The Central Statistics Office (CSO) in Ireland report that 62% of adults were overweight or obese in 2017(1). In 2019, the latest data from the national longitudinal study “Growing Up in Ireland” showed that 15% of five-year olds were overweight and 5% were obese(2). According to WHO projections, Ireland may have the highest prevalence of overweight and obesity in Europe by 2030(3). The International Agency for Research on Cancer (IARC) state there is sufficient evidence for a cancer preventative effect of the absence of excess body fatness for 13 different cancer sites(4). Recent USA based research showed that the risk of developing an obesity-related cancer seems to be increasing in a stepwise manner in successively younger birth cohorts(5), therefore we aimed to investigate trends in the age of obesity related cancer diagnosis.

National Cancer Registry Ireland (NCRI) data was obtained for 25 to 84 year olds between 1997 and 2016 for several cancer types and 5-year intervals were constructed to categorise age (25–29, 30–34 etc.) and year of diagnosis (1997–2001, 2002–2006 etc.). An age-period-cohort model was fitted to each cancer type, resulting in estimated temporal trends in incidence (expressed as annual percent change (APC) per year) and significance of the trend across age groups was tested using a Wald test(5).

Analysis of NCRI data reflect significantly increasing incidence of colorectal cancer with decreasing age in Ireland (n = 40703, p = 0.00). Despite the limitation of low numbers for several cancer sites in younger age groups, similar trends were observed for a number of cancer types including thyroid and cervical cancer (p < 0.05). Although other factors (such as early detection) may contribute to these observations, addressing the large increase in incidence of obesity and colorectal cancer among young adults is essential. Research to identify exposure to modifiable risk factors, including nutrition and physical activity, throughout life in Ireland is necessary to examine potential determinants of such trends.

Global diabetes prevalence is projected to rise from 382 million in 2013 to 592 million in 2030 with type 2 diabetes (T2D) accounting for 90–95% of cases(1,2). Approximately 4.7 million people in the UK had diabetes in 2018, this is estimated to rise to over 5 million by 2025(2) costing the NHS £15.1bn by 2035(3).

Low-fat vegan (LFV) diets are associated with improvements in T2D biomarkers(4) however, to date no systematic review has evaluated the effectiveness of LFV diets in prevention and management of T2D. The aim of this study was to assess effects of LFV diets on T2D biomarkers in controlled trials for those with, or at risk of developing T2D.

The review followed the PRISMA checklist for systematic reviews and used Cochrane Collaboration search strategies, which support the identification of randomised controlled trials (RCT's)(5). Searches were conducted and titles and abstracts screened for relevance by two independent review authors (AW or KEL). Studies that evaluated a LFV diet to reduce the risk of, or manage T2D using controlled trials were included with publication range from January 2003 to June 2018.

Search criteria identified 130 articles, which reduced to 21 upon abstract and title screening. Further application of inclusion/exclusion criteria led to identification of six key studies. Four studies were RCT's, and two were cluster-controlled trials at corporate sites, one of which was randomised. In three of the studies participants were previously diagnosed with T2D, in two of the studies participants had a BMI ≥ 25 kg/m2 and/or were previously diagnosed with T2D, and one study had participants with BMI between 28 and 40 kg/m2. Four studies had a control group (non-intervention) two of which had a conventional diabetes diet (American Dietetic Association(6)) as a comparison group. Mean study length was 28.3 weeks; ranging from 16–74 weeks. An intention-to-treat and medication adjusted analysis was carried out; significant effects for BMI, body weight, HbA1c, fasting plasma glucose, LDL, HDL and total cholesterol were reported in most of the studies.

Results from this systematic review demonstrate that a LFV diet is an effective alternative therapy to improve T2D biomarkers in those with, or at elevated risk of developing the disease. The LFV diet showed consistent positive effects on T2D biomarkers. Its effects are comparable to conventional T2D diets and energy restrictive diets, despite absence of restrictions on energy intake or portion sizes.

Iodine is essential in foetal development through being an integral part of the thyroid hormones. Severe iodine deficiency is associated with foetal growth restriction and preterm delivery. Less is known about the potential impact of mild-to-moderate iodine deficiency on these outcomes.

The aim of this study was to investigate whether maternal iodine intake in pregnancy was associated with birth weight (BW) z-score (i.e. BW adjusted for gestational length and sex) and preterm delivery (before week 37).

The study population included 77,995 singleton pregnancies from The Norwegian Mother and Child Cohort Study recruited in gestational week 15 in the period 2002–2008. Habitual iodine intake was calculated from a validated food frequency questionnaire covering the first half of pregnancy. Use of supplements was reported in questionnaires. Urinary iodine concentration (UIC) was measured in gestational week 18 in a subsample of n = 2795 women. Median iodine intake from food was 121 μg/day and median UIC was 69 μg/L. Median UIC < 150 μg/L is considered insufficient in pregnant women. Median birthweight was 3610 g and 5.0% were born before gestational week 37. Associations were modelled flexibly by use of restricted cubic splines, and adjusted for age, parity, pre-pregnancy BMI, education, smoking in pregnancy, energy intake, and fibre intake.

In non-users of iodine-containing supplements (n = 48,958), a low habitual iodine intake from food (lower than about 150 μg/day) was associated with a lower mean BW z-score (p < 0.001). Compared to an intake of 150 μg/day (reference), mean z-score was 0.04 SD lower at 100 μg/day and 0.12 SD lower at 75 μg/day. Results were similar when using UIC as the exposure (n = 2795, p = 0.017). Any use of iodine containing supplements in pregnancy was associated with 0.03 (95% CI: 0.01, 0.04) SD increase in BW z-score compared to no use (n = 77,949, p < 0.001).

A low habitual iodine intake from food (lower than about 100 μg/day) was associated with increased risk of preterm delivery (p = 0.003). Compared to an intake of 100 μg/day (reference), 75 μg/day was associated with 10% increased risk, and 50 μg/day with 28% increased risk. Use of an iodine-containing supplement was not associated with the risk of preterm delivery (OR: 0.97 (95%CI: 0.91, 1.04, p = 0.42)).

Inadequate iodine intake is prevalent in women of childbearing age in otherwise well-nourished populations. Our results indicate that mild-to-moderate iodine deficiency in pregnancy is associated with restricted foetal growth and increased risk of preterm delivery.

Breast cancer is the most common cancer among women worldwide. Diet and inflammation have been associated with carcinogenesis. However, there are limited studies regarding both blood levels of inflammatory markers and diet with respect to the risk of breast cancer. We assessed the association between serum inflammatory profiles and breast cancer (BC) risk with including dietary patterns (DPs). This case-control study involved 420 women aged 40–79 years from north-eastern Poland, including 190 newly-diagnosed breast cancer cases. The serum concentrations of C-reactive protein (CRP), interleukin-6 (IL-6) and leukocytes (including neutrophil and lymphocyte count) were marked in 129 post-menopausal women (82 controls, 47 cases) using a Cobas Integra 400plus auto-analyser, an immune-analyser Cobas e411 (Roche Diagnostics®) and haematology analyser MEK-7300 (Nihon Kohden®), respectively. A posteriori inflammatory profiles were derived with a Principal Component Analysis (PCA). A logistic regression analysis was performed. The odds ratios (ORs) and 95% confidence interval (95% CI) were calculated. The ORs were adjusted for: age, BMI, socioeconomic status, overall physical activity, smoking status, abuse of alcohol, age at menarche, number of children, oral contraceptive use, hormone-replacement therapy use, family history of breast cancer, vitamin/mineral supplements use, hormone receptor status of breast cancer and PCA-driven DPs (‘Non-Healthy’, ‘Prudent’, and ‘Margarine and Sweetened Dairy’) score. Two serum inflammatoryprofiles were identified. The ‘CRP-IL-6’ profile was loaded heavily by the interleukin-6 (factor loading 0.78) and C-reactive protein (factor loading 0.75). The ‘Neutrophil-Lymphocyte’ profile was loaded heavily by the lymphocyte (factor loading 0.86) and neutrophil count (factor loading 0.70). The risk of BC was six times higher (OR = 6.05; 95%Cl: 1.93–18.91; p < 0.05) in the third tertile of serum ‘CRP-IL-6’ profile compared to the first or second tertile. The risk of BC was three times higher (OR = 3.11; 95%Cl: 1.05–9.20; p < 0.05) at the level of serum IL-6 > 3.10 pg/mL and seven times higher (OR = 7.48; 95%Cl: 2.31–24.29; p < 0.05) at the level of serum neutrophil count > 3.90 103/μL. No significant association between BC risk and serum ‘Neutrophil-Lymphocyte’ profile and also CRP or lymphocyte count considered separately was revealed. Concluding, the elevated serum levels of both C-reactive protein and interleukin-6, as well as elevated levels of interleukin-6 and neutrophil count considered separately, were associated with higher risk of postmenopausal breast cancer, independently of dietary patterns. Therefore, women should be screened for blood concentrations of multiple potential pro-inflammatory markers in the breast cancer prevention.

A cancer diagnosis serves as a teachable moment for patients to consider lifestyle changes to cope with treatment and for future health. Diet and nutrition is an important, yet neglected aspect of survivorship care. Patients diagnosed with a cancer in the pelvis (anal, bladder, rectal and cancers of the reproductive organs) may benefit from dietary modifications that could improve quality of life and may influence survival. This project aims to explore pelvic cancer patients’ dietary habits and experiences of nutrition support in cancer care.

People who have been diagnosed with a pelvic cancer and either undergoing (n = 266) or having completed radiotherapy treatment 6–24 months previously (n = 405) were invited to fill in a survey about their current dietary habits and nutritional support received after diagnosis. Data were analysed using descriptive statistics and logistic regression.

Respondents [n = 251; mean (SD) age 68.8 (10.5) years] reported high obesity rates (63% classified as overweight or obese) and presence of treatment side effects (82%) and comorbidities (75%). There was no significant difference in obesity rates, presence of side effects or comorbidities between on-treatment and post-treatment patients (P > 0.01 for all comparisons). Two thirds of respondents (n = 170) reported at least one dietary change since diagnosis and 31% (n = 77) reported supplement use. However, the number of respondents who reported a change per food group was low. Most notable changes included increased intake of vegetables (33%), fruit (33%) and oily fish (29%) and reduction of sugary foods (48%), alcohol (41%) and processed meats (40%). Forty-three percent reported receipt of dietary advice from the healthcare team; mainly leaflets from cancer specialist nurses. Receipt of dietary support from the healthcare team was a significant predictor of dietary change (OR 4.02, 95% CI: 2.01–8.05). Thirty-two percent of respondents (n = 79) reported looking for advice from other sources, with the internet being the most popular (66% of them looked at medical and/or charity websites). Survivors were interested in receiving dietary support in relation to weight management, management of side effects and future health; however, they were uncertain of the benefits of diet in improving low appetite and the feeling of fatigue.

Results from this study highlight an increased need for dietary support in pelvic cancer populations post-treatment. Emphasis should be given to the role of the healthcare team in the delivery of nutritional information to influence patients make healthy dietary choices.

The world's population is getting older. By 2050 it is projected that 22% of the global population will be over 60 years of age. Aging is characterized by a decline in physiological functions and an increase in malnutrition and associated disease. Oxidative stress is associated with several age-related conditions, including cardiovascular disease, obesity, neurodegenerative disease and cancer. The negative effects of reactive oxygen species (ROS) are neutralized by antioxidants, which can be endogenously synthesised or consumed through the diet. Cheese is a nutritious dairy food and previous research has highlighted the antioxidant potential of some cheese types. Physiological changes associated with age can impact digestion processes, thus the ability to efficiently release bioactive compounds, including antioxidants, may be impaired with age. Cheese products varying by milk source, texture and fat content were subjected to simulated gastrointestinal in vitro digestion (SGID) using a modified model to account for physiological changes associated with aging. The antioxidant potential of undigested and digested cheese (n = 11) was determined in vitro using DPPH radical scavenging, ferric reducing antioxidant power (FRAP) and total phenolic content (TPC) assays. Cellular antioxidant potential was also investigated using the reduced glutathione assay (GSH) in the human macrophage cell line (U937). Experiments were carried out in triplicate. Statistical analysis by T-test demonstrated that digestion significantly increased (P < 0.05) FRAP and TPC of all cheese products analysed. Of the eleven cheese products investigated, goats’ cheese displayed a significant increase (P < 0.05) in radical scavenging properties (89.8% inhibition). Low fat white cheddar had significantly higher (P < 0.05) GSH levels compared to full fat white cheddar (0.16 ± 0.01, 0.11 ± 0.01 μmol GSH/mg protein, respectively). Findings from this study suggest that despite physiological changes associated with age, cheese digestates retain, and in some cases, possess higher antioxidant potential than undigested samples. Cheese demonstrates potential as a functional food for the elderly by neutralizing the negative effects associated with age-related oxidative stress.

The growing problem with the irreproducibility of the results of metabolic research urged us to analyse potential critical factors contributing to the disparities. Microbiota composition constitutes crucial variable in animal research. Gut flora changes depending on multiple factors including fibre intake. Mice derive fibre from diet but, especially under fasted conditions, also from bedding ingestion. Our project aims at assessing the impact of cage bedding on several aspects of metabolic research. Mice were divided into an overnight (ON) fasted, caloric restricted (CR) and ad libitum groups. In order to evaluate the effects of different beddings, mice were housed in cages with the most commonly used beddings: wooden, corncob or cellulose. Additionally, ON-fasted mice were kept without bedding or on metal grid preventing coprophagy. Mice bedding consumption and faeces production were measured. Oral glucose tolerance test (OGTT) was performed. Metabolites composition and microbial diversity were analysed in caecum content. Mice organ weight and gene expression measurements were conducted. Mice show preference towards wooden bedding and consume it in highest amount followed by corncob and cellulose beddings. As a result of the bedding ingestion, caecum weight increases in the CR compared to ad libitum fed mice. CR leads to a reduced faeces production and it is the lowest in the CR wood and CR cellulose groups. Housing on metal grid results in greater body weight loss and decreases stomach and caecum size. ON-fasted mice housed without bedding or on grids show lower glucose levels when submitted to OGTT. CR influences the caecal bacterial diversity as well as the caecal metabolic profile. In the CR group, mice on the corncob bedding show a distinct caecal microbial composition and caecal metabolic profile compared to wooden and cellulose groups. CR is associated with increased abundance of Marinifilaceaefrom the phylum Bacteroidales, and Erysipelotrichaceaefrom the phylum Firmicutes and with a decreased abundance of Lachnospiraceaefrom the phylum Firmicutes. The main metabolites effected by CR are organic acids, fructose, sucrose and glutamine. The type of bedding exerts the strongest effect on valeric acid, lysine, threonine and organic acids. CR- and ON-fasting-associated weight loss, bedding consumption and fibre accumulation in the caecum depends on the bedding type. CR mice housed on wooden and cellulose beddings show similarities and are distinguishable from CR corncob group. Thus, depending on the composition and consumed amount, bedding intake affects numerous parameters and shapes the microbiome.

Strong evidence exists linking poor diet to increased risk of overweight, obesity and non-communicable diseases. Reformulation initiatives, whereby the food industry gradually reduces energy, sodium, sugar, fat and saturated fat content of their products, are identified as important strategies to improve dietary intakes. The analysis aimed to examine the impact of voluntary reformulation, changes to products on the market and changes in consumer preferences on dietary intakes in Ireland from 2005 to 2017. Data on composition and volume sales (kg/ year) of products available on the market in 2005 and 2017 were collected from 15 Food Drink Ireland (FDI) member companies via online templates. These products were assigned to appropriate food and beverage groups identified in four Irish University Nutrition Alliance (IUNA) surveys of preschool children (1–4 years), children (5–12 years), teenagers (13–17 years) and adults (18–90 years). Assignment of FDI products to IUNA foods and beverages was carried out using weighted distributions for a given group of foods. The weightings were taken from the sales volumes of similar products relative to one another in a given category in a given year. Monte Carlo simulations were used to run the IUNA survey consumption data with both sets of weighted composition data from 2005 and 2017. The Creme Global intake model was used to estimate daily energy and nutrient intakes for all four populations during 2005 and 2017. The Wilcoxon-signed rank test was used to test for differences between the two years. Changes in both the products available on the market and market share of these products were observed from 2005 to 2017. The nutrient with the greatest intake reduction between the two years for all ages was sugar. Children and teens were the most affected, where total sugar intakes reduced by 3.2g/d and 2.7g/d, respectively. This reduction was primarily driven by the beverage category. There were modest saturated fat intake reductions observed for teens and adults (0.2g/d and 0.5g/d, respectively). Energy, total fat and sodium intakes for all ages remained relatively stable between the two years. This analysis highlights the impact of not only food industry efforts but also consumer choices on nutrient intakes in Ireland. It is worth noting that the data collected predates the sugar tax on sugar-sweetened beverages in Ireland. Reductions in sugar intakes were not compensated by total fat or energy increases.

There is a concern that the bad habits and lifestyles of adolescents such as the consumption of foods rich in sugars and fats in free moments, like school recess acts as a factor that intervenes in the increase of weight. The aim of this study was to determine the relationship between the nutritional status and the food consumption inside the school in adolescents of Mexico City. A sample of 1457 first-grade adolescents from high schools was obtained. A food consumption frequency questionnaire was applied inside the school. Likewise, anthropometric measures were taken and through the use of Who Anthro Plus program, the nutritional status diagnosis was obtained. The obtained data was analyzed with the statistical package IBM SPSS Statistics® version 20.0 for Windows. It was found that adolescents who frequently consume sweets, snacks, flavored waters, bottled juices, confectionery products and ice cream have a higher prevalence of obesity (15.2%, 17.4%, 21.4%, 17.7%, 16.2% and 19.9%, respectively), compared to adolescents who never consume these products (11.6%, 12.6%, 11.9%, 10.5%, 14.6%, and 12.1%, respectively). Likewise, it was found that adolescents who frequently consume fruit, water, yogurt, vegetables and milk have a lower prevalence of obesity (10.2%, 12.2%, 10.3%, 10.1% and 9.35, respectively) in comparison with adolescents who never consume these foods during school hours (19.1%, 18.6%, 17.8%, 15.9% and 16.3%, respectively), finding a highly significant difference (p < 0.01) in the consumption of flavored water, fruit, natural water and yogurt. It was found that the food consumed during school hours has an impact on the nutritional status of adolescents, similar to the results found in the literature, so it is suggested to promote healthy habits and lifestyles at school and at home so that adolescents opt for healthy snack choices.

Short dietary assessment instruments are often included to large-scale twin studies that focus on multiple health behaviors and disease outcomes. Due to their feasibility and low cost, food frequency questionnaires (FFQs) are usually employed to provide data on dietary intake and diet quality. Diet quality have been linked with eating styles such as meal and snacking frequency. However, little is known about how diet quality is related with various eating styles. We aimed to construct a food-based diet quality score (DQS) from a short FFQ to estimate diet quality and to examine its association with eating styles and nutrient intakes in Finnish twins. Participants were 3592 twin individuals (764 dizygotic [DZ] pairs and 430 monozygotic [MZ] pairs) aged 31–37 from the wave 5 of the FinnTwin16 study. The DQS (range 0–12 points) was constructed from a short FFQ that included dark bread, white bread, fruits and berries, vegetables, fish, whole grain, fast food, fat free or reduced-fat milk or milk products, sugar-sweetened soft drinks or juices, energy drinks, butter and margarine/vegetable oils. Eating styles were self-reported by questionnaire. Nutrient intakes were calculated from food diaries completed in a subsample of 249 twin individuals (45 same-sex DZ pairs and 60 MZ pairs). Data were analyzed using Pearson correlations and multiple regression models adjusted for covariates in individuals and pairwise analyses. Both in individuals and within twin pairs, a higher diet quality was associated with health-conscious eating, having breakfast, fewer evening meals, and a higher frequency and regularity of eating. No associations were found with night eating styles, food as a reward or food restricting/overeating. Among individuals, the DQS was positively correlated with the intakes of protein, fiber and magnesium and negatively with the intakes of total fat, saturated fat and sucrose (rmax = 0.30 for magnesium). The associations with saturated fat, fiber and magnesium remained significant within same-sex DZ twin pairs (rmax = 0.40 for magnesium) and with sucrose within MZ twin pairs (r = -0.42). The DQS is an easy-to-use short tool for ranking individuals and twin pairs according to overall diet quality. The DQS reflects eating styles and intakes of key nutrients related to diet quality.

A high sugar intake is increasingly discussed to affect health. Preceding analyses based on dietary records, collected between 1985–2016, suggest a decreasing time trend in total [TS], added [AS] and free [FS] sugar intake (calculated as % of energy intake [%E]) since 2005 - most notably from 2010 onwards - among children and adolescents in Germany. While TS [%E] intake decreased with age, the youngest children had the lowest AS [%E] intake and the oldest children had the lowest FS [%E] intake, with minor differences in the other age groups. However, self-reported data is likely introducing bias due to selective underreporting, especially of sugar-rich foods, which can distort results. Therefore, we analysed time and age trends in dietary sugar intake using urinary biomarkers.

Urinary fructose excretion (FE), sucrose excretion (SE) and the sum of both (FE + SE) as predictive biomarkers for sugar intake was measured by UPLC-MS/MS in 997 24-h urine samples from the DONALD (Dortmund Nutritional and Anthropometric Longitudinally Designed) study cohort collected from 8.5–16.5-year-olds (239 boys, 253 girls) between 1990–2016. Time and age trends of log-transformed FE, SE and FE + SE were analyzed using polynomial mixed-effects regression models. Sex, creatinine and urea excretion, urine volume, total daily energy intake, overweight status, type of weekday, maternal overweight and educational status as well as maternal employment were considered as potentially confounding factors.

The mean (± SD) FE was 26.1 ± 20.9 mg/d, SE 33.4 ± 38.3 mg/d, and FE + SE 59.6 ± 49.1 mg/d. While FE increased significantly with age (linear trend: p = 0.0147), there were no significant age trends for SE and FE + SE in adjusted models. Between 1990–2016 FE as well as FE + SE decreased significantly (linear trend: p = 0.0280 and p < 0.0001, respectively). SE showed a tendency towards a negative quadratic time trend (p = 0.0574).

The decreasing time trend of urinary sugar excretion supports our previous observed time trends in dietary sugar intake based on dietary records among children and adolescents. In contrast, the observed age trend in dietary sugar was not confirmed by the biomarker trend analysis, suggesting increasing underreporting with age during adolescence. Although predictive biomarkers do not allow conclusions on the amount of the respective dietary intake, the combination of biomarker and dietary record data points towards the need of further public health measures to support the observed decline in sugar intake among children and adolescents in Germany.

The project is funded by the German Federal Ministry of Food and Agriculture.

Native maize is the main cereal in the diet in rural communities in México, especially in those with a traditional maize-based food-pattern. Although maize has a high nutritional value, its consumption in rural communities has declined in recent decades. Traditional food replacement by modern and processed foods seems to adversely affect the composition of the diet. The aims of this study were: to evaluate the consumption of maize, its influence on nutrient intake in Matlatzincas women, and to analyze the association between the intake of maize food products with specific nutrients within the context of the most current dietary pattern in this indigenous group from central México. In this longitudinal study, we assessed diet over a one-year period, using eight 24-h dietary recalls and a food frequency questionnaire in a rural representative sample of 92 indigenous women aged 19 to 90 years. We derived dietary patterns using principal component analysis based on the intake of ten predefined food groups according to the Mexican equivalents food system. A specific food group based on maize products (maize-food-group) such as tortilla, sopes, pozole, atole, pinole, tacos, etc., was created to identify separately maize consumption in the habitual diet. The percentage of daily energy intake attributed to maize-food-group as well as the intake of macro and micronutrients were estimated. Based on the nutrimental composition of each food group, we analyzed habitual intakes of energy (kcal/d), carbohydrates, proteins, lipids, fiber, vitamins A, D, C and B12 (mg/d), and micronutrients: phosphorus, calcium, iron and zinc (mg/d). Analysis of variance and Pearsońs correlation test were used to evaluate the association between the tertiles of the consumption of maize-food-group and nutrient intake. Additionally, a diet score was calculated to assess diet quality according to maize-food-group consumption. All dietary data were energy-adjusted for the analysis. The consumption of maize-based-foods corresponded to 23.0% of the total caloric intake. Consumption of maize was higher among older women. Consumption of maize-food-group was positively correlated with the ingest of proteins, lipids, dietary fiber, calcium, phosphorus, iron, B12 vitamin, D vitamin and energy (p < 0.05). Better diet quality was associated with a high consumption of maize-food-group. We conclude that the high consumption of maize food products and its positive correlation with the intake of recommended nutrients for healthy diet in adult women, needs interventions in public health to promote a better diet-quality aimed to increase the intake of maize-based food group.

Pesticides are used in large quantities in current agriculture all over Europe. Some pesticides are suspected of having negative effects on human health. While the risk associated with occupational exposure is getting well-documented, evidence of dietary exposure to these molecules in the general population is lacking. The aim of this study is to describe dietary exposure to pesticides among NutriNet-Santé cohort participants’.

NutriNet-Santé is a web-based cohort of French middle-aged adults launched in 2009. BioNutriNet is an ancillary study launched in 2014, aiming to measure organic and conventional food consumption and evaluate different dimensions of diet sustainability. Dietary intake of organic and conventional foods was assessed using a 264 items self-administered semi-quantitative food frequency questionnaire. Exposure to 25 commonly used pesticides was evaluated using contamination data from CVUA Stuttgart accounting for farming system. Dietary pesticide exposure profiles were identified using Non-Negative Matrix factorization (NMF), specially adapted for non-negative data with excess zeros. The NMF scores were then introduced in a hierarchical clustering process. Identified clusters were described in terms of sociodemographics, dietary patterns and exposure to pesticides.

A total of 34,442 participants were selected. Four clusters were identified. All clusters seemed to be exposed to the same molecules, with gradual intensity. The first cluster was characterized by the lowest energy-intake, highest organic food consumption (26.9%) and lowest dietary pesticide exposure; participants had highest consumption levels of organic food groups, including fruits and vegetables. Proportion of male participants and of rural residents was higher than in other groups. Clusters 2 and 3 were characterized by intermediate energy intake, and low organic food consumption. High conventional fruits and vegetables intake and moderate pesticide exposure were observed in cluster 2, composed of a lower smokers’ proportion. Cluster 3 individuals were less exposed to pesticides than clusters 2 and 4, except for Spinosad and Pyrethrins. Cluster 4 was characterized by the highest energy intake, lowest organic food consumption (12.2%), and highest pesticide exposure ; high consumption levels of conventional fruits, vegetables and a higher proportion of vegetarians were observed. This cluster was composed of 85.6% female participants, mainly retired (47.5%), mostly living in cities of 200 000 inhabitants and more (48.2%), and physically active.

Dietary pesticide exposures seem to vary across the clusters and depending on the proportion of organic food in the diet. High consumers of conventional fruits and vegetables consumers seem to be more exposed to our selection of pesticides.