It has been speculated that vegetarians or vegans may have higher risks of fractures than meat eaters, but there is limited evidence from prospective cohorts. We aimed to assess the risks of total and site-specific fractures in people of different diet groups, in a prospective cohort with a large proportion of non-meat eaters.

In EPIC-Oxford, dietary information was collected at baseline (1993–2001) and at follow-up around 14 years later (≈2010). Participants were categorised into five diet groups (≈20,106 regular meat eaters: ≥ 50 g of meat per day, ≈9,274 low meat eaters: < 50 g of meat per day, ≈8,037 fish eaters, ≈15,499 vegetarians and ≈1,982 vegans, with minor variations in numbers for each outcome after pre-specified exclusions) at both time points. Using multivariable Cox regression adjusted for socio-demographic, lifestyle, and physiological confounders, we estimated the risks of total and site-specific fractures (arm, wrist, hip, leg, ankle, and other main sites i.e. clavicle, rib and vertebra) in the different diet groups, with outcomes identified through record linkage.

Over an average of 17.6 years of follow-up, we observed 3,941 cases of total fractures, 566 arm fractures, 889 wrist fractures, 945 hip fractures, 366 leg fractures, 520 ankle fractures, and 467 other main site fractures. Compared with meat eaters, vegetarians had marginally higher risks of total fractures (hazard ratios and 95% confidence intervals: 1.10; 1.00–1.20) and arm fractures (1.28; 1.01–1.63), while vegans had significantly higher risks of total fractures (1.44; 1.21–1.72) and leg fractures (2.06; 1.22–3.47), and marginally higher risks of arm fractures (1.60, 1.01–2.54). For hip fractures, the risks were higher in fish eaters (1.28; 1.03–1.59), vegetarians (1.27; 1.05–1.55) and vegans (2.35; 1.67–3.30, p-heterogeneity < 0.0001) than regular meat eaters. There were no significant differences in risks of wrist, ankle or other main site fractures by diet groups. Overall, the significant associations appeared stronger without adjustment for body mass index (e.g. 1.52; 1.27–1.81 in vegans for total fractures), and were slightly attenuated with additional adjustment for total protein (1.41; 1.17–1.69) or dietary calcium (1.32; 1.10–1.59).

In conclusion, non-meat eaters, especially vegans, had higher risks of either total or some site-specific fractures, particularly hip fractures. The higher risks might be partly explained by the lower body mass index in these diet groups, but differences in dietary intakes of protein and calcium are likely relevant as well. Given the observational design of this study, causality and potential mechanisms should be further investigated.

Meat intake is thought to play a role in the risk of cancer. The Third Expert Report of the World Cancer Research Fund/American Institute for Cancer Research concluded that red meat was a probable cause and processed meat a convincing cause of colorectal cancer. However, evidence for associations between red and processed meat intake and other cancer sites is limited. Furthermore, few studies have investigated the association between poultry intake and cancer risk. Therefore, this study aimed to examine the associations between red, processed meat and poultry intake and incidence for 20 common cancer sites.

We analysed data from 475,264 participants (54 % women) in UK Biobank. Participants were aged 37–73 years and cancer free at baseline. Cancer diagnosis and death due to cancer without prior diagnosis during follow-up were determined using data-linkage with cancer and death registries (with follow-up until 31 March 2016 for England and Wales and until 31 October 2015 for Scotland, respectively). Information on meat consumption was based on a touchscreen questionnaire completed at baseline covering type and frequency of meat intake. We used multivariable-adjusted Cox proportional hazards models to determine the association between baseline meat intake and cancer incidence. Analyses of lung cancer risk were restricted to never smokers. All analyses were adjusted for socio-demographic, lifestyle and women-specific factors.

Over a mean 6.9 (SD 1.3) years of follow-up, 28,431 participants were diagnosed with any type of cancer. Red meat intake was positively associated with risk for colorectal cancer (n cases = 3,164; Hazard ratio (HR) per 50 g/day higher intake 1.22, 95% Confidence Interval (CI) 1.05–1.41), breast cancer (n cases = 5,536; 1.12, 1.01–1.24) and prostate cancer (n cases = 5,807; 1.16, 1.03–1.30). Processed meat intake was positively associated with risk for colorectal cancer (n cases = 3,189; HR per 20 g/day higher intake 1.17, 95% CI 1.06–1.30). Poultry intake was positively associated with risk for cancers of the lymphatic and hematopoietic tissues (n cases = 2,431; HR per 30g/day increment in intake 1.16, 95%-CI 1.03, 1.32).

In summary, higher intakes of red and processed meat were associated with a higher risk of colorectal cancer. Red meat consumption was also positively associated with risk of breast and prostate cancer, but these associations are not supported by most previous prospective studies. The positive association of poultry intake with cancers of the lymphatic and hematopoietic tissues requires further investigation.

The evidence of associations between individual foods and dietary fibre with subtypes of stroke (ischaemic and haemorrhagic) is not conclusive. We aimed to investigate this in a large prospective cohort.

We analysed data on 418,329 men and women from the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Consumption of various animal-sourced foods (red and processed meat, poultry, fish, dairy, egg), plant-sourced foods (fruit and vegetables, legumes, nuts and seeds) and dietary fibre was assessed using validated country-specific questionnaires, calibrated with 24-hour recalls. Using multivariable Cox regressions adjusted for energy intake and socio-demographic, lifestyle and physiological confounders, we estimated hazard ratios of fatal and non-fatal ischaemic, haemorrhagic and total (i.e. ischaemic, haemorrhagic and unspecified) stroke associated with calibrated increment differences in consumption of each food or dietary fibre.

Over an average of 12.7 years of follow-up, we observed 4281 cases of ischaemic stroke, 1430 cases of haemorrhagic stroke, and 7378 cases of total stroke. For ischaemic stroke, lower risks were observed with higher consumption of fruit and vegetables (hazard ratio (HR); 95% confidence interval (CI) for per 200g/d of calibrated intake, 0.87; 0.82–0.93) and dietary fibre (per 10g/d, HR 0.77; 95% CI 0.69–0.86) (p-trend < 0.001 for both); more modest inverse associations were also observed for milk (per 200g/d, HR 0.95; 95% CI 0.91–0.99, p-trend = 0.02), yogurt (per 100g/d, HR 0.91; 95% CI 0.85–0.97, p-trend = 0.004) and cheese (per 30g/d, HR 0.88; 95% CI 0.81–0.97, p-trend = 0.008), while a modest positive association was observed with higher red meat consumption (per 50g/d, HR 1.14; 95% CI 1.02–1.27, p-trend = 0.02). For haemorrhagic stroke, higher risk was associated with higher egg consumption (per 20g/d, HR 1.25; 95% CI 1.09–1.43, p-trend = 0.002). For total stroke, associations were consistent with those of both subtypes; we observed inverse associations for fruit and vegetables (HR 0.89, 95% CI 0.85–0.93), dietary fibre (HR 0.80, 95% CI 0.74–0.86), yogurt (HR 0.91, 95% CI 0.87–0.96), cheese (HR 0.88, 95% CI 0.82–0.94), and positive associations for red and processed meat (HR 1.18, 95% CI 1.05–1.33) and egg (HR 1.07, 95% CI 1.01–1.14).

To conclude, risk of ischaemic stroke was inversely associated with consumption of fruit and vegetables, dietary fibre and dairy foods, and positively associated with red meat, while risk of haemorrhagic stroke was positively associated with egg consumption. Causality of the associations cannot be determined in this observational study.

There is evidence that plant-based diets might be associated with a lower risk of IHD; however, previous studies have not reported on intake of subtypes of fruit and vegetables and sources of dietary fibre. This study aims to assess the associations of major plant foods, their subtypes and dietary fibre with risk of ischaemic heart disease (IHD) in the European Prospective Investigation into Cancer and Nutrition (EPIC)-CVD Consortium.

We conducted a prospective analysis of 490,311 men and women in ten European countries without a history of myocardial infarction or stroke at recruitment. Dietary intake was assessed using validated questionnaires and calibrated with 24-hour recall data. Cox regression models, adjusted for IHD risk factors, were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).

During a mean of 12.6 years follow-up, we documented 8504 myocardial infarction cases or deaths from IHD. Participants consuming at least eight portions (80 grams each) of fruits and vegetables a day had a 10% lower risk of IHD (HR 0.90, 95% CI: 0.82–0.98) compared with those consuming fewer than three portions a day. The risk of IHD was 6% (95% CI 0.90–0.99; P-trend = 0.009) lower for a 200 g/day higher intake of fruit and vegetables combined, 3% (0.95–1.00; P-trend = 0.021) lower for a 100 g/ day higher fruit intake, and 8% (0.86–0.97; P-trend = 0.006) lower for a 50 g/ day higher intake of bananas. Moreover, risk of IHD was 9% (0.83–0.99; P-trend = 0.032) lower for a 10g/ day higher intake of nuts and seeds, and 10% (0.82–0.98; P-trend = 0.020) lower for a 10g/ day higher intake of total dietary fibre. No associations were observed between legumes, total vegetables and other subtypes of fruit and vegetables and IHD risk.

The results from this large prospective study suggest that higher intakes of fruit and vegetables combined, total fruit, bananas, nuts and seeds, and total fibre are associated with a lower risk of IHD. Given the observational design of this study, causality and potential mechanisms should be further investigated.

Prebiotics are a subtype of dietary fibre selectively fermented by beneficial bacterial in the colon. Preclinical evidence has suggested that prebiotics may be associated with a decreased risk of colorectal cancer. However, the association between dietary intake of prebiotics and colorectal cancer risk has not been investigated prospectively. This study aims to prospectively investigate the association between total prebiotic intake and colorectal cancer risk. Further characterisation of the association by prebiotic sub-type (fructans and galacto-oligosaccharides (GOSs)) and colorectal cancer sub-site (colon cancer and rectal cancer) were secondary objectives.

A total of 53,700 men and women living in England and Scotland who were enrolled in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Oxford study, were included in the analysis and followed up for incident colorectal cancers. Validated semi-quantitative food frequency questionnaires administered at baseline were used to calculate daily fructan, GOS and total prebiotic intake. We used multivariable Cox proportional hazards models to assess associations between prebiotic intake and risk of colorectal cancer.

A total of 574 incident cases of colorectal cancer were identified during a mean of 16.1 years of follow-up. Total prebiotic, fructan and GOS intake were not significantly associated with colorectal cancer risk. The hazard ratios for those in the highest fourths of total prebiotic, fructan and GOS intake compared to those in the lowest fourths were 0.87 (95% confidence intervals (CI) 0.66–1.14; P for trend = 0.3), 0.91 (95% CI 0.70–1.18; P for trend = 0.4), and 0.87 (95% CI 0.66–1.15; P for trend = 0.4) respectively. The associations remained nonsignificant when colorectal cancer sub-sites were investigated separately.

The results from this observational study do not support an association between prebiotic intake and colorectal cancer risk. Given the biological plausibility of a role for prebiotics in reducing colorectal cancer risk and since the non-significant association between prebiotic intake and colorectal cancer risk observed in the current study may be due to the small number of cases and the healthy profile of the cohort, further epidemiological research is needed to characterise the association between dietary prebiotic intake and colorectal cancer incidence.

To examine timing of eating across ten European countries.

Cross-sectional analysis of the European Prospective Investigation into Cancer and Nutrition (EPIC) calibration study using standardized 24 h diet recalls collected during 1995–2000. Eleven predefined food consumption occasions were assessed during the recall interview. We present time of consumption of meals and snacks as well as the later:earlier energy intake ratio, with earlier and later intakes defined as 06.00–14.00 and 15.00–24.00 hours, respectively. Type III tests were used to examine associations of sociodemographic, lifestyle and health variables with timing of energy intake.

Ten Western European countries.

In total, 22 985 women and 13 035 men aged 35–74 years (n 36 020).

A south–north gradient was observed for timing of eating, with later consumption of meals and snacks in Mediterranean countries compared with Central and Northern European countries. However, the energy load was reversed, with the later:earlier energy intake ratio ranging from 0·68 (France) to 1·39 (Norway) among women, and from 0·71 (Greece) to 1·35 (the Netherlands) among men. Among women, country, age, education, marital status, smoking, day of recall and season were all independently associated with timing of energy intake (all P<0·05). Among men, the corresponding variables were country, age, education, smoking, physical activity, BMI and day of recall (all P<0·05).

We found pronounced differences in timing of eating across Europe, with later meal timetables but greater energy load earlier during the day in Mediterranean countries compared with Central and Northern European countries.

To describe the development of the Oxford WebQ, a web-based 24 h dietary assessment tool developed for repeated administration in large prospective studies; and to report the preliminary assessment of its performance for estimating nutrient intakes.

We developed the Oxford WebQ by repeated testing until it was sufficiently comprehensive and easy to use. For the latest version, we compared nutrient intakes from volunteers who completed both the Oxford WebQ and an interviewer-administered 24 h dietary recall on the same day.

Oxford, UK.

A total of 116 men and women.

The WebQ took a median of 12·5 (interquartile range: 10·8–16·3) min to self-complete and nutrient intakes were estimated automatically. By contrast, the interviewer-administered 24 h dietary recall took 30 min to complete and 30 min to code. Compared with the 24 h dietary recall, the mean Spearman's correlation for the 21 nutrients obtained from the WebQ was 0·6, with the majority between 0·5 and 0·9. The mean differences in intake were less than ±10 % for all nutrients except for carotene and vitamins B12 and D. On rare occasions a food item was reported in only one assessment method, but this was not more frequent or systematically different between the methods.

Compared with an interviewer-based 24 h dietary recall, the WebQ captures similar food items and estimates similar nutrient intakes for a single day's dietary intake. The WebQ is self-administered and nutrients are estimated automatically, providing a low-cost method for measuring dietary intake in large-scale studies.

Vegetarians and vegans exclude certain food sources of vitamin D from their diet, but it is not clear to what extent this affects plasma concentrations of 25-hydroxyvitamin D (25(OH)D). The objective was to investigate differences in vitamin D intake and plasma concentrations of 25(OH)D among meat eaters, fish eaters, vegetarians and vegans.

A cross-sectional analysis.

United Kingdom.

Plasma 25(OH)D concentrations were measured in 2107 white men and women (1388 meat eaters, 210 fish eaters, 420 vegetarians and eighty-nine vegans) aged 20–76 years from the European Prospective Investigation into Cancer and Nutrition (EPIC)–Oxford cohort.

Plasma 25(OH)D concentrations reflected the degree of animal product exclusion and, hence, dietary intake of vitamin D; meat eaters had the highest mean intake of vitamin D (3·1 (95 % CI 3·0, 3·2) μg/d) and mean plasma 25(OH)D concentrations (77·0 (95 % CI 75·4, 78·8) nmol/l) and vegans the lowest (0·7 (95 % CI 0·6, 0·8) μg/d and 55·8 (95 % CI 51·0, 61·0) nmol/l, respectively). The magnitude of difference in 25(OH)D concentrations between meat eaters and vegans was smaller (20 %) among those participants who had a blood sample collected during the summer months (July–September) compared with the winter months (38 %; January–March). The prevalence of low plasma concentrations of 25(OH)D (<25 nmol/l) during the winter and spring ranged from <1 % to 8 % across the diet groups.

Plasma 25(OH)D concentrations were lower in vegetarians and vegans than in meat and fish eaters; diet is an important determinant of plasma 25(OH)D in this British population.

To compare the average out-of-home (OH) consumption of foods and beverages, as well as energy intake, among populations from 10 European countries and to describe the characteristics of substantial OH eaters, as defined for the purpose of the present study, in comparison to other individuals.

Cross-sectional study. Dietary data were collected through single 24-hour dietary recalls, in which the place of consumption was recorded. For the present study, substantial OH eaters were defined as those who consumed more than 25% of total daily energy intake at locations other than the household premises. Mean dietary intakes and the proportion of substantial OH eaters are presented by food group and country. Logistic regression analyses were used to estimate the odds of being a substantial OH eater in comparison to not being one, using mutually adjusted possible non-dietary determinants.

Ten European countries participating in the European Prospective Investigation into Cancer and Nutrition (EPIC).

The subjects were 34 270 individuals, 12 537 men and 21 733 women, aged 35–74 years.

The fraction of energy intake during OH eating was generally higher in northern European countries than in the southern ones. Among the food and beverage groups, those selectively consumed outside the home were coffee/tea/waters and sweets and, to a lesser extent, cereals, meats, added lipids and vegetables. Substantial OH eating was positively associated with energy intake and inversely associated with age and physical activity. Substantial OH eating was less common among the less educated compared with the more educated, and more common during weekdays in central and north Europe and during the weekend in south Europe.

Eating outside the home was associated with sedentary lifestyle and increased energy intake; it was more common among the young and concerned in particular coffee/tea/waters and sweets.