Sepsis – syndrome of infection complicated by organ dysfunction – is responsible for over 750 000 hospitalisations and 200 000 deaths in the USA annually. Despite potential nutritional benefits, the association of diet and sepsis is unknown. Therefore, we sought to determine the association between adherence to a Mediterranean-style diet (Med-style diet) and long-term risk of sepsis in the REasons for Geographic Differences in Stroke (REGARDS) cohort. We analysed data from REGARDS, a population-based cohort of 30 239 community-dwelling adults age ≥45 years. We determined dietary patterns from a baseline FFQ. We defined Med-style diet as a high consumption of fruit, vegetables, legumes, fish, cereal and low consumption of meat, dairy products, fat and alcohol categorising participants into Med-style diet tertiles (low: 0–3, moderate: 4–5, high: 6–9). We defined sepsis events as hospital admission for serious infection and at least two systematic inflammatory response syndrome criteria. We used Cox proportional hazard models to determine the association between Med-style diet tertiles and first sepsis events, adjusting for socio-demographics, lifestyle factors, and co-morbidities. We included 21 256 participants with complete dietary data. Dietary patterns were: low Med-style diet 32·0 %, moderate Med-style diet 42·1 % and high Med-style diet 26·0 %. There were 1109 (5·2 %) first sepsis events. High Med-style diet was independently associated with sepsis risk; low Med-style diet referent, moderate Med-style diet adjusted hazard ratio (HR) 0·93 (95 % CI 0·81, 1·08), high Med-style diet adjusted HR=0·74 (95 % CI 0·61, 0·88). High Med-style diet adherence is associated with lower risk of sepsis. Dietary modification may potentially provide an option for reducing sepsis risk.

Findings from observational studies on the associations between vitamin E intake and stroke risk remain controversial, and the dose–response relationship between vitamin E intake and risk of stroke remains to be determined. We conducted a meta-analysis of prospective studies aiming to clarify the relationships between vitamin E intake and risk of stroke. Relevant studies were identified by searching online databases through to June 2018. We computed summary relative risks (RR) with corresponding 95 % CI. Among 3156 articles retrieved from online databases and relevant bibliographies, nine studies involving 3284 events and 220 371 participants were included in the final analyses. High dietary vitamin E intake was inversely associated with the risk of overall stroke (RR=0·83, 95 % CI 0·73, 0·94), and with the risk of stroke for individuals who were followed-up for <10 (RR=0·84, 95 % CI 0·72, 0·91). There was a non-linear association between dietary vitamin E intake and stroke risk (P=0·0249). Omission of any single study did not alter the summary result. In conclusion, this meta-analysis suggests that there is a significant inverse relationship between dietary vitamin E intake and stroke risk. This meta-analysis provides evidence that a higher dietary vitamin E intake is associated with a lower stroke risk.

The prevalence of impaired fasting glucose (IFG) and diabetes mellitus (DM) has reportedly increased significantly among Chinese children and adolescents. We aimed to examine the prevalence of IFG and DM, the disparities in sex and region and related risk factors among Chinese children and adolescents. A total of 16 434 Chinese children aged 6–17 years were selected from a national cross-sectional survey, and fasting glucose was measured for all participants. Overall, mean fasting plasma glucose (FPG) concentration was (4·64 (sd 0·51)) mmol/l, and the prevalence of DM and IFG was 0·10 and 1·89 %, respectively. Compared with girls, boys had higher FPG concentration (4·69 v. 4·58 mmol/l, r 0·107, P<0·001) and IFG prevalence (2·67 v. 1·07 %, rφ 0·059, P<0·001). Compared with rural children and adolescents, urban children and adolescent had higher FPG concentration (4·65 v. 4·62 mmol/l, r 0·029, P<0·001) and DM prevalence (0·15 v. 0·05 %, rφ 0·016, P<0·01). In addition, self-reported fried foods intake and overweight/obesity were positively associated with IFG, and the proportion of consuming fried foods more than or equal to once per week and overweight/obesity prevalence in boys and urban children and adolescents were significantly higher than girls and rural children and adolescents, respectively (P<0·05). Although the prevalence of IFG and DM was relatively low in Chinese children and adolescents, sex and region disparities were observed, which may be associated with differences in overweight/obesity prevalence and dietary factors.

The Nestlé Nutritional Profiling System (NNPS) has been developed to guide food and beverage reformulation. The WHO published guidelines to develop and validate nutrient profiling systems. The objective was to conduct validation tests of the NNPS following principles of the WHO guidelines. French (Individual and National Survey on food Consumption 2006–2007) and the USA (National Health and Nutrition Examination Surveys 2011–2012) nationally representative dietary surveys were used. NNPS outcomes (PASS, FAIL, out-of-scope) of foods were compared with the validated UK Ofcom nutrient profiling system outcomes. Contributions of NNPS outcomes to energy intakes were compared between diets nutritional quality classes defined by two methods: based on a food-based quality indicator (Programme National Nutrition Santé Guideline Score in France, Healthy Eating Index 2010 in the USA) or on a combination of three nutrient-based indicators (mean adequacy ratio, mean excess ratio and energy density). In both countries, food items with a NNPS FAIL outcome had a lower nutritional quality according to the UK Ofcom, with an overall agreement between the two systems of 75·7 % in France and 68·8 % in the USA. In both countries, a high (respectively, low) contribution of NNPS PASS (respectively, NNPS FAIL) was positively associated with diet healthiness. Absolute associations were stronger between the contribution of NNPS FAIL products and measures of diet healthiness. Foods and beverages reaching NNPS standards appeared to have a higher nutritional quality and would be more likely to contribute to healthier diets, mainly linked to a reduction of nutrients to limit.

Although minerals are linked to several reproductive outcomes, it is unknown whether dietary minerals are associated with ovulatory function. We hypothesised that low intakes of minerals would be associated with an increased risk of anovulation. We investigated associations between dietary mineral intake and both reproductive hormones and anovulation in healthy women in the BioCycle Study, which prospectively followed up 259 regularly menstruating women aged 18–44 years who were not taking mineral supplements for two menstrual cycles. Intakes of ten selected minerals were assessed through 24-h dietary recalls at up to four times per cycle in each participant. Oestradiol, progesterone, luteinising hormone (LH), follicle-stimulating hormone (FSH), sex-hormone-binding globulin and testosterone were measured in serum up to eight times per cycle. We used weighted linear mixed models to evaluate associations between minerals and hormones and generalised linear models for risk of anovulation. Compared with Na intake ≥1500 mg, Na intake <1500 mg was associated with higher levels of FSH (21·3 %; 95 % CI 7·5, 36·9) and LH (36·8 %; 95 % CI 16·5, 60·5) and lower levels of progesterone (−36·9 %; 95 % CI −56·5, −8·5). Na intake <1500 mg (risk ratio (RR) 2·70; 95 % CI 1·00, 7·31) and Mn intake <1·8 mg (RR 2·00; 95 % CI 1·02, 3·94) were associated with an increased risk of anovulation, compared with higher intakes, respectively. Other measured dietary minerals were not associated with ovulatory function. As essential minerals are mostly obtained via diet, our results comparing insufficient levels with sufficient levels highlight the need for future research on dietary nutrients and their associations with ovulatory cycles.

CHD is becoming an increasing priority worldwide, as it is one of the main causes of death in low- and middle-income countries lately. This study aims to evaluate the association between beverage consumption patterns and the risk of CHD among Mexican adult population. We performed a cross-sectional analysis using data from 6640 adults participating in the Health Workers’ Cohort Study. Factor analysis was performed to identify beverage patterns using sex-specific Framingham prediction algorithms to estimate CHD risk. The prevalence of moderate to high CHD risk was 17·8 %. We identified four major beverage consumption patterns, which were categorised as alcohol, coffee/tea, soft drinks and low-fat milk. We observed a lower risk of CHD (OR=0·61; 95 % CI 0·46, 0·80; and OR=0·58; 95 % CI 0·43, 0·79, respectively) among participants in the upper quintile of alcohol or low-fat milk consumption compared with those in the bottom quintile. In contrast, a higher consumption of soft drinks was positively associated with CHD risk (OR=1·64; 95 % CI 1·21, 2·20) when compared with other extreme quintiles. Finally, coffee/tea consumption was not significantly associated with CHD risk. Our findings suggest that a beverage pattern characterised by a higher intake of sugar-sweetened beverages may be associated with an increased risk of CHD among the Mexican adult population, whereas patterns of moderate alcohol intake and low-fat milk may be associated with a reduced risk.

The aims of this study were to describe which and when food textures are offered to children between 4 and 36 months in France and to identify the associated factors. An online cross-sectional survey was designed, including questions about 188 food texture combinations representing three texture levels: purées (T1), soft small pieces (T2) and hard/large pieces and double textures (T3). Mothers indicated which combinations they already offered to their child. A food texture exposure score (TextExp) was calculated for all of the texture levels combined and for each texture level separately. Associations between TextExp and maternal and child characteristics and feeding practices were explored by multiple linear regressions, per age class. Answers from 2999 mothers living in France, mostly educated and primiparous, were analysed. Over the first year, children were mainly exposed to purées. Soft and small pieces were slowly introduced between 6 and 22 months, whereas hard/large pieces were mainly introduced from 13 months onwards. TextExp was positively associated with children’s number of teeth and ability to eat alone with their finger or a fork. For almost all age classes, TextExp was higher in children introduced to complementary feeding earlier, lower for children who were offered only commercial baby foods and higher for those who were offered only home-made/non-specific foods during the second year. Our study shows that until 12 months of age the majority of French children were exposed to pieces to a small extent. It provides new insights to further understand the development of texture acceptance during a key period for the development of eating habits.

About 57 % of the pregnant European women have 25-hydroxyvitamin D (25(OH)D) concentrations below 50 nmol/l. However, as data on the impact of gestational vitamin D deficiency on maternal and fetal health are limited, the WHO does not advocate vitamin D supplementation as part of routine antenatal care. We explored associations between first trimester maternal 25(OH)D status and childhood cognition at 5–6 years of age (n 1854, primarily Caucasian). Median serum 25(OH)D was determined at 13 (interquartile range 12–14) weeks of gestation. Childhood attention, motor fluency and flexibility and executive function were assessed using the Amsterdam Neuropsychological Tasks. Restricted cubic splines and linear regression analyses were used to analyse the data while adjusting for many maternal and child related covariates. Higher 25(OH)D status (nmol/l) was associated with better attention and executive functioning as shown by a faster reaction time (β −0·30 (sd 0·14) ms, P=0·03), faster response speed (β −0·58 (sd 0·21) ms, P=0·006), and better response speed stability (β −0·45 (sd 0·17) ms, P=0·009). No associations were observed of serum 25(OH)D with motor fluency and flexibility. Associations were most pronounced among children of African origin (n 205) as compared with those of Caucasian or another origin, for example attention (reaction time, β −2·06 (sd 0·70) ms, P=0·004) and executive function (response speed, β −1·95 (sd 0·94) ms, P=0·04). Concluding, maternal 25(OH)D status was significantly associated with childhood attention and executive function, while no associations were observed for 25(OH)D status with motor fluency and flexibility.

High protein intake in young children is associated with excess gains in weight and body fat, but the specific role of different protein sources has yet to be described. The study aimed to investigate the role of different types of protein in the post-weaning stage on weight, BMI and overweight/obesity at 60 months. Intakes of animal, dairy and plant protein and a dietary pattern characterising variation in protein types at 21 months of age were estimated using a 3-d diet diary in a cohort of 2154 twins; weight and height were recorded every 3 months from birth to 60 months. Longitudinal mixed-effect models investigated the associations between sources of protein intake or dietary pattern scores and BMI, weight and overweight/obesity from 21 months up to 60 months. Adjusting for confounders, dairy protein intake at 21 months was positively associated with greater weight (46 (95 % CI 21, 71) g and BMI up to 60 months (0·04 (95 % CI 0·004, 0·070) kg/m2) and the odds of overweight/obesity at 3 years (OR 1·12; 95 % CI 1·00, 1·24). Milk showed associations of similar magnitude. A dietary pattern low in dairy protein and high in plant protein was associated with lower weight gain up to 60 months, but not overweight/obesity. Intake of dairy products in early childhood is most strongly associated with weight gain, compared with other protein sources. A dietary pattern characterised by lower protein intake and greater protein source diversity at 2 years may confer a lower risk of excess weight gain.

The protein leverage hypothesis proposes that the need to prioritise protein intake drives excess energy intake (EI) when the dietary ratio of protein to fat and carbohydrate is reduced. We hypothesised that cats may become prone to overconsuming energy content when moderate protein diets were offered, and considered the potential influence of fat and carbohydrate on intake. To determine the effect of dietary protein and macronutrient profile (MNP) on EI, weight and body composition, cats (1–4 years) were offered food in excess of energy requirements (ER). A total of six diets were formulated, containing moderate (approximately 7 % w/w; approximately 22 % metabolisable energy (ME)) or high (approximately 10 % w/w; approximately 46 % ME) protein and varying levels of carbohydrate and fat. For 4 weeks, 120 cats were offered 100 % of their individual ER of a diet at the MNP selected by adult cats (50:40:10 protein energy ratio:fat energy ratio:carbohydrate energy ratio). EI, body weight (BW), body composition, activity and palatability were measured. Subsequently, cats were offered one of the six diets at 200 % of their individual ER for 4 weeks when measurements were repeated. Cats offered excess high protein diets had higher EI (kJ/kg) throughout, but at 4 weeks BW was not significantly different to baseline. Cats offered excess moderate protein diets reduced EI and gradually lost weight (average loss of 0·358 (99 % CI 0·388, 0·328) kg), irrespective of fat:carbohydrate and initial palatability. The data do not support the protein leverage hypothesis. Furthermore, cats were able to adapt intake of a wet diet with high protein in an overfeeding environment within 28 d.

The aim of this article was to investigate the mechanism of appetite suppression induced by high-fat diets (HFD) in blunt snout bream (Megalobrama amblycephala). Fish (average initial weight 40·0 (sem 0·35) g) were fed diets with two fat levels (6 and 11 %) with four replicates. HFD feeding for 30 d could significantly increase the weight gain rate, but feeding for 60 d cannot. Food intake of M. amblycephala began to decline significantly in fish fed the HFD for 48 d. HFD feeding for 60 d significantly reduced the expression of neuropeptide Y and elevated the expression of cocaine- and amphetamine-regulated transcript (CART), actions both in favour of suppression of appetite. The activation of fatty acid sensing was partly responsible for the weakened appetite. In addition, inflammatory factors induced by the HFD may be involved in the regulation of appetite by increasing the secretion of leptin and then activating the mammalian target of rapamycin (mTOR). Lipopolysaccharide (LPS, 2·0 mg/kg of fish weight) was administered to induce inflammation, and sampling was performed after 3, 6, 9, 12, 18, 24 and 48 h of LPS injection. Within 6–24 h of LPS injection, the food intake and appetite of M. amblycephala decreased significantly, whereas the mRNA expression of leptin and mTOR increased significantly. Our results indicate that inflammatory cytokines may be the cause of appetite suppression in M. amblycephala fed a HFD.

Genetic variants associated with dietary intake may be important as factors underlying the development of obesity. We investigated the associations between the obesity candidate genes (fat mass and obesity-associated (FTO), melanocortin-4 receptor (MC4R), leptin (LEP) and leptin receptor) and total energy intake and percentage of energy from macronutrients and ultra-processed foods before and during pregnancy. A sample of 149 pregnant women was followed up in a prospective cohort in Rio de Janeiro, Brazil. A FFQ was administered at 5–13 and 30–36 weeks of gestation. Genotyping was performed using real-time PCR. Associations between polymorphisms and the outcomes were investigated through multiple linear regression and ANCOVA having pre-pregnancy dietary intake as a covariate. The A-allele of FTO-rs9939609 was associated with a −6·5 % (95 % CI −12·3, −0·4) decrease in the percentage of energy from protein and positively associated with the percentage of energy from carbohydrates before pregnancy (β=2·6; 95 % CI 0·5, 4·8) and with a 13·3 % (95 % CI 0·7, 27·5) increase in the total energy intake during pregnancy. The C-allele of MC4R-rs17782313 was associated with a −7·6 % (95 % CI −13·8, −1·0) decrease in the percentage of energy from protein, and positively associated with the percentage of energy from ultra-processed foods (β=5·4; 95 % CI 1·1, 9·8) during pregnancy. ANCOVA results revealed changes in dietary intake from pre-pregnancy to pregnancy for FTO-rs9939609 (percentage of energy from ultra-processed foods, P=0·03), MC4R-rs17782313 (total energy intake, P=0·02) and LEP-rs7799039 (total energy intake, P=0·04; percentage of energy from protein, P=0·04). These findings suggest significant associations between FTO-rs9939609, MC4R-rs17782313 and LEP-rs7799039 genes and the components of dietary intake in pregnant women.

Nutrient profiling (NP) is a method for evaluating the healthfulness of foods. Although many NP models exist, most have not been validated. This study aimed to examine the content and construct/convergent validity of five models from different regions: Australia/New Zealand (FSANZ), France (Nutri-Score), Canada (HCST), Europe (EURO) and Americas (PAHO). Using data from the 2013 UofT Food Label Information Program (n15342 foods/beverages), construct/convergent validity was assessed by comparing the classifications of foods determined by each model to a previously validated model, which served as the reference (Ofcom). The parameters assessed included associations (Cochran–Armitage trend test), agreement (κ statistic) and discordant classifications (McNemar’s test). Analyses were conducted across all foods and by food category. On the basis of the nutrients/components considered by each model, all models exhibited moderate content validity. Although positive associations were observed between each model and Ofcom (all Ptrend<0·001), agreement with Ofcom was ‘near perfect’ for FSANZ (κ=0·89) and Nutri-Score (κ=0·83), ‘moderate’ for EURO (κ=0·54) and ‘fair’ for PAHO (κ=0·28) and HCST (κ=0·26). There were discordant classifications with Ofcom for 5·3 % (FSANZ), 8·3 % (Nutri-Score), 22·0 % (EURO), 33·4 % (PAHO) and 37·0 % (HCST) of foods (all P<0·001). Construct/convergent validity was confirmed between FSANZ and Nutri-Score v. Ofcom, and to a lesser extent between EURO v. Ofcom. Numerous incongruencies with Ofcom were identified for HCST and PAHO, which highlights the importance of examining classifications across food categories, the level at which differences between models become apparent. These results may be informative for regulators seeking to adapt and validate existing models for use in country-specific applications.

Because obesity is associated with many co-morbidities, including diabetes mellitus, this study evaluated the second-meal effect of a commercial prebiotic, inulin-type fructans, and the effects of the prebiotic on faecal microbiota, metabolites and bile acids (BA). Nine overweight beagles were used in a replicated 3×3 Latin square design to test a non-prebiotic control (cellulose) against a low (equivalent to 0·5 % diet) and high dose (equivalent to 1·0 % diet) of prebiotic over 14-d treatments. All dogs were fed the same diet twice daily, with treatments provided orally via gelatin capsules before meals. On days 13 or 14 of each period, fresh faecal samples were collected, dogs were fed at 08.00 hours and then challenged with 1 g/kg body weight of maltodextrin in place of the 16.00 hours meal. Repeated blood samples were analysed for glucose and hormone concentrations to determine postprandial incremental AUC (IAUC) data. Baseline glucose, insulin and active glucagon-like peptide-1 levels were similar between all groups (P>0·10). Glucose and insulin IAUC after glucose challenge appeared lower following the high dose, but did not reach statistical relevance. Prebiotic intervention resulted in an increase in relative abundance of some Firmicutes and a decrease in the relative abundance of some Proteobacteria. Individual and total faecal SCFA were significantly increased (P<0·05) following prebiotic supplementation. Total concentration of excreted faecal BA tended to increase in dogs fed the prebiotic (P=0·06). Our results indicate that higher doses of inulin-type prebiotics may serve as modulators of gut microbiota, metabolites and BA pool in overweight dogs.

Dementia is a leading cause of morbidity and mortality without pharmacologic prevention or cure. Mounting evidence suggests that adherence to a Mediterranean dietary pattern may slow cognitive decline, and is important to characterise in at-risk cohorts. Thus, we determined the reliability and validity of the Mediterranean Diet and Culinary Index (MediCul), a new tool, among community-dwelling individuals with mild cognitive impairment (MCI). A total of sixty-eight participants (66 % female) aged 75·9 (sd 6·6) years, from the Study of Mental and Resistance Training study MCI cohort, completed the fifty-item MediCul at two time points, followed by a 3-d food record (FR). MediCul test–retest reliability was assessed using intra-class correlation coefficients (ICC), Bland–Altman plots and κ agreement within seventeen dietary element categories. Validity was assessed against the FR using the Bland–Altman method and nutrient trends across MediCul score tertiles. The mean MediCul score was 54·6/100·0, with few participants reaching thresholds for key Mediterranean foods. MediCul had very good test–retest reliability (ICC=0·93, 95 % CI 0·884, 0·954, P<0·0001) with fair-to-almost-perfect agreement for classifying elements within the same category. Validity was moderate with no systematic bias between methods of measurement, according to the regression coefficient (y=−2·30+0·17x) (95 % CI −0·027, 0·358; P=0·091). MediCul over-estimated the mean FR score by 6 %, with limits of agreement being under- and over-estimated by 11 and 23 %, respectively. Nutrient trends were significantly associated with increased MediCul scoring, consistent with a Mediterranean pattern. MediCul provides reliable and moderately valid information about Mediterranean diet adherence among older individuals with MCI, with potential application in future studies assessing relationships between diet and cognitive function.

National dietary guidelines are directed at the general population. However, these guidelines may be perceived as unrealistic by a substantial part of the population, as they differ considerably from individual consumption patterns and preferences. Personalised dietary recommendations will probably improve adherence, and it has been shown that these recommendations can be derived by mathematical optimisation methods. However, to better account for risks and benefits of specific foods, the background exposure to nutrients and contaminants needs to be considered as well. This background exposure may come from other foods and supplements, and also from environmental sources like the air and the sun. The objective of this study was therefore to analyse the effect of including individual variation in background exposure when modelling personalised dietary recommendations for fish. We used a quadratic programming model to generate recommended fish intake accounting for personal preference by deviating as little as possible from observed individual intake. Model constraints ensure that the modelled intake meets recommendations for EPA, DHA and vitamin D without violating tolerable exposure to methyl mercury, dioxins and dioxin-like polychlorinated biphenyls. Several background exposures were analysed for 3016 Danish adults, whose food intakes and body weights were reported in a national dietary survey. We found that the lower nutrient constraints were critical for the largest part of the study population, and that a total of 55% should be advised to increase their fish intake. The modelled fish intake recommendations were particularly sensitive to the vitamin D background exposure.

Ultra-processed foods provide 58 % of energy intake and 89 % of added sugars in the American diet. Nevertheless, the association between ultra-processed foods and excess weight has not been investigated in a US sample. The present investigation therefore aims to examine the association between ultra-processed foods and excess weight in a nationally representative sample of US adults. We performed a cross-sectional analysis of anthropometric and dietary data from 15 977 adults (20–64 years) participating in the National Health and Nutrition Examination Survey 2005–2014. Dietary data were collected by 24-h recall. Height, weight and waist circumference (WC) were measured. Foods were classified as ultra-processed/non-ultra-processed according to the NOVA classification. Multivariable linear and logistic regression was used to evaluate the association between ultra-processed food consumption (% energy) and BMI, WC and odds of BMI≥25 kg/m2, BMI≥30 kg/m2 and abdominal obesity (men: WC≥102 cm, women: WC≥88 cm). Prevalence of BMI≥25 kg/m2, BMI≥30 kg/m2 and abdominal obesity was 69·2, 36·1 and 53·0 %, respectively. Consuming ≥74·2 v. ≤36·5 % of total energy from ultra-processed foods was associated with 1·61 units higher BMI (95 % CI 1·11, 2·10), 4·07 cm greater WC (95 % CI 2·94, 5·19) and 48, 53 and 62 % higher odds of BMI≥25 kg/m2, BMI≥30 kg/m2 and abdominal obesity, respectively (OR 1·48; 95 % CI 1·25, 1·76; OR 1·53; 95 % CI 1·29, 1·81; OR 1·62; 95 % CI 1·39, 1·89, respectively; Pfor trend<0·001 for all). A significant interaction between being female and ultra-processed food consumption was found for BMI (F4,79=4·89, P=0·002), WC (F4,79=3·71, P=0·008) and BMI≥25 kg/m2 (F4,79=5·35, P<0·001). As the first study in a US population, our findings support that higher consumption of ultra-processed food is associated with excess weight, and that the association is more pronounced among women.

Several descriptive studies on the intake of polyphenols, mostly flavonoids, have been published, especially in Europe and the USA, but insufficient data are still available in Latin-American countries, where different types of foods are consumed and different dietary habits are observed. The goal of this cross-sectional study was to estimate dietary intakes of polyphenols, including grand total, total per classes and subclasses and individual compounds, and to identify their main food sources in Mexican women. The Mexican Teachers’ Cohort includes 115 315 female teachers, 25 years and older, from twelve states of Mexico, including urban and rural areas. Dietary data were collected in the period 2008–2011 using a validated FFQ, and individual polyphenol intake was estimated using food composition data from the Phenol-Explorer database. Median total polyphenol intake was the highest in Baja California (750 mg/d) and the lowest in Yucatan (536 mg/d). The main polyphenols consumed were phenolic acids (56·3–68·5 % total polyphenols), followed by flavonoids (28·8–40·9 %). Intake of other polyphenol subclasses (stilbenes, lignans and others) was insignificant. Coffee and fruits were the most important food sources of phenolic acids and flavonoids, respectively. Intake of a total of 287 different individual polyphenols could be estimated, of which forty-two were consumed in an amount ≥1 mg/d. The most largely consumed polyphenols were several caffeoylquinic acids (ranging from 20 and 460 mg/d), ferulic acid, hesperidin and proanthocyanidins. This study shows a large heterogeneity in intakes of individual polyphenols among Mexican women, but a moderate heterogeneity across Mexican states. Main food sources were also similar in the different states.