Women in South Asia, including Nepal, have some of the poorest nutritional indicators globally, leading to poor maternal and child health outcomes. Nepal also suffers from high levels of household food insecurity, and newly married women are at high risk. Intra-household relationships may mediate the relationship between food insecurity and women’s nutrition in Nepal for newly married women. Our aim is to understand how newly married, preconception, women’s food consumption changes when she enters her husband’s home, compared with her natal home. We also explore whether relationship quality with husbands and mothers-in-law mediates the association between food insecurity and eating less high-quality food, using structural equation modelling.

Cross-sectional survey data.

Rural Nepal in 2018.

Data were collected from 200 newly married, preconception women.

Women had poor diet quality, and most ate fewer high-quality foods important for pregnancy in their marital, compared with natal, home. Higher quality relationships with mothers-in-laws mediated the association between food insecurity and a woman eating fewer high-quality foods in her marital, compared with natal, home. Relationship quality with husbands was not associated with changes in food consumption.

Preconception, newly married women in Nepal are eating less high-quality foods important for women’s health during the preconception period – a key period for avoiding adverse maternal and infant health outcomes. Relationships with mothers-in-law are key to women’s access to high-quality food, suggesting that interventions aiming to improve maternal and child nutrition should target all household members.

To identify correlates and underlying beliefs regarding the adolescents’ intention to abstain from consuming sugar-sweetened beverages (SSB) and the consumption of ≤1 daily portion of SSB.

Correlational study.

Region of Chaudière-Appalaches in the province of Quebec, Canada.

311 adolescents aged 13–18 years completed a self-administrated online questionnaire based on the Reasoned Action Approach. Frequency and quantity of different types of SSB within the past month were measured.

Total mean SSB intake was 882·6 ml/d (654·0 kJ/d ). Only 11·3 % abstained from SSB within the last month. Intention to abstain from SSB was explained by identification as SSB abstainers (β = 0·47), perceived norm (β = 0·32), attitude (β = 0·30), age 13–14 years (β = –0·27) and perception of the school environment (β = 0·14), which explained 66 % of the variance. Consumption of ≤1 daily portion of SSB was explained by the intention to abstain (OR = 1·55; 95 % CI 1·14, 2·11), perceived behavioural control to abstain (OR = 1·80; 95 % CI 1·29, 2·52), sex (girls v. boys: OR = 2·34; 95 % CI 1·37, 3·98) and socio-economic status (advantaged v. disadvantaged school: OR = 2·08; 95 % CI 1·21, 3·56). Underlying beliefs (i.e. more energy, decreased risk of addiction and friends’ approval) associated with intention as well as perceived barriers (e.g. access to SSB, after an activity that makes you thirsty), and facilitating factors (e.g. access to water) linked to SSB consumption were identified.

The results can inform public health interventions to decrease SSB consumption and their associated health problems among adolescents.

To obtain projections of the prevalence of childhood malnutrition indicators up to 2030 and to analyse the changes of wealth-based inequality in malnutrition indicators and the degree of contribution of socio-economic determinants to the inequities in malnutrition indicators in Bangladesh. Additionally, to identify the risk factors of childhood malnutrition.

Cross-sectional study. A Bayesian linear regression model was used to estimate trends and projections of malnutrition. For equity analysis, slope index, relative index and decomposition in concentration index were used. Multilevel logistic models were used to identify risk factors of malnutrition.

Household surveys in Bangladesh from 1996 to 2014.

Children under the age of 5 years.

A decreasing trend was observed for all malnutrition indices. In 1990, predicted prevalence of stunting, wasting and underweight was 55·0, 15·9 and 61·8 %, respectively. By 2030, prevalence is projected to reduce to 28·8 % for stunting, 12·3 % for wasting and 17·4 % for underweight. Prevalence of stunting, wasting and underweight were 34·3, 6·9 and 32·8 percentage points lower in the richest households than the poorest households. Contribution of the wealth index to child malnutrition increased over time and the largest contribution of pro-poor inequity was explained by wealth index. Being an underweight mother, parents with a lower level of education and poorer households were the key risk factors for stunting and underweight.

Our findings show an evidence-based need for targeted interventions to improve education and household income-generating activities among poor households to reduce inequalities and reduce the burden of child malnutrition in Bangladesh.

To describe out-of-home consumption according to the purpose and extent of industrial processing and also evaluate the association between eating out and ultra-processed food consumption, taking account of variance within and between individuals.

Cross-sectional study.

Brazil.

The study was based on the Individual Food Intake of the Brazilian Household Budget Survey, carried out with 34 003 individuals aged 10 years or more, between May 2008 and May 2009. All food items were classified according to food processing level. The habit of eating out was evaluated through the frequency of days each individual reported eating out, described according to sociodemographic characteristics. The contribution of food energy per group and subgroup was estimated according to the frequency of eating out. In addition, multilevel modelling was employed to evaluate the association between eating out and ultra-processed food consumption.

In Brazil, culinary preparations accounted for most of the energy eaten out. However, it was possible to observe a higher contribution of ultra-processed foods, especially sugary beverages and ready-to-eat meals, as the frequency of out-of-home consumption increased. Compared with food consumption exclusively at home, eating out increased the consumption of ultra-processed foods by 0·41 percentage points within and between individuals.

In Brazil, the same individual and different individuals had greater consumption of ultra-processed foods when they ate out of home compared with when they ate at home. So, it is necessary to implement public policies which discourage the out-of-home consumption of ultra-processed foods and that provide affordable and accessible less-processed food options.

To determine the level of vitamin D and to identify the association between vitamin D and depressive symptoms in apparently healthy Korean male adults.

A retrospective study design. Among 43 513 participants between 1 March and 30 November 2018, after eliminating participants with a history of depression or vitamin D deficiency, 9058 were included. To determine the level of vitamin D, serum 25-hydroxyvitamin D [25(OH)D] was measured. To assess the level of depression, the Korean version of the Center for Epidemiologic Studies Depression Scale (CES-D) was used.

South Korea.

Male adults who underwent routine health check-ups.

The average vitamin D level was 22·31 ± 7·09 ng/ml as 25(OH)D, while the number of subjects in the vitamin D insufficiency group with a finding of <20 ng/ml was 3783 (41·8 %). The mean CES-D score in all subjects was 8·31 ± 5·97 points, and the proportion of the depressive symptoms group with a score of ≥16 was 8·71 %. The OR of patients in the depressive symptoms group also being in the insufficiency group was found to be 1·49 (95 % CI 1·12, 2·00).

A total of 41·8 % of apparently healthy male adults had vitamin D levels <20 ng/ml. We identified an association between vitamin D insufficiency and depressive symptoms in apparently healthy Korean male adults.

To determine how children interpret terms related to food processing; whether their categorisation of foods according to processing level is consistent with those used in research; and whether they associate the degree of processing with healthfulness.

Qualitative data were collected from ten focus groups. Focus groups were audio-recorded, transcribed verbatim, and thematic analysis was conducted.

Four elementary and afterschool programmes in a large, urban school district in the USA that served predominantly low-income, racial/ethnic minority students.

Children, 9–12 years old, in the fourth–sixth grades (n 53).

The sample was 40 % male, 47 % Hispanic with a mean age of 10·4 ± 1·1 years. Children’s understanding of unprocessed foods was well aligned with research classifications, while concordance of highly processed foods with research categorisations varied. Five primary themes regarding the way children categorised foods according to their processing level emerged: type and amount of added ingredients; preparation method; packaging and storage; change in physical state or sensory experience; and growing method. Most children associated processing level with healthfulness, describing unprocessed foods as healthier. The most common reason provided for the unhealthfulness of processed foods was added ingredients, including ‘chemicals’ and ‘sugar’.

The current study demonstrated that children have a working knowledge of processing that could be leveraged to encourage healthier eating patterns; however, their understanding is not always consistent with the classification systems used in research. The vocabulary used by researchers and consumers to talk about processing must be reconciled to translate findings into actionable messages.

In the current meta-analysis, we aimed to systematically review and summarize eligible studies for the association between dietary inflammatory index (DII) and blood pressure, hypertension (HTN) and glucose homeostasis biomarkers.

In a systematic search of PubMed, Scopus and Google Scholar electronic databases up to February 2019, relevant studies were included in the literature review. Observational studies evaluating the association between DII and HTN, hyperglycaemia, systolic blood pressure (SBP), diastolic blood pressure (DBP), fasting blood glucose (FBG), insulin, homeostatic model assessment of insulin resistance (HOMA-IR) and glycated Hb (HbA1c) were included.

Not applicable.

Total numbers of studies were as follows: OR for DII and HTN (n 12), OR for DII and hyperglycaemia (n 9), HTN prevalence (n 9), mean (sd) of SBP and DII (n 12), mean (sd) of DBP and DII (n 10), mean (sd) of FBS and DII (n 13), mean (sd) of HbA1c and DII (n 3), mean (sd) of insulin and DII (n 6), mean (sd) of HOMA-IR and DII (n 7). Higher DII scores were associated with higher odds of HTN (OR = 1·13; 95 % CI 1·01, 1·27; P < 0·001), SBP (weighted mean difference (WMD) = 1·230; 95 % CI 0·283, 2·177; P = 0·011), FBS (WMD = 1·083; 95 % CI 0·099, 2·068; P = 0·031), insulin (WMD = 0·829; 95 % CI 0·172, 1·486; P = 0·013), HbA1c (WMD = 0·615; 95 % CI 0·268, 0·961; P = 0·001) and HOMA-IR (WMD = 0·192; 95 % CI 0·023, 0·361; P = 0·026) values compared with lowest DII categories.

Lower inflammatory content of diets for prevention of cardiovascular risk factors is recommended.

Obesity is a risk factor for severe complications and death from the coronavirus disease 2019 (COVID-19). Public health efforts to control the pandemic may alter health behaviors related to weight gain, inflammation, and poor cardiometabolic health, exacerbating the prevalence of obesity, poor immune health, and chronic diseases.

We reviewed how the pandemic adversely influences many of these behaviors, specifically physical activity, sedentary behaviors, sleep, and dietary intakes, and provided individual level strategies that may be used to mitigate them.

At the community level and higher, public health and health care professionals need to advocate for intervention strategies and policy changes that address these behaviors, such as increasing nutrition assistance programs and creating designated areas for recreation and active transportation, to reduce disparities among vulnerable populations.

The long-lasting impact of the pandemic on health behaviors, and the possibility of a second COVID-19 wave, emphasize the need for creative and evolving, multi-level approaches to assist individuals in adapting their health behaviors to prevent both chronic and infectious diseases.

This study systematically reviewed literature on the neighbourhood food environment in relation to diet and obesity among residents in China.

A keyword search of peer-reviewed articles was performed in Cochrane Library, PubMed, and Web of Science. Eligibility criteria include study designs: longitudinal/cohort studies or cross-sectional studies; study participants: people of all ages; exposures: neighbourhood food environment (e.g. restaurants, supermarkets, wet markets, fast-food restaurants, or convenience stores); outcomes: diet and/or body weight status; and country: China.

Seventeen studies met all criteria and were included. Among the eight studies that assessed the neighbourhood food environment in relation to diet, six reported at least one statistically significant relationship in the expected direction, whereas the remaining two exclusively reported null effects. Among the eleven studies that assessed the neighbourhood food environment in relation to body weight or overweight/obesity, ten reported a significant association whereas the remaining one reported a null relationship. Variety, density, and proximity of food outlets were positively associated with local residents’ dietary diversity, portion size, and daily caloric intake. Density and proximity of fast-food restaurants and convenience stores were positively associated with local residents’ adiposity in some but not all studies. Evidence linking any specific food outlet type to diet/obesity remains lacking due to the small number of studies and heterogeneities in food environment measures, geographical locations, and population subgroups.

The neighbourhood food environment may influence diet and obesity among Chinese residents but the evidence remains preliminary. Future studies adopting an experimental study design and objective/validated environment and dietary measures are warranted.

To compare the prevalence of malnutrition (undernutrition and excess weight) by wealth, education level, ethnicity and urban/rural areas in Mexican children and women of reproductive age.

We compared the prevalence of overweight, obesity, wasting/underweight, stunting/short stature and anaemia by socioeconomic and ethnic indicators. For each indicator, we estimated prevalence ratios (PR) adjusted by all other socioeconomic and ethnic indicators. We analysed if results differed by urban/rural areas.

Mexican National Health and Nutrition Survey 2012.

Children <5 years, non-pregnant women 11–19 years and non-pregnant women 20–49 years (n 33 244).

In most age groups, belonging to non-indigenous households, with high wealth, high education and in urban areas were inversely associated with stunting or short stature (PR ranging from 0·40 to 0·83), and wealth and education were inversely associated with anaemia (PR ranging from 0·53 to 0·78). The prevalence of overweight was similar across subgroups among children <5 years; however, among women 11–19 years, wealth, non-indigenous household and urban areas were positively associated (PR ranging from 1·16 to 1·33); and among women 20–49 years, education was inversely associated (PR 0·83).

Socially disadvantaged populations have a higher prevalence of undernutrition, whereas the prevalence of excess weight is either equal (children <5 years), slightly lower (women 11–19 years) or even higher (women 20–49 years) with lower education. These results highlight the need for specific actions to address social inequalities in malnutrition in the Mexican population.

To examine public commitments for encouraging United States consumers to make healthy dietary purchases with their Supplemental Nutrition Assistance Program (SNAP) benefits among of prevalent SNAP-authorised retailers.

National SNAP-authorised retail landscape in addition to stores located in California and Virginia, two states targetted for a Partnership for a Healthier America pilot social marketing campaign.

SNAP-authorised retailers with the most store locations in selected settings.

A review of retailers’ publicly available business information was conducted (November 2016–February 2017). Webpages and grey literature sources were accessed to identify corporate social responsibility (CSR) reports and commitments describing strategies to encourage healthy consumer purchases aligned with the 2015–2020 Dietary Guidelines for Americans. Evidence was organised using a marketing-mix and choice-architecture (MMCA) framework to characterise strategies used among eight possible types (i.e. place, profile, portion, pricing, promotion, priming, prompting and proximity).

Of the SNAP-authorised retailers (n 38) reviewed, more than half (n 20; 52·6 %) provided no information in the public domain relevant to the research objective. Few retailers (n 8; 21·1 %) had relevant CSR information; grey literature sources (n 52 articles across seventeen retailers) were more commonly identified. SNAP-authorised retailers in majority committed to increasing the number of healthy products available for purchase (profile).

Substantial improvements are needed to enhance the capacity and commitments of SNAP-authorised retailers to use diverse strategies to promote healthy purchases among SNAP recipients. Future research could explore feasible approaches to improve dietary behaviours through sector changes via public–private partnerships, policy changes, or a combination of government regulatory and voluntary business actions.

To investigate changes in socio-economic inequalities in growth in height, weight, BMI and grip strength in children born during 1955–1993 in Guatemala, a period of marked socio-economic-political change.

We modelled longitudinal data on height, weight, BMI and hand grip strength using Super-Imposition by Translation and Rotation (SITAR). Internal Z-scores summarising growth size, timing and intensity (peak growth velocity, e.g. cm/year) were created to investigate inequalities by socio-economic position (SEP; measured by school attended). Interactions of SEP with date of birth were investigated to capture secular changes in inequalities.

Urban and peri-urban schools in the region of Guatemala City, Guatemala.

Participants were 40 484 children and adolescents aged 3–19 years of Ladino and Maya ancestry (nobservations 157 067).

The difference in height (SITAR size) between lowest and highest SEP decreased from −2·0 (95 % CI −2·2, −1·9) sd to −1·4 (95 % CI −1·5, −1·3) sd in males, and from −2·0 (95 % CI −2·1, −1·9) sd to −1·2 (95 % CI −1·3, −1·2) sd in females over the study period. Inequalities also reduced for weight, BMI and grip strength, due to greater secular increases in lowest-SEP groups. The puberty period was earlier and shorter in higher-SEP individuals (earlier SITAR timing and higher SITAR intensity). All SEP groups showed increases in BMI intensity over time.

Inequality narrowed between the 1960s and 1990s. The lowest-SEP groups were still >1 sd shorter than the highest. Risks remain for reduced human capital and poorer population health for urban Guatemalans.

We aimed to estimate the cost-effectiveness of brief weight-loss counselling by dietitian-trained practice nurses, in a high-income-country case study.

A literature search of the impact of dietary counselling on BMI was performed to source the ‘best’ effect size for use in modelling. This was combined with multiple other input parameters (e.g. epidemiological and cost parameters for obesity-related diseases, likely uptake of counselling) in an established multistate life-table model with fourteen parallel BMI-related disease life tables using a 3 % discount rate.

New Zealand (NZ).

We calculated quality-adjusted life-years (QALY) gained and health-system costs over the remainder of the lifespan of the NZ population alive in 2011 (n 4·4 million).

Counselling was estimated to result in an increase of 250 QALY (95 % uncertainty interval −70, 560 QALY) over the population’s lifetime. The incremental cost-effectiveness ratio was 2011 $NZ 138 200 per QALY gained (2018 $US 102 700). Per capita QALY gains were higher for Māori (Indigenous population) than for non-Māori, but were still not cost-effective. If willingness-to-pay was set to the level of gross domestic product per capita per QALY gained (i.e. 2011 $NZ 45 000 or 2018 $US 33 400), the probability that the intervention would be cost-effective was 2 %.

The study provides modelling-level evidence that brief dietary counselling for weight loss in primary care generates relatively small health gains at the population level and is unlikely to be cost-effective.

To compare the distribution of malnutrition by socio-economic indicators (SEI) in Peruvian children under 5 years and women of reproductive age (WRA).

We analysed data from the National Demographic and Family Health Survey. WHO criteria were used to define malnutrition indicators (overweight/obesity combined (OW); wasting/underweight; stunting/short stature; anaemia). Linear combination test was used to compare the prevalence of malnutrition by SEI (wealth index as a proxy of socio-economic status (SES); education; ethnicity). Prevalence ratio (PR) was used to describe disparities and associations between malnutrition and SEI.

Peru (2015).

Children (n 22 833) under 5 years and WRA (n 33 503; 5008 adolescents and 28 495 adults).

The most prevalent form of malnutrition was anaemia (32·0 %) in children and OW in adolescent and adult WRA (31·3 and 65·1 %, respectively). Adjusted models showed that stunting and anaemia were significantly lower among children with high SES (PR = 0·25, 0·67), high-educated mothers (PR = 0·26, 0·76) and higher in indigenous children (PR = 1·3, 1·2); conversely, OW was higher among those with high SES and high-educated mothers (PR = 1·8, 1·6) compared with their lowest counterparts. In WRA, stunting/short stature was lower among those with high SES, high education and higher in indigenous adult women. OW in adolescents and adults was higher in high SES (PR = 1·4, 1·1), lower in indigenous adult women (PR = 0·84) and lower in high-educated adult women (PR = 0·86).

In the studied population, the distribution of malnutrition was associated with SEI disparities. Effective policies that integrate actions to overcome the double burden of malnutrition and reduce disparities are needed.

To explore the conceptualisation of healthy food by citizens and how they judge the healthiness of ultra-processed foods.

Four focus groups were conducted using a semi-structured discussion guide. Focus group discussions were held about the concept of healthy food, what characterise a healthy product and healthiness perception of ultra-processed products. Transcripts of the focus groups were analysed following an inductive coding approach.

Uruguay, one of the Latin American countries with the highest prevalence of overweight and obesity.

Fifty-two adult Uruguayan participants, diverse in terms of gender, age, educational level and socio-economic status.

In agreement with previous studies on lay perceptions of healthy eating, the conceptualisation of healthy food was mainly focused on food characteristics. Although participants regarded lack of processing as a cue for healthiness, they did not categorise all ultra-processed products as unhealthy. Albeit some product categories were automatically regarded as unhealthy, participants considered that other categories could include healthy and unhealthy products. In such cases, they explicitly referred to several simplified cognitive strategies to judge whether an ultra-processed product is healthy or not. Results showed that participants tended to rely on simple cues, such as label design, nutrient claims, brand, price and country of origin as indicators of product healthiness.

Healthiness perception of ultra-processed products seems to be largely influenced by heuristics, which stresses the need to implement policies that make the potential negative effects of ultra-processed products salient.

To assess whether a community water service is associated with the frequency of sugar-sweetened beverages (SSB) consumption, obesity, or perceived health status in rural Alaska.

We examined the cross-sectional associations between community water access and frequency of SSB consumption, body mass index categories, and perceived health status using data from the 2013 and 2015 Alaska Behavioral Risk Factor Surveillance System (BRFSS). Participants were categorized by zip code to ‘in-home piped water service’ or ‘no in-home piped water service’ based on water utility data. We evaluated the univariable and multivariable (adjusting for age, household income and education) associations between water service and outcomes using log-linear survey-weighted generalized linear models.

Rural Alaska, USA.

Eight hundred and eighty-seven adults, aged 25 years and older.

In unadjusted models, participants without in-home water reported consuming SSB more often than participants with in-home water (1·46, 95 % CI: 1·06, 2·00). After adjustment for potential confounders, the effect decreased but remained borderline significant (1·29, 95 % CI: 1·00, 1·67). Obesity was not significantly associated with water service but self-reported poor health was higher in those communities without in-home water (1·63, 95 % CI: 1·05, 2·54).

Not having access to in-home piped water could affect behaviours surrounding SSB consumption and general perception of health in rural Alaska.

Children breast-fed during infancy consume more fruits and vegetables than formula-fed children. This pattern is likely due, in part, to infant learning from flavours of the mother’s diet transmitted through breast milk, but more research is needed to understand associations between early flavour exposures and later dietary patterns. We examined whether breast-feeding and maternal fruit and vegetable consumption during nursing were synergistically associated with higher child fruit and vegetable consumption.

Prospective cohort study of breast-feeding duration, maternal diet postpartum and child diet. Complete breast-feeding and maternal diet data were available for 1396 mother–child dyads; multiple imputation was used for missing data in other variables. In separate multivariable logistic regression models, we estimated the adjusted odds of high child fruit or vegetable consumption at 12 months or 6 years as a function of breast-feeding duration, maternal fruit or vegetable consumption during nursing, and their interaction.

The Infant Feeding Practices Study II and Year 6 Follow-Up.

Mother–child dyads followed from birth to 6 years during 2005–2012 in the USA.

Longer breast-feeding duration was associated with high child fruit and vegetable consumption at 12 months. At 6 years, the interaction between breast-feeding duration and maternal vegetable consumption was associated with high child vegetable consumption.

Higher maternal vegetable consumption and longer breast-feeding duration were synergistically associated with high child vegetable consumption at 6 years, independent of sociodemographic characteristics and fruit and vegetable availability. Exposures to vegetable flavours through breast milk may promote later child vegetable consumption.

To examine how household food insecurity is related to adolescent weight status and disordered eating.

Cross-sectional, population-based study. Adolescents self-reported unhealthy weight control behaviours, binge eating and meal frequency; weight status was measured. Household food insecurity was assessed by asking parents to respond to the validated six-item US Household Food Security Survey Module.

Adolescents surveyed within Minneapolis/St. Paul public middle and high schools completed surveys at school, and their parents/guardians were surveyed by mail during the 2009–2010 academic year.

Ethnically/racially diverse, primarily low-income adolescents (mean age: 14·4 years, range: 10–22 years) and their parents/guardians (n 2285 dyads).

More than one-third (38·9 %) of the adolescents experienced past-year household food insecurity, 43·2 % reported disordered eating and 39·6 % were overweight. Generalised regression models showed that food insecure (FI) compared with food secure (FS) adolescents had higher prevalence of overweight (FI: 42·3 % v. FS: 37·9 %, P = 0·039), lower breakfast consumption (FI: 4·1 times/week v. FS: 4·4 times/week, P = 0·005) and greater use of unhealthy weight control behaviours (FI: 49·0 % v. FS: 39·5 %, P < 0·001) in unadjusted models. Models adjusted for parental education, ethnicity/race, sex and age found that food insecurity was associated with higher prevalence of unhealthy weight control behaviours (FI: 44·5 % v. FS: 37·8 %, P = 0·007), but not with weight status or other eating behaviours.

These results suggest that food insecurity may be an independent risk factor for unhealthy weight control behaviours, indicating a need to approach these intersecting issues in a comprehensive manner.

We aimed to profile vegetable consumption and its association with dietary and sociodemographic factors.

Secondary analysis of a nationally representative nutrition survey. ‘Vegetables’ refers to non-discretionary ‘vegetables and legumes/beans’ as defined by the Australian Dietary Guidelines (ADG). Prevalence of vegetable consumption, frequency of intake, proportion meeting ADG recommendations, most popular food groups, intake at each reported eating occasion, and the profile of high and low vegetable consumers (based on the median servings) were determined.

Australian 2011–2012 National Nutrition and Physical Activity Survey.

Children and adolescents aged 2 to 18 years (n 2812).

Vegetables were consumed by 83·0% (95% CI 81·6, 84·4%) of participants, but the median vegetable servings was less than a third of the ADG recommendations. ‘Leaf and stalk vegetables’ and ‘potatoes’ were the most popular vegetable-dense food groups at lunch and dinner, respectively. Sixty-four percent had vegetables once a day, and predominantly at dinner. Vegetable frequency was positively associated with daily vegetable servings and variety. Participants who consumed vegetables twice a day generally had vegetables at both lunch and dinner and had nearly double the servings (2·6, sd 1·9) of those who consumed them once (1·5, sd 1·5). High vegetable consumers were older, had higher total energy, but lower discretionary energy intake and were less likely to be at risk of metabolic complications.

Increasing the frequency of vegetable consumption may assist with increasing daily vegetable servings. A focus on consuming vegetables at lunch may assist with increasing both total servings and variety.