Previous studies revealed that consuming spicy food reduced mortality from CVD and lowered stroke risk. However, no studies reported the relationship between spicy food consumption, stroke types and dose–response. This study aimed to further explore the association between the frequency of spicy food intake and the risk of stroke in a large prospective cohort study. In this study, 50 174 participants aged 30–79 years were recruited. Spicy food consumption data were collected via a baseline survey questionnaire. Outcomes were incidence of any stroke, ischaemic stroke (IS) and haemorrhagic stroke (HS). Multivariable-adjusted Cox proportional hazard models estimated the association between the consumption of spicy food and incident stroke. Restricted cubic spline analysis was used to examine the dose–response relationship. During the median 10·7-year follow-up, 3967 strokes were recorded, including 3494 IS and 516 HS. Compared with those who never/rarely consumed spicy food, those who consumed spicy food monthly, 1–2 d/week and 3–5 d/week had hazard ratio (HR) of 0·914 (95 % CI 0·841, 0·995), 0·869 (95 % CI 0·758, 0·995) and 0·826 (95 % CI 0·714, 0·956) for overall stroke, respectively. For IS, the corresponding HR) were 0·909 (95 % CI 0·832, 0·994), 0·831 (95 % CI 0·718, 0·962) and 0·813 (95 % CI 0·696, 0·951), respectively. This protective effect showed a U-shaped dose–response relationship. For obese participants, consuming spicy food ≥ 3 d/week was negatively associated with the risk of IS. We found the consumption of spicy food was negatively associated with the risk of IS and had a U-shaped dose–response relationship with risk of IS. Individuals who consumed spicy food 3–5 d/week had a significantly lowest risk of IS.

Observational research, mainly prospective cohort studies (PCS), has represented a long-standing challenge for those attempting to draw up consistent policy recommendations in the area of diet and health. This has been due to the inherent limitations in ascribing causality from observed associations due to problems of confounding of the findings and publication and citation bias. Developments in nutritional epidemiology research over the past 20–30 years have enabled causal criteria to be derived from observational studies and the totality of the primary literature to be reviewed objectively, reducing previous focus on narrative accounts of individual studies. The gold standard approach to assessing causal relationships is via randomised controlled trials (RCT), but neither RCT nor PCS provide direct evidence for biological plausibility, which is a key criterion for assessing causality. Although extensive mechanistic data are available in the literature, a systematic approach to select and assess quality and relevance of published studies has not been available. This limits their use in the development of diet and health policy. Recent studies have investigated a proposed two-step framework and novel methodologies for integrating heterogeneous data from cell, animal and human studies. Pilot and feasibility studies have shown this to be a useful novel approach to studies of diet and cancer, but further refinements are required, including development of appropriate quality criteria which are less dependent on RCT designs. Future studies are needed to fully verify the approach and its potential for use in other diet–disease relationships.

Dietary and lifestyle evolutionary discordance is hypothesised to play a role in the aetiology of CVD, including CHD and stroke. We aimed to investigate associations of a previously reported, total (dietary plus lifestyle) evolutionary-concordance (EC) pattern score with incident CVD, CHD and stroke. We used multivariable Cox proportional hazards regression to investigate associations of the EC score with CVD, CHD and stroke incidence among USA Black and White men and women ≥45 years old in the prospective REasons for Geographic and Racial Differences in Stroke study (2003–2017). The EC score comprised seven equally weighted components: a previously reported dietary EC score (using Block 98 FFQ data) and six lifestyle characteristics (alcohol intake, physical activity, sedentary behaviour, waist circumference, smoking history and social network size). A higher score indicates a more evolutionary-concordant dietary/lifestyle pattern. Of the 15 467 participants in the analytic cohort without a CVD diagnosis at baseline, 1563 were diagnosed with CVD (967 with CHD and 596 with stroke) during follow-up (median 11·0 years). Among participants in the highest relative to the lowest EC score quintile, the multivariable-adjusted hazards ratios and their 95 % CI for CVD, CHD and stroke were, respectively, 0·73 (0·62, 0·86; P trend < 0·001), 0·72 (0·59, 0·89; P trend < 0·001) and 0·76 (0·59, 0·98; P trend = 0·01). The results were similar by sex and race. Our findings support that a more evolutionary-concordant diet and lifestyle pattern may be associated with lower risk of CVD, CHD and stroke.

The present study aimed to investigate the association of the Chinese visceral adiposity index (CVAI) and its 6-year change with hypertension risk and compare the ability of CVAI and other obesity indices to predict hypertension based on the Rural Chinese Cohort Study. Study participants were randomly recruited by a cluster sampling procedure, and 10 304 participants ≥18 years were included. Modified Poisson regression was used to derive adjusted relative risks (RR) and 95 % CI. We identified 2072 hypertension cases during a median of 6·03 years of follow-up. The RR for the highest v. lowest CVAI quartile were 1·29 (95 % CI 1·05, 1·59) for men and 1·53 (95 % CI 1·22, 1·91) for women. Per-sd increase in CVAI was associated with hypertension for both men (RR 1·09, 95 % CI 1·02, 1·16) and women (RR 1·14, 95 % CI 1·06, 1·22). Also, the area under the receiver operating characteristic curve value for hypertension was higher for CVAI than the four other obesity indices for both sexes (all P < 0·05). Finally, per-sd increase in CVAI change was associated with hypertension for both men (RR 1·26, 95 % CI 1·16, 1·36) and women (RR 1·23, 95 % CI 1·15, 1·30). Similar results were observed in sensitivity analyses. CVAI and its 6-year change are positively associated with hypertension risk. CVAI has better performance in predicting hypertension than other visceral obesity indices for both sexes. The current findings suggest CVAI as a reliable and applicable predictor of hypertension in rural Chinese adults.

A higher dietary intake or serum concentration of betaine has been associated with greater lean body mass in middle-aged and older adults. However, it remains unknown whether betaine intake is associated with age-related loss of skeletal muscle mass (SMM). We assessed the association between dietary betaine intake and relative changes in SMM after 3 years in middle-aged adults. A total of 1242 participants aged 41–60 years from the Guangzhou Nutrition and Health Study 2011–2013 and 2014–2017 with body composition measurements by dual-energy X-ray absorptiometry were included. A face-to-face questionnaire was used to collect general baseline information. After adjustment for potential confounders, multiple linear regression found that energy-adjusted dietary betaine intake was significantly and positively associated with relative changes (i.e. percentage loss or increase) in SMM of legs, limbs and appendicular skeletal mass index (ASMI) over 3 years of follow-up (β 0·322 (se 0·157), 0·309 (se 0·142) and 0·303 (se 0·145), respectively; P < 0·05). The ANCOVA models revealed that participants in the highest betaine tertile had significantly less loss in SMM of limbs and ASMI and more increase in SMM of legs over 3 years of follow-up, compared with those in the bottom betaine tertile (all Ptrend < 0·05). In conclusion, our findings suggest that elevated higher dietary betaine intake may be associated with less loss of SMM of legs, limbs and ASMI in middle-aged adults.

Although the association between fruit consumption and CHD risk has been well studied, few studies have focused on flavonoid-rich fruits (FRF), in particular strawberries and grapes. We aimed to verify the association of total and specific FRF consumption with risk of CHD by a large prospective cohort study. A total of 87 177 men and women aged 44–75 years who were free of CVD and cancer at study baseline were eligible for the present analysis. FRF consumption was assessed using a FFQ. Cox proportional hazards regression models were used to estimate the hazard ratios (HR) of CHD in relation to FRF consumption with adjustment for potential risk factors and confounders. During a mean follow-up of 13·2 years, we identified 1156 incident CHD cases. After full adjustment for covariates including demographics, lifestyles and dietary factors, the HR were 0·93 (95 % CI 0·77, 1·11), 0·91 (95 % CI 0·75, 1·11), 0·84 (95 % CI 0·67, 1·04) and 0·78 (95 % CI 0·62, 0·99) for the second, third, fourth and fifth quintiles compared with the lowest quintile of FRF consumption. Regarding specific fruits, we observed a significant inverse association for citrus fruit consumption and a borderline inverse association for strawberry consumption, while no association was observed for apple/pear or grape consumption. Although the associations appeared to be stronger in women, they were not significantly modified by sex. Higher consumption of FRF, in particular, citrus fruits, may be associated with a lower risk of developing CHD.

Fish harbour many types of nutrients that are beneficial for preventing cognitive decline. Therefore, habitual fish intake might contribute to a lower risk of incident dementia. However, few prospective cohort studies have investigated fish consumption in relation to incident dementia, and their findings have been inconsistent. To investigate the association between fish consumption and the risk of incident dementia, we collected data on the consumption of fish and other foods using an FFQ in a baseline survey of individuals aged ≥65 years living in Ohsaki City, Japan. After 5·7 years of follow-up, the incidence of dementia was 1118 (8·5 %) among 13 102 participants. We then used a multivariate-adjusted Cox model to estimate hazard ratios (HR) and 95 % CI. Compared with subjects with the lowest fish intake (Q1), the multivariate HR were 0·90 (95 % CI 0·74, 1·11) for Q2, 0·85 (95 % CI 0·73, 0·99) for Q3 and 0·84 (95 % CI 0·71, 0·997) for Q4 (Ptrend = 0·029). Such associations were also observed even after excluding participants who were diagnosed with dementia in the first 2 years of follow-up and those who had poorer cognitive function at baseline. In conclusion, an association was observed between higher fish consumption and a lower risk of incident dementia among healthy elderly people without disability. These findings suggest that habitual fish intake may be beneficial for the prevention of dementia.

The association of chocolate consumption with risk of gestational diabetes has not been examined. We aimed to investigate the prospective association between chocolate consumption and risk of gestational diabetes in a large birth cohort in Japan. A total of 97 454 pregnant women with a median gestational age of 12 weeks were recruited from January 2011 to March 2014. Data on demographic information, disease history, socio-economic status, lifestyle and dietary habits were obtained at the study enrolment. Dietary intake during the past 12 months before study enrolment was assessed through a semi-quantitative FFQ. The logistic regression was used to obtain the OR of gestational diabetes in relation to chocolate consumption. Among 84 948 women eligible for the analysis, 1904 cases of gestational diabetes (2·2 %) were identified during the period of pregnancy. After controlling for potential confounding factors including age, smoking status, drinking status, education level, occupation, pre-pregnant BMI, depression, previous history of macrosomia babies, parity, physical activity and dietary factors, women in the highest quartile of chocolate consumption, compared with those in the lowest quartile, had a significantly lower risk of developing gestational diabetes (OR 0·78, 95 % CI 0·67, 0·90; P for trend = 0·002). Stratified analyses suggested that the association was not significantly modified by pre-pregnancy BMI, age, parity, smoking status or drinking status. The present prospective cohort study provided evidence that chocolate consumption was associated with a significant lower risk of gestational diabetes in Japanese women.

Recent dairy product studies have suggested that fermented rather than non-fermented dairy products might provide benefits on cardiovascular health, but the evidence is inconclusive. Therefore, we investigated whether fermented and non-fermented dairy products have distinct associations with the risk of incident CHD in a population with high dairy product intake. The present study included a total of 1981 men, aged 42–60 years, from the Kuopio Ischaemic Heart Disease Risk Factor Study, with no CHD at baseline. Dietary intakes were assessed with instructed 4-d food records. We used Cox’s proportional hazards regression model to estimate the associations with the risk of CHD. Fatal and non-fatal CHD events were ascertained from national registries. During a mean follow-up of 20·1 years, 472 CHD events were recorded. Median intakes were 105 g/d for fermented (87 % low-fat products) and 466 g/d for non-fermented dairy products (60 % low-fat products). After adjusting for potential confounders, those in the highest (v. lowest) intake quartile of fermented dairy products had 27 % (95 % CI 5, 44; P-trend=0·02) lower risk of CHD. In contrast, those in the highest intake quartile of non-fermented dairy products had 52 % (95 % CI 13, 104; P-trend=0·003) higher risk of CHD. When analysed based on fat content, low-fat (<3·5 % fat) fermented dairy product intake was associated with lower risk (hazard ratio in the highest quartile=0·74; 95 % CI 0·57, 0·97; P-trend=0·03), but high-fat fermented dairy and low-fat or high-fat non-fermented dairy products had no association. These results suggest that fermented and non-fermented dairy products can have opposite associations with the risk of CHD.

It has not been well established whether dietary folate intake reduces the risk of diabetes development. We aimed to clarify the prospective association between dietary folate intake and type 2 diabetes (T2D) risk among 7333 Korean adults aged 40 years or older who were included in the Multi-Rural Communities Cohort. Dietary folate intake was estimated from all 106 food items listed on a FFQ, not including folate intake from supplements. Two different measurements of dietary folate intake were used: the baseline consumption and the average consumption from baseline until just before the end of follow-up. The association between folate intake and T2D risk was determined through a modified Poisson regression model with a robust error estimator controlling for potential confounders. For 29 745 person years, 319 cases of diabetes were ascertained. In multivariable analyses, dietary folate intake was inversely associated with risk of T2D for women, not for men. For women, the incidence rate ratio of diabetes in the third tertile compared with the first tertile was 0·57 (95 % CI 0·38–0·87, Pfor trend=0·0085) in the baseline consumption model and 0·64 (95 % CI 0·43–0·95, Pfor trend=0·0244) in the average consumption model. These inverse associations was found in both normal fasting blood glucose group and impaired fasting glucose group among women. Among non-users of multinutrients and vitamin supplements, the significant inverse association remained. Thus, higher dietary intake of folate is prospectively associated with lower risk of diabetes for women.

Previous studies that have investigated the association between B-vitamin supplement use and risk for cataract yield conflicting results. The aim of this study was to examine the association between use of high-dose B-vitamin supplements (approximately 10 times recommended daily intake) and risk for age-related cataract in a population-based prospective study of 13 757 women from the Swedish Mammography Cohort and 22 823 men from the Cohort of Swedish Men. Dietary supplement use and potential confounders were assessed using a questionnaire at baseline. Information on cataract diagnosis and extraction was obtained through linkage to registers. During the follow-up period between January 1998 and December 2011, we identified 8395 cataract cases (3851 for women and 4544 for men). The use of B vitamins plus other supplements and B vitamins only was associated with 9 % (95 % CI 2, 17) and 27 % (95 % CI 12, 43) increased risk for cataract, respectively. The hazard ratios for use of B vitamins only and risk for cataract stratified by different age groups were as follows: <60 years: 1·88 (95 % CI 1·47, 2·39); 60–69 years: 1·21 (95 % CI 0·96, 1·53); and ≥70 years: 1·09 (95 % CI 0·91, 1·31) (Pinteraction=0·002). Our results suggest that the use of high-dose B-vitamin supplements was associated with an increased risk for cataract. This association might be confined to younger participants.

Arterial wall thickening, stimulated by low-grade systemic inflammation, underlies many cardiovascular events. As diet is a significant moderator of systemic inflammation, the dietary inflammatory index (DIITM) has recently been devised to assess the overall inflammatory potential of an individual’s diet. The primary objective of this study was to assess the association of the DII with common carotid artery–intima-media thickness (CCA–IMT) and carotid plaques. To substantiate the clinical importance of these findings we assessed the relationship of DII score with atherosclerotic vascular disease (ASVD)-related mortality, ischaemic cerebrovascular disease (CVA)-related mortality and ischaemic heart disease (IHD)-related mortality more. The study was conducted in Western Australian women aged over 70 years (n 1304). Dietary data derived from a validated FFQ (completed at baseline) were used to calculate a DII score for each individual. In multivariable-adjusted models, DII scores were associated with sub-clinical atherosclerosis: a 1 sd (2·13 units) higher DII score was associated with a 0·013-mm higher mean CCA–IMT (P=0·016) and a 0·016-mm higher maximum CCA–IMT (P=0·008), measured at 36 months. No relationship was seen between DII score and carotid plaque severity. There were 269 deaths during follow-up. High DII scores were positively associated with ASVD-related death (per sd, hazard ratio (HR): 1·36; 95 % CI 1·15, 1·60), CVA-related death (per sd, HR: 1·30; 95 % CI 1·00, 1·69) and IHD-related death (per sd, HR: 1·40; 95 % CI 1·13, 1·75). These results support the hypothesis that a pro-inflammatory diet increases systemic inflammation leading to development and progression of atherosclerosis and eventual ASVD-related death.

A lower BMI has been reported among consumers of organic foods, but this relationship has never been examined in a prospective design study. Our aim was to prospectively investigate the association between frequency of organic food consumption and weight change. We analysed data from 62 224 participants of the NutriNet-Santé cohort (78 % women, mean age=45 years) with information on consumption frequency of organic foods, dietary intake and repeated anthropometric data. For sixteen products, participants reported their consumption frequency of labelled organic foods (never, occasionally, most of the time). An organic score (OS) with a maximum of thirty-two points was computed. The associations of the OS (modeled as quartiles (Q)) with change in BMI during follow-up (on average 3·1 years) and with the risk of overweight and obesity were estimated by ANCOVA and multivariable logistic regression. A lower BMI increase was observed across quartiles of the OS (mean difference Q4 v. Q1=−0·16 (95 % CI −0·32, −0·01). An increase in the OS was associated with a lower risk of overweight and obesity (among non-overweight and non-obese participants at inclusion): OR for Q4 v. Q1 were 0·77 (95 % CI 0·68, 0·86) and 0·69 (95 % CI 0·58, 0·82), respectively. Concerning obesity risk, the association was stronger among participants with higher adherence to nutritional guidelines. This study supports a strong protective role of consumption frequency of organic foods with regard to the risk of overweight and obesity that depends on overall dietary quality. Upon confirmation, these results may contribute to fine-tune nutritional guidelines by accounting for farming practices in food production.

There is limited evidence on the prospective association of time spent in activity intensity (sedentary (SED), moderate (MPA) or vigorous (VPA) physical activity) and dietary intake with adiposity indicators in young people. This study aimed to assess associations between (1) baseline objectively measured activity intensity, dietary energy density (DED) and 4-year change in adiposity and (2) 4-year change in activity intensity/DED and adiposity at follow-up. We conducted cohort analyses including 367 participants (10 years at baseline, 14 years at follow-up) with valid data for objectively measured activity (Actigraph), DED (4-d food diary), anthropometry (waist circumference (WC), %body fat (%BF), fat mass index (FMI), weight status) and covariates. Linear and logistic regression models were fit, including adjustment for DED and moderate-to-vigorous physical activity. Results showed that baseline DED was associated with change in WC (β for 1kJ/g difference: 0·71; 95% CI 0·26, 1·17), particularly in boys (1·26; 95% CI 0·41, 2·16 v. girls: 0·26; 95% CI −0·34, 0·87), but not with %BF, FMI or weight status. In contrast, baseline SED, MPA or VPA were not associated with any of the outcomes. Change in DED was negatively associated with FMI (β for 1kJ/g increase: −0·86; 95% CI −1·59, −0·12) and %BF (−0·86; 95% CI −1·25, −0·11) but not WC (−0·27; 95% CI −1·02, 0·48). Change in SED, MPA and VPA did not predict adiposity at follow-up. In conclusion, activity intensity was not prospectively associated with adiposity, whereas the directions of associations with DED were inconsistent. To inform public health efforts, future studies should continue to analyse longitudinal data to further understand the independent role of different energy-balance behaviours in changes in adiposity in early adolescence.

Vitamin D deficiency is common during pregnancy and higher in Hispanic as compared with non-Hispanic white women. However, the association between vitamin D deficiency and adverse pregnancy outcomes remains unclear and may vary across ethnic groups, in part because of genetic variation in the metabolism of vitamin D. Few studies have included Hispanic women. Therefore, we investigated this association among 237 participants in the Behaviors Affecting Baby and You Study, a randomised trial of an exercise intervention among ethnically diverse prenatal care patients in Massachusetts. Baseline serum 25-hydroxyvitamin D (25(OH)D) was measured at 15·2 (sd 4·7) weeks’ gestation. Information on adverse pregnancy outcomes was abstracted from medical records. Mean 25(OH)D was 30·4 (sd 12·0) ng/ml; 53·2 % of participants had insufficient (<30 ng/ml) and 20·7 % had deficient (<20 ng/ml) 25(OH)D levels. After adjusting for month of blood draw, gestational age at blood draw, gestational age at delivery, age, BMI and Hispanic ethnicity, women with insufficient and deficient vitamin D had infants with birth weights 139·74 (se 69·16) g (P=0·045) and 175·52 (se 89·45) g (P=0·051) lower compared with women with sufficient vitamin D levels (≥30 ng/ml). Each 1 ng/ml increase in 25(OH)D was associated with an increased risk for gestational diabetes mellitus among Hispanic women only (relative risk 1·07; 95 % CI 1·03, 1·11) in multivariable analysis. We did not observe statistically significant associations between maternal vitamin D status and other pregnancy outcomes. Our findings provide further support for an adverse impact of vitamin D deficiency on birth weight in Hispanic women.

Diets with a high glycaemic index (GI) or glycaemic load (GL) have been hypothesised to increase the risk of diabetes, CVD and some cancers. In the present study, the associations of dietary GI and GL with the risk of all-cause and cause-specific mortality were prospectively examined in a general population in Japan, where white rice is the main contributor of dietary GI and GL. A total of 28 356 residents of Takayama City, Japan, who responded to a self-administered questionnaire in 1992 were included in the present analyses. Dietary intake was assessed using a validated FFQ. Mortality was ascertained over 16 years. In men, dietary GI was found to be significantly inversely associated with the risk of all-cause and non-cancer, non-cardiovascular mortality; the hazard ratios (HR) for the highest v. lowest quartile were 0·80 (95 % CI 0·68, 0·95) and 0·64 (95 % CI 0·49, 0·84), respectively. Dietary GL was found to be significantly inversely associated with the risk of all-cause, cancer, and non-cancer, non-cardiovascular mortality; the HR for the highest v. lowest quartile were 0·71 (95 % CI 0·59, 0·86), 0·71 (95 % CI 0·52, 0·99) and 0·64 (95 % CI 0·48, 0·87), respectively. The results obtained for the GL derived from white rice, but not from other foods, closely mirrored those obtained for overall GL. In women, dietary GI was found to be significantly positively associated with the risk of cardiovascular mortality; the HR for the highest v. lowest quartile was 1·56 (95 % CI 1·15, 2·13). The results of the present study suggest potential favourable effects of dietary GI and GL on mortality in men, but an association between high GI and an increased risk of cardiovascular mortality in women.

Cancer is a leading cause of death, and the dietary pattern in Korea is changing rapidly from a traditional Korean diet to a Westernised diet. In the present study, we investigated the effects of dietary factors on cancer risk with a prospective cohort study. Among 26 815 individuals who participated in cancer screening examinations from September 2004 to December 2008, 8024 subjects who completed a self-administered questionnaire concerning demographic and lifestyle factors, and a 3 d food record were selected. As of September 2013, 387 cancer cases were identified from the National Cancer Registry System, and the remaining individuals were included in the control group. The hazard ratio (HR) of cancer for the subjects older than or equal to 50 years of age was higher (HR 1·80, 95 % CI 1·41, 2·31; P< 0·0001) than that for the other subjects. Red meat consumption, Na intake and obesity (BMI ≥ 25 kg/m2) were positively associated with overall cancer incidence in men (HR 1·41, 95 % CI 1·02, 1·94; P= 0·0382), gastric cancer (HR 2·34, 95 % CI 1·06, 5·19; P= 0·0365) and thyroid cancer (HR 1·56, 95 % CI 1·05, 2·31; P= 0·0270), respectively. Participants who had at least three dietary risk factors among the high intakes of red meat and Na, low intakes of vegetables and fruits, and obesity suggested by the World Cancer Research Fund/American Institute for Cancer Research at baseline tended to have a higher risk of cancer than the others (HR 1·26, 95 % CI 0·99, 1·60; P= 0·0653). In summary, high intakes of red meat and Na were significant risk factors of cancer among Koreans.

Coffee consumption has been shown to be associated with various health outcomes, but no comprehensive review and meta-analysis of the association between coffee consumption and total mortality has been conducted. To quantitatively assess this association, we conducted a meta-analysis of prospective cohort studies. Eligible studies were identified by searching the PubMed and EMBASE databases for all articles published through June 2013 and reviewing the reference lists of the retrieved articles. Pooled relative risks (RR) with 95 % CI were calculated using a random-effects model. We identified twenty studies of coffee consumption and total mortality, including 129 538 cases of deaths among the 973 904 participants. The RR of total mortality for the high v. low category of coffee consumption was 0·86 (95 % CI 0·80, 0·92). The pooled RR for studies using ≥ 2–4 cups/d as a cut-off for the high category was similar to that for studies using ≥ 5–9 cups/d as the cut-off. By geographical region, the inverse association tended to be stronger for the eight studies conducted in Europe (RR 0·78, 95 % CI 0·70, 0·88) and three studies carried out in Japan (RR 0·82, 95 % CI 0·73, 0·92) than for the nine studies conducted in the USA (RR 0·92, 95 % CI 0·84, 1·00). The inverse association was similar for men (RR 0·81, 95 % CI 0·73, 0·90) and women (RR 0·84, 95 % CI 0·79, 0·89). A weak, but significant, inverse association was found with moderate coffee consumption (1–2 cups/d; RR 0·92, 95 % CI 0·87, 0·98). High decaffeinated coffee consumption was also found to be associated with a lower risk of death, but the data are limited. Our findings indicate that coffee consumption is associated with a reduced risk of total mortality.

Observational studies have suggested that the intake of flavonoids is associated with a decreased risk of CVD. However, the results of these studies remain controversial. The aim of the present study was to evaluate the association between dietary flavonoid intake and CVD risk by conducting a systematic review of prospective cohort studies. Electronic reference databases were searched to identify studies that met the pre-stated inclusion criteria. The studies were assessed for eligibility and data were extracted by two authors independently. For each study, relative risks (RR) and 95 % CI were extracted and pooled using either a fixed-effects or a random-effects model. Generalised least-squares trend estimation analysis was used to evaluate dose–response relationships. The inclusion criteria were met by fourteen prospective cohort studies. The intakes of anthocyanidins (RR 0·89, 95 % CI 0·83, 0·96), proanthocyanidins (RR 0·90, 95 % CI 0·82, 0·98), flavones (RR 0·88, 95 % CI 0·82, 0·96), flavanones (RR 0·88, 95 % CI 0·82, 0·96) and flavan-3-ols (RR 0·87, 95 % CI 0·80, 0·95) were inversely associated with the risk of CVD when comparing the highest and lowest categories of intake. A similar association was observed for flavonol intake and CVD risk. Sensitivity and subgroup analyses further supported this association. The summary RR for CVD for every 10 mg/d increment in flavonol intake was 0·95 (95 % CI 0·91, 0·99). The present systematic review suggests that the dietary intakes of six classes of flavonoids, namely flavonols, anthocyanidins, proanthocyanidins, flavones, flavanones and flavan-3-ols, significantly decrease the risk of CVD.