Identifying trends in dietary salt sources is essential for effectively reducing salt/Na intake. This study aimed to examine the trends in dietary salt sources among Japanese adults using the 2007–2019 National Health and Nutrition Survey data collected from 95 581 adults aged ≥ 20 years. Dietary intake was estimated using the 1-d household-based dietary record. Foods reported as potential sources of salt intake in Japan and other countries were categorised into twenty-one groups. Salt intake for each food group was adjusted using the density method based on the energy intake. Trends in dietary salt intake based on food sources by sex and age groups (20–39 years, 40–59 years and ≥ 60 years) were analysed using the Joinpoint Regression Program. Salt intake for each age group in both men and women decreased from 2007 (5·3 g/1000 kcal–6·4 g/1000 kcal) to 2019 (4·9 g/1000 kcal–5·6 g/1000 kcal). The major dietary source of salt continued to be seasonings such as soya sauce and soyabean paste (approximately 70 %). Salt intake from seasonings decreased over time in adults aged ≥ 40 years but did not change in those aged 20–39 years. Additionally, a decreasing salt intake from unprocessed fish and shellfish and an increasing salt intake from unprocessed meat were observed across all age categories for both sexes. This study demonstrated that a strategy targeting different age groups may be needed to reduce salt consumption from seasonings among the Japanese population. Further studies on salt content in seasonings and continued monitoring of trends in dietary salt sources are required.

Iodine supply is crucial during pregnancy to ensure that the proper thyroid function of mother and baby support fetal brain development. Little is known about iodine status or its dietary determinants in pregnant women in the Republic of Cyprus. We therefore recruited 128 pregnant women at their first-trimester ultrasound scan to a cross-sectional study. We collected spot-urine samples for the measurement of urinary iodine concentration (UIC, µg/l) and creatinine concentration (Creat, g/l), the latter of which allows us to correct for urine dilution and to compute the iodine-to-creatinine ratio (UI/Creat). Women completed a FFQ and a general questionnaire. We used a General Linear model to explore associations between maternal and dietary characteristics with UI/Creat. The median UIC (105 µg/l) indicated iodine deficiency according to the WHO criterion (threshold for adequacy = 150 µg/l), and the UI/Creat was also low at 107 µg/g. Only 32 % (n 45) of women reported the use of iodine-containing supplements; users had a higher UI/Creat than non-users (131 µg/g v. 118 µg/g), though this difference was NS in the adjusted analysis (P = 0·37). Of the dietary components, only egg intake was significantly associated with a higher UI/Creat in adjusted analyses (P = 0·018); there was no significant association with milk, dairy products or fish intake. Our results suggest that pregnant women in Cyprus have inadequate iodine status and are at risk of mild-to-moderate iodine deficiency. Further research on dietary sources in this population is required.

We examined cross-sectional and longitudinal associations of dietary factors with caries experience in a population sample of 487 children aged 6–9 years at baseline examinations of the Physical Activity and Nutrition in Children (PANIC) Study. Altogether, 406 of these children attended 2-year follow-up examinations. Food consumption and eating frequency were assessed using 4-day food records, diet quality using the Baltic Sea Diet Score (BSDS) and eating behaviour using the Children’s Eating Behavior Questionnaire. Caries experience was examined clinically. The cross-sectional associations of dietary factors with caries experience at baseline were analysed using linear regression and the longitudinal associations of dietary factors with a change in caries experience over follow-up using generalised mixed-effects regression adjusted for other risk factors. A higher consumption of high-fibre grain products (standardised regression coefficient β = −0·16, P = 0·003) and milk (β = −0·11, P = 0·025) and higher BSDS (β = −0·15, P = 0·007) were associated with lower caries experience, whereas a higher consumption of potatoes (β = 0·11, P = 0·048) and emotional overeating (β = 0·12, P = 0·025) were associated with higher caries experience. Higher snacking frequency (fixed coefficient β = 0·07, P = 0·033), desire to drink (β = 0·10, P = 0·046), slowness in eating (β = 0·12, P = 0·027) and food fussiness (β = 0·12, P = 0·018) were associated with higher caries experience, whereas enjoyment of food (β = −0·12, P = 0·034) and higher BSDS (β = −0·02, P = 0·051) were associated with lower caries experience.

Food insecurity has been associated with poor diet, but few studies focused on the postpartum period – an important time for women’s health. We examined associations between food security and diet quality in postpartum women and assessed whether participation in federal food assistance programmes modified this potential relation. Using longitudinal data, we analysed the association between food security at 3 months postpartum and a modified Alternate Healthy Eating Index-2010 (AHEI) at 6 months postpartum (excluding alcohol). We conducted multivariable linear regressions examining associations between food security and AHEI. We assessed two food assistance programmes as potential effect modifiers. The sample included 363 postpartum women from the Nurture study, located in the Southeastern USA (2013–2017). Among women, 64·4 % were Black and 45·7 % had a high school diploma or less. We found no evidence of an interaction between food security and two federal food assistance programmes. In adjusted models, marginal, low and very low food security were not associated with AHEI. However, low (β: −0·64; 95 % CI −1·15, −0·13; P = 0·01) and very low (β: −0·57; 95 % CI −1·02, −0·13; P = 0·01) food security were associated with greater trans fat intake. Food security status was not associated with overall diet quality but was associated with higher trans fat (low and very low) and more moderate alcohol (marginal) intake. Future studies should assess the consistency and generalisability of these findings.

As, Pb and Hg are common environmental contaminants in low- and middle-income countries. We investigated the association between child toxicant exposure and growth and development and determined if this association was mitigated by Se concentration. Toxicant concentrations in fingernail samples, anthropometry and Bayley’s Scales of Infant Development, 3rd edition domains were assessed in 36-month-old children whose mothers had been part of a randomised controlled trial in rural Vietnam. Multivariable regression analyses were performed to estimate the effect of toxicant exposure on clinical outcomes with adjustments for potential confounders and interaction with fingernail Se concentration. We analysed 658 children who had data for at least one physical or developmental outcome, and at least one toxicant measurement, and each of the covariates. Fingernail As concentration was negatively associated with language (estimate per 10 % increase in As: −0·19, 95 % CI: (–0·32, −0·05)). Pb was negatively associated with cognition (estimate per 10 % increase in Pb: −0·08 (–0·15, −0·02)), language (estimate per 10 % increase in Pb: −0·18 (–0·28, −0·10)) and motor skills (estimate per 10 % increase in Pb: −0·12 (–0·24, 0·00)). Hg was negatively associated with cognition (estimate per 10 % increase in Hg: −0·48, (–0·72, −0·23)) and language (estimate per 10 % increase in Hg −0·51, (–0·88, −0·13)) when Se concentration was set at zero in the model. As Se concentration increased, the negative associations between Hg and both cognition and language scores were attenuated. There was no association between toxicant concentration and growth. As, Pb and Hg concentrations in fingernails of 3-year-old children were associated with lower child development scores. The negative association between Hg and neurological development was reduced in magnitude with increasing Se concentration. Se status should be considered when assessing heavy metal toxicants in children and their impact on neurodevelopmental outcomes.

A healthy lifestyle during adolescence is associated with insulin sensitivity or liver enzyme levels and thus might contribute to the prevention of non-alcoholic fatty liver disease (NAFLD). Therefore, we examined the association between adherence to a hypothesis-based lifestyle score including dietary intake, physical activity, sedentary behaviour, sleep duration and BMI in adolescence and fatty liver indices in early adulthood. Overall, 240 participants of the DOrtmund Nutritional and Anthropometric Longitudinally Designed study completed repeated measurements of lifestyle score factors during adolescence (females: 8·5–15·5 years, males: 9·5–16·5 years). Multivariable linear regression models were used to investigate the association between adolescent lifestyle scores and NAFLD risk (hepatic steatosis index (HSI) and fatty liver index (FLI)) in early adulthood (18–30 years). Participants visited the study centre 4·9 times during adolescence and achieved on average 2·8 (min: 0·6, max: 5) out of five lifestyle score points. Inverse associations were observed between the lifestyle score and fatty liver indices (HSI: ß=−5·8 % (95 % CI −8·3, −3·1), P < 0·0001, FLI: ß=−32·4 % (95 % CI −42·9, −20·0), P < 0·0001) in the overall study population. Sex-stratified analysis confirmed these results in men, while inverse but non-significant associations were observed in women (P > 0·05). A higher lifestyle score was associated with lower HSI and FLI values, suggesting that a healthy lifestyle during adolescence might contribute to NAFLD prevention, predominantly in men. Our findings on repeatedly measured lifestyle scores in adolescents and their association with NAFLD risk in early adulthood warrant confirmation in larger study populations.

The association between alcohol intake and the risk of glioma has been widely studied, but these results have yielded conflicting findings. Therefore, we conducted this systematic review and updated meta-analysis to systematically evaluate the association between alcohol intake and the risk of glioma. A systematic literature search of relevant articles published in PubMed, Web of Science, CNKI and Wan fang databases up to December 2021 was conducted. Pooled estimated of relative risk (RR) and 95 % CI were calculated using fixed-effects models. A total of eight articles with three case–control studies involving 2706 glioma cases and 2 189 927 participants were included in this meta-analysis. A reduced risk of glioma was shown for the low–moderate alcohol drinking v. non-drinking (RR = 0·87; 95 % CI (0·78, 0·97); P = 0·014). In addition, there was no evidence of an increased risk of glioma in the heavy alcohol drinking compared with non-drinking (RR = 0·89; 95 % CI (0·67, 1·18); P = 0·404). The findings suggest an inverse association between low–moderate alcohol drinking and the risk of glioma, in the absence, however, of a dose–response relationship. More prospective studies are needed to provide further insight into the association between alcohol drinking and glioma risk.

Hyperhomocysteinaemia (HHcy) is associated with all-cause mortality in some disease states. However, the correlation between HHcy and the risk of mortality in the general population has rarely been researched. We aimed to evaluate the association between HHcy and all-cause and cause-specific mortality among adults in the USA. This study analysed data from the National Health and Nutrition Examination Survey database (1999–2002 survey cycle). A multivariable Cox regression model was built to evaluate the correlation between HHcy and all-cause and cause-specific mortality. Smooth curve fitting was used to analyse their dose-dependent relationship. A total of 8442 adults aged 18–70 years were included in this study. After a median follow-up period of 14·7 years, 1007 (11·9 %) deaths occurred including 197 CVD-related deaths, 255 cancer-related deaths and fifty-eight respiratory disease deaths. The participants with HHcy had a 93 % increased risk of all-cause mortality (hazard ratio (HR) 1·93; 95 % CI (1·48, 2·51)), 160 % increased risk of CVD mortality (HR 2·60; 95 % CI (1·52, 4·45)) and 82 % increased risk of cancer mortality (HR 1·82; 95 % CI (1·03, 3·21)) compared with those without HHcy. For unmeasured confounding, E-value analysis proved to be robust. In conclusion, HHcy was associated with high risk of all-cause and cause-specific (CVD, cancer) mortality among adults aged below 70 years.

Some evidence suggests that breast-feeding may modify food preferences in the later years of life. The present study aimed to provide a systematic review of observational studies investigating the association between exposure to breast milk and its duration with data-driven or hypothesis-driven (or diet quality scores) dietary patterns over 1 year of age. The databases of PubMed, Scopus and Web of Science were searched for observational studies published from January 2010 until July 2021, which led to the identification of twenty-two eligible articles. There was considerable heterogeneity between studies in terms of assessment of exposure and outcome. Of the eleven studies that assessed data-driven dietary patterns, ten reported a significant association for at least one identified dietary pattern. Overall, being breast-fed and a longer duration of any/exclusive breast-feeding were associated with higher scores on healthy dietary patterns characterised mainly by high loadings of fruits, vegetables and whole grains. In comparison, a negative association was found for unhealthy dietary patterns rich in foods with high content of added sugar, salt and saturated fats. In terms of diet quality scores, nine out of eleven studies reported a significant positive association between the duration of any breast-feeding and adherence to recommended healthy diets or dietary guidelines. In conclusion, the evidence from this review was generally in support of the hypothesis indicating breast-feeding is associated with healthy dietary patterns at later ages. However, due to the methodological limitations in the available studies, further research is warranted to elucidate solid evidence on this topic.

Sarcopenic obesity is defined as the presence of high fat mass and low muscle mass combined with low physical function, and it is closely related with the onset of cardiovasular diseases (CVD). The existing anthropometric indices, which are being utilised in clinical practice as predictors of CVD, may also be used to screen sarcopenic obesity, but their feasibility remained unknown. Using cross-sectional data of 2031 participants aged 70–84 years (mean age, 75·9 ± 3·9 years; 49·2 % women) from the Korean Frailty and Aging Cohort Study, we analysed the association of anthropometric indices, including body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHtR) and weight-adjusted waist index (WWI) with sarcopenic obesity. Body composition was measured using dual-energy X-ray absorptiometry. Higher WWI, WHtR and WC quartiles were associated with higher risk of sarcopenic obesity; the odds ratio (OR) of sarcopenic obesity were highest in the fourth quartile of the WWI (OR: 10·99, 95 % CI: 4·92–24·85, P for trend < 0·001). WWI provided the best diagnostic power for sarcopenic obesity in men (area under the receiver operating characteristic curve: 0·781, 95 % CI: 0·751–0·837). No anthropometric indices were significantly associated with sarcopenic obesity in women. WWI was the only index that was negatively correlated with physical function in both men and women. WWI showed the strongest association with sarcopenic obesity, defined by high fat mass and low muscle mass combined with low physical function only in older men. No anthropometric indices were associated with sarcopenic obesity in older women.

Higher dietary protein, alone or in combination with physical activity (PA), may slow the loss of age-related muscle strength in older adults. We investigated the longitudinal relationship between protein intake and grip strength, and the interaction between protein intake and PA, using four longitudinal ageing cohorts. Individual participant data from 5584 older adults (52 % women; median: 75 years, IQR: 71·6, 79·0) followed for up to 8·5 years (mean: 4·9 years, SD: 2·3) from the Health ABC, NuAge, LASA and Newcastle 85+ cohorts were pooled. Baseline protein intake was assessed with food frequency questionnaires and 24-h recalls and categorized into < 0·8, 0·8–<1·0, 1·0–<1·2 and ≥ 1·2 g/kg adjusted body weight (aBW)/d. The prospective association between protein intake, its interaction with PA, and grip strength (sex- and cohort-specific) was determined using joint models (hierarchical linear mixed effects and a link function for Cox proportional hazards models). Grip strength declined on average by 0·018 SD (95 % CI: –0·026, –0·006) every year. No associations were found between protein intake, measured at baseline, and grip strength, measured prospectively, or rate of decline of grip strength in models adjusted for sociodemographic, anthropometric, lifestyle and health variables (e.g., protein intake ≥ 1·2 v· < 0·8 g/kg aBW/d: β = –0·003, 95 % CI: –0·014, 0·005 SD per year). There also was no evidence of an interaction between protein intake and PA. We failed to find evidence in this study to support the hypothesis that higher protein intake, alone or in combination with higher PA, slowed the rate of grip strength decline in older adults.

The aim of the study was to investigate the association between pre-gestational carbohydrate quality index (CQI) and the incidence of gestational diabetes mellitus (GDM). Data from the ‘Seguimiento Universidad de Navarra’ (SUN) cohort were used, which includes 3827 women who notified at least one pregnancy between December 1999 and December 2019. We used a validated semi-quantitative 136-item FFQ to evaluate dietary exposures at baseline and at 10-year follow-up. The CQI was defined by four criteria: glycaemic index, whole-grain/total-grain carbohydrate, dietary fibre intake and solid/total carbohydrate ratio. We fitted generalised estimating equations with repeated measurements of the CQI to assess its relationship with incident GDM. A total of 6869 pregnancies and 202 new cases of incident GDM were identified. The inverse association between the global quality of carbohydrate and the development of GDM was not statistically significant: OR the highest v. the lowest CQI category: 0·67, 95 % CI (0·40, 1·10), Pfor trend = 0·10. Participants at the highest CQI category and with daily carbohydrate amounts ≥50 % of total energy intake had the lowest incidence of GDM (OR = 0·29 (95 % CI (0·09, 0·89)) compared with those with the lowest quality (lowest CQI) and quantity (≤40 %). Further studies are needed to overcome the limitations of our study. Those studies should jointly consider the quality and the quantity of dietary carbohydrates, as the quality might be of importance, especially in women with a higher intake of carbohydrates.

The current study aims to describe the consumption of ultra-processed foods, from 2 to 4 years old, and evaluate its association with growth outcomes during the same period. It is a prospective cohort study using data from the 2015 Pelotas-Brazil Birth Cohort. Outcomes assessed at the 2- and 4-year-old follow-ups were BMI-for-age Z-score and length/height-for-age Z-score. The exposure was a score of ultra-processed food consumption calculated at each follow-up by summing up the positive answers for the consumption of nine specific items/subgroups of ultra-processed foods: (i) instant noodles; (ii) soft drink; (iii) chocolate powder in milk; (iv) nuggets, hamburger or sausages; (v) packaged salty snacks; (vi) candies, lollipops, chewing gum, chocolate or jelly; (vii) sandwich cookie or sweet biscuit; (viii) juice in can or box or prepared from a powdered mix and (ix) yogurt. Crude and adjusted analyses between the score of ultra-processed foods and the outcomes were run using generalised estimating equations. Prevalence of consumption of ultra-processed foods increased from 2 to 4 years old, for all evaluated items/subgroups, except yogurt. In prospective analyses, higher scores of ultra-processed food consumption were associated with higher BMI-for-age Z-score and lower length/height-for-age Z-score, after adjustment for confounders. Ultra-processed food consumption, measured using a short questionnaire with low research burden, increased from 2 to 4 years old and was related to deleterious growth outcomes in early childhood. These results reinforce the importance of avoiding the consumption of these products in childhood to prevent the double burden of malnutrition and non-communicable chronic diseases throughout the life.

Using a sample of US adults aged 65 years and older, we examined the role of dietary quality in cystatin C change over 4 years and whether this association varied by race/ethnicity. The Health and Retirement Study provided observations with biomarkers collected in 2012 and 2016, participant attributes measured in 2012, and dietary intake assessed in 2013. The sample was restricted to respondents who were non-Hispanic/Latino White (n 789), non-Hispanic/Latino Black (n 108) or Hispanic/Latino (n 61). Serum cystatin C was constructed to be equivalent to the 1999–2002 National Health and Nutrition Examination Survey (NHANES) scale. Dietary intake was assessed by a semi-quantitative FFQ with diet quality measured using an energy-adjusted form of the Alternative Healthy Eating Index-2010 (AHEI-2010). Statistical analyses were conducted using autoregressive linear modelling adjusting for covariates and complex sampling design. Cystatin C slightly increased from 1·2 mg/l to 1·3 mg/l over the observational period. Greater energy-adjusted AHEI-2010 scores were associated with slower increase in cystatin C from 2012 to 2016. Among respondents reporting moderately low to low dietary quality, Hispanic/Latinos had significantly slower increases in cystatin C than their non-Hispanic/Latino White counterparts. Our results speak to the importance of considering racial/ethnic determinants of dietary intake and subsequent changes in health in ageing populations. Further work is needed to address measurement issues including further validation of dietary intake questionnaires in diverse samples of older adults.

Many dietary guidelines recommend restricting the consumption of processed red meat (PRM) in favour of healthier foods such as fish, to reduce the risk of chronic conditions such as hypertension and diabetes. The objective of this study was to estimate the potential effect of replacing PRM for fatty fish, lean fish, red meat, eggs, pulses, or vegetables, on the risk of incident hypertension and diabetes. This was a prospective study of women in the E3N cohort study. Cases of diabetes and hypertension were based on self-report, specific questionnaires, and drug reimbursements. In the main analysis, information on regular dietary intake was assessed with a single food history questionaire, and food substitutions were modelled using cox proportional hazard models. 95 % confidence intervals were generated via bootstrapping. 71 081 women free of diabetes and 45 771 women free of hypertension were followed for an average of 18·7 and 18·3 years, respectively. 2681 incident cases of diabetes and 12 327 incident cases of hypertension were identified. Relative to PRM, fatty fish was associated with a 15 % lower risk of diabetes (HR = 0·85, 95 CI (0·73, 0·97)) and hypertension (HR = 0 85 (0·79, 0·91)). Between 3 and 10 % lower risk of hypertension or diabetes was also observed when comparing PRM with vegetables, unprocessed red meat or pulses. Relative to PRM, alternative protein sources such as fatty fish, unprocessed red meat, vegetables or pulses was associated with a lower risk of hypertension and diabetes.

Epidemiological studies on the association between the low carbohydrate diet (LCD) score and CVD risk factors have limited and inconsistent results. Data are from the baseline survey of Community-based Cohort Study on Nervous System Diseases. A total of 4609 adults aged ≥ 18 years were included in the study. Dietary data were assessed using a validated semi-quantitative FFQ. Multivariable logistic regression analyses were used to estimate relationships of three LCD scores with low HDL-cholesterol, high LDL-cholesterol, hypercholesterolaemia, hypertriacylglycerolaemia, impaired fasting glucose (IFG), high blood pressure and hyperuricaemia after adjusting for potential confounders. A higher LCD score was negatively associated with low HDL-cholesterol (OR: 0·65 (95 % CI 0·50, 0·83), P = 0·0001) and IFG (OR: 0·65 (95 % CI 0·51, 0·81), P = 0·001) after the final adjustment. However, there are sex differences in this result. Males in the highest quintile of the animal-based or plant-based LCD scores showed a decreased risk of low HDL-cholesterol, and females in the highest quintile of the animal-based or plant-based LCD scores showed a decreased risk of IFG than those in the lowest quintile of the LCD scores. These results suggest that sex differences should be considered when using LCD to treat dyslipidaemia and reduce fasting blood glucose.

The present study evaluated the association of food addiction (FA), the change of the BMI/age z-score and the consumption of ultra-processed foods in overweight students undergoing a 16-month, multicomponent intervention in the school environment. FA was investigated using the Yale Food Addiction Scale for Children, and the dietary assessment was estimated using the semi-quantitative FFQ in overweight 9–11-year-old students (BMI/age z-score ≥ 1) of both sexes at their baseline and after the intervention (n 120). Among the schoolchildren, 33·4 % had FA in at least one of the two assessments. The analysis of mixed-effects models to assess the effect of the intervention and the change of the BMI/age z-score between evaluations showed that the occurrence of FA influenced the maintenance of weight (time#FA, β = 0·30, 95 % CI 0·05, 0·54, P = 0·016). Weight loss was observed only in individuals who did not present FA (BMI/age z-score = −0·3). When evaluating the effect of the intervention and the dietary variables, we verified a reduction in the consumption of sugary milk-based drinks −71·13 kJ (–17 kcal), P = 0·04 only in non-FA students at the end of the study. FA has been identified as an underlying factor with therapeutic relevance, and an enhanced understanding of FA can open new paths for the prevention and management of obesity.

Avocado is a fruit rich in dietary fibre, potassium, Mg, mono and PUFA and bioactive phytochemicals, which are nutritional components that have been associated with cardiovascular health. Yet, despite the boom in avocado consumption, we lack evidence on its association with CVD risk in the general population. To estimate the prospective association between avocado consumption and incident hypertension in Mexican women, we estimated the association in participants from the Mexican Teachers’ Cohort who were ≥ 25 years, free of hypertension, CVD and cancer at baseline (n 67 383). We assessed baseline avocado consumption with a semi-quantitative FFQ (never to six or more times per week). Incident hypertension cases were identified if participants self-reported a diagnosis and receiving treatment. To assess the relation between categories of avocado consumption (lowest as reference) and incident hypertension, we estimated incidence rate ratios (IRR) and 95 % CI using Poisson regression models and adjusting for confounding. We identified 4002 incident cases of hypertension during a total of 158 706 person-years for a median follow-up of 2·2 years. The incidence rate of hypertension was 25·1 cases per 1000 person-years. Median avocado consumption was 1·0 (interquartile range: 0·23, 1·0) serving per week (half an avocado). After adjustment for confounding, consuming 5 + servings per week of avocado was associated with a 17 % decrease in the rate of hypertension, compared with non- or low consumers (IRR = 0·83; 95 % CI: 0·70, 0·99; P trend = 0·01). Frequent consumption of avocado was associated with a lower incidence of hypertension.

Mechanistic studies have suggested that antioxidants have beneficial effects on age-related macular degeneration (AMD). This study aimed to investigate the association between the types and sources of dietary vitamin and carotenoid intakes and AMD risk in China. A matched case–control study of 260 AMD cases and 260 matched controls was performed. The participants were interviewed for dietary information and potential confounders, and comprehensive ophthalmic examinations were performed. Conditional logistic models were used to estimate the odds ratio (OR) and 95 % confidence interval (CI) of specific vitamins and carotenoids and their main sources. When comparing the extreme quartiles, the ORs (95 % CI) were 0·30 (0·10, 0·88) for lutein and 0·28 (0·11, 0·74) for β-cryptoxanthin. The associations for other dietary vitamin and carotenoid intakes were generally weaker and non-significant. Higher intakes of spinach and egg, which are important sources of lutein, were associated with a reduced odds of AMD. ORs (95% CIs) comparing extreme categories were 0·42 (0·20, 0·88) for spinach and 0·52 (95% CI: 0·27, 0·98) for egg. Participants who were in the highest category of both egg intake and spinach intake had a much greater reduced odds of having AMD (OR: 0·23; 95% CI: 0·08, 0·71) than those in the lowest category of egg intake and spinach intake. In conclusion, a higher intake of lutein and lutein-rich foods was associated with a significantly decreased odds of AMD. These findings provide further evidence of the benefits of lutein and lutein-rich foods in the prevention of AMD.