasional drinkers (<1 time/week) served as the reference group. In fully adjusted analyses, hazard ratios (95% CI) for mortality were 1.02 (0.94-1.11) for <7 drinks/week, 1.14 (1.02-1.28) for 7 to <14 drinks/week, 1.13 (0.96-1.35) for 14 to <21 drinks/week, and 1.45 (1.16-1.81) for ≥21 drinks/week. No level of regular alcohol consumption was linked to reduced all-cause mortality.
Shows an association, not proof of cause and effect.
Goulden R Am J Med, 2016. PMID 26524703
2016 / Prospective cohort study
Researchers evaluated the association between dietary protein intake (total, animal, and plant) and 6-year changes in grip strength among 1,746 participants in the Framingham Offspring cohort. Overall, higher protein intake was associated with less decline in grip strength (all p for trend ≤ .05), with lowest-quartile participants losing 0.17% to 0.27% per year and highest-quartile participants gaining 0.52% to 0.60% per year. Among participants aged 60 years or older (n = 646), significant protective trends against grip strength loss were observed for total and animal protein (all p for trend < .03), but not plant protein, while trends were not statistically significant in those under 60.
Shows an association, not proof of cause and effect.
McLean RR, Mangano KM, Hannan MT, et al. J Gerontol A Biol Sci Med Sci, 2016. PMID 26525088
2016 / Longitudinal cohort study
Researchers conducted a longitudinal analysis of 69 women aged 35 to 72 undergoing bariatric surgery to evaluate changes in hot flushes and vaginal dryness from preoperative to 6-month postoperative visits. The reported degree of bother of symptoms associated with hot flushes decreased from pre- to post-surgery (p < 0.01), leading authors to conclude that bother may decrease after surgery. There was no significant change in the prevalence of hot flushes or vaginal dryness in the overall study sample.
Shows an association, not proof of cause and effect.
Goughnour SL, Thurston RC, Althouse AD, et al. Climacteric, 2016. PMID 26555182
2016 / Cross-sectional general population survey
A general population survey of Australians examined the risk of non-fatal stroke or transient ischemic attack (TIA) among past-year cannabis users. Out of 7,455 participants, there were 153 stroke/TIA cases (2.1%). After adjusting for age cohort, past-year cannabis users had 3.3 times the rate of stroke/TIA (95% CI 1.8-6.3, p<0.001), which attenuated to an incidence rate ratio (IRR) of 2.3 after adjusting for stroke-related covariates including tobacco (95% CI 1.1-4.5). The elevated risk was specific to those who used cannabis weekly or more often (IRR 4.7, 95% CI 2.1-10.7), with no elevation observed for less frequent use.
Shows an association, not proof of cause and effect.
Hemachandra D, McKetin R, Cherbuin N, et al. Aust N Z J Public Health, 2016. PMID 26558539
2016 / Prospective cohort study
This prospective cohort study assessed the relationship between dietary patterns and prostate cancer incidence among 26,346 men in the Adventist Health Study-2. Over follow-up, 1,079 incident prostate cancer cases were identified, and approximately 8% of the cohort adhered to a vegan diet. Vegan diets showed a statistically significant protective association with prostate cancer risk (HR: 0.65; 95% CI: 0.49, 0.85). In race-stratified analyses, the association remained statistically significant for white participants (HR: 0.63; 95% CI: 0.46, 0.86), while black vegans had a similar point estimate that was not statistically significant (HR: 0.69; 95% CI: 0.41, 1.18).
Shows an association, not proof of cause and effect.
Tantamango-Bartley Y, Knutsen SF, Knutsen R, et al. Am J Clin Nutr, 2016. PMID 26561618
2016 / Cross-sectional twin study
Researchers examined 40 healthy monozygotic twin pairs, including 26 pairs discordant for BMI, to assess the impact of acquired obesity on adipose tissue SIRT and NAD+ pathway expression. In heavier co-twins, expression of SIRT1, SIRT3, SIRT5, NAMPT, NMNAT2, NMNAT3, and NRK1 was significantly down-regulated, and PARP activity trended higher. Controlling for twin-shared factors, SIRT1, SIRT3, NAMPT, NMNAT3, and NRK1 negatively correlated with adiposity, while SIRT1 and SIRT5 positively correlated with insulin sensitivity. Genes involved in the mitochondrial unfolded protein response were also significantly down-regulated in heavier co-twins.
Shows an association, not proof of cause and effect.
Jukarainen S, Heinonen S, Rämö JT, et al. J Clin Endocrinol Metab, 2016. PMID 26574954
2016 / Quasi-experimental policy analysis / difference-in-differences
This study evaluated the effect of restaurant trans fat bans on cardiovascular disease (CVD) mortality rates across New York State jurisdictions utilizing within-county variation and differential policy rollout timing. The results showed that the policy led to a 4.5% decrease in CVD mortality rates, equating to 13 fewer CVD deaths per 100,000 persons annually. The averted deaths were estimated to be worth approximately $3.9 million per 100,000 persons per year.
Shows an association, not proof of cause and effect.
Restrepo BJ, Rieger M J Health Econ, 2016. PMID 26620830