Food security is defined as being able to access enough food that will help maintain an active, healthy lifestyle for those living in a household. While there are no studies on food security issues among deaf people, research shows that communication barriers early in life are linked to poor physical and mental health outcomes. Childhood communication barriers may also risk later food insecurity.

A single food security screener question found to have 82 % sensitivity in classifying families who are at risk for food insecurity was taken from the six-item US Household Food Security Survey Module. Questions related to food insecurity screener, depression diagnosis and retrospective communication experience were translated to American Sign Language and then included in an online survey. Over 600 deaf adult signers (18–95 years old) were recruited across the USA.

After adjusting for covariates, deaf adults who reported being able to understand little to none of what their caregiver said during their formative years were about five times more likely to often experience difficulty with making food last or finding money to buy more food, and were about three times more likely to sometimes experience this difficulty, compared with deaf adults who reported to being able to understand some to all of what their caregiver said.

Our results have highlighted a marked risk for food insecurity and related outcomes among deaf people. This should raise serious concern among individuals who have the potential to effect change in deaf children’s access to communication.

To unravel the concept of nutrition awareness, as it relates to risky personal nutrition-related behaviours, and to assess the sociodemographic and psychosocial correlates of nutrition awareness.

Data were collected in a cross-sectional study with the aid of a face-to-face interview-assisted questionnaire that was based on the Precaution Adoption Process Model and Stages of Change Model.

Dutch consumer homes.

Six hundred and three Dutch adults aged 18 to 80 years, selected from a panel.

Our model explains nutrition awareness well (explained variance 53.7%). Psychosocial correlates were involvement with nutrition, health awareness, association with healthy food, perceived relevance of eating less fat, association with necessity, perceived relevance of vitamins, and perceived attributes of independent organisations. Sociodemographic correlates were gender and age. The relationship between nutrition awareness and nutrition-related behaviours proved to be very complicated.

The value of our study is that it unravels the concept of nutrition awareness. Understanding the correlates of nutrition awareness can contribute to a more effective application of behavioural change models. Our results support increasing involvement with nutrition through personalising and tailoring to the motivational stage.

To compare estimates from one day with usual intake estimates to evaluate how the adjustment for within-person variability affected nutrient intake and adequacy in Mexican children.

In order to obtain usual nutrient intakes, the National Cancer Institute’s method was used to correct the first 24 h dietary recall collected in the entire sample (n 2045) with a second 24 h recall collected in a sub-sample (n 178). We computed estimates of one-day and usual intakes of total energy, fat, Fe, Zn and Na.

2012 Mexican National Health and Nutrition Survey.

A total of 2045 children were included: 0–5·9 months old (n 182), 6–11·9 months old (n 228), 12–23·9 months old (n 537) and 24–47·9 months old (n 1098). From these, 178 provided an additional dietary recall.

Although we found small or no differences in energy intake (kJ/d and kcal/d) between one-day v. usual intake means, the prevalence of inadequate and excessive energy intake decreased somewhat when using measures of usual intake relative to one day. Mean fat intake (g/d) was not different between one-day and usual intake among children >6 months old, but the prevalence of inadequate and excessive fat intake was overestimated among toddlers and pre-schoolers when using one-day intake (P<0·05). Compared with usual intake, estimates from one day yielded overestimated prevalences of inadequate micronutrient intakes but underestimated prevalences of excessive intakes among children aged >6 months.

There was overall low variability in energy and fat intakes but higher for micronutrients. Because the usual intake distributions are narrower, the prevalence of inadequate/excessive intakes may be biased when estimating nutrient adequacy if one day of data is used.

People eat more unhealthy foods when served more (portion size effect) and when the food is served in larger units (unit size effect). The present study aimed to examine whether these effects can be used for the good: to increase vegetable consumption among children.

A 2×2 between-subjects experiment was conducted at two schools. Pupils were presented in class with cucumber that varied both in unit size (one piece v. pre-sliced) and portion size (one-third v. two-thirds of a cucumber). Children ate ad libitum during the morning break and filled in a survey.

Primary schools in the centre of the Netherlands.

Primary-school pupils (n 255) aged 8–13 years.

Children ate 54 % more cucumber when served a large compared with a smaller portion (difference of 49 g; P<0·001). Large units did not impact consumption (P=0·58), but were considered as less convenient to eat than small units (P=0·001).

Findings suggest that children’s vegetable intake can be improved by serving larger portions in smaller-sized pieces.

To identify whether picky eating during childhood is associated with dietary intake, weight status and disordered eating behaviour during young adulthood.

A population-based study using data from young adults who responded online or by mail to the third wave of the Project EAT (Eating and Activity in Teens and Young Adults) study in 2008–2009. Participants retrospectively reported the extent to which they were a picky eater in childhood, sociodemographic characteristics, disordered eating behaviours, usual dietary intake, and weight and height.

Participants were initially recruited in the Minneapolis/St. Paul metropolitan area of Minnesota, USA, in 1998–1999.

The analytic sample included 2275 young adults (55 % female, 48 % non-Hispanic White, mean age 25·3 (sd 1·6) years).

Young adults who reported picky eating in childhood were found to currently have lower intakes of fruit, vegetables and whole grains, and more frequent intakes of snack foods, sugar-sweetened beverages and foods from fast-food restaurants. No associations were observed between picky eating in childhood and young adults’ weight status, use of weight-control strategies or report of binge eating.

While young adults who report picky eating during childhood are not at higher risk for disordered eating, those who were picky eaters tend to have less healthy dietary intake. Food preferences and dietary habits established by picky eaters during childhood may persist into adulthood.

The present study examined the energy (kilojoule) content of Australian fast-food menu items over seven years, before and after introduction of menu board labelling, to determine the impact of the introduction of the legislation.

Analysis of the median energy contents per serving and per 100g of fast-food menu items. Change in energy content of menu items across the years surveyed and differences in energy content of standard and limited-time only menu items were analysed.

Five of Australia’s largest fast food chains: Hungry Jack’s, KFC, McDonald’s, Oporto and Red Rooster.

All standard and limited-time only menu items available at each fast-food chain, collected annually for seven years, 2009–2015.

Although some fast-food chains/menu item categories had significant increases in the energy contents of their menus at some time points during the 7-year period, overall there were no significant or systematic decreases in energy following the introduction of menu labelling (P=0·19 by +17 kJ/100 g, P=0·83 by +8 kJ/serving). Limited-time only items were significantly higher in median energy content per 100 g than standard menu items (+74 kJ/100 g, P=0·002).

While reformulation across the entire Australian fast-food supply has the potential to positively influence population nutrient intake, the introduction of menu labelling legislation in New South Wales, Australia did not lead to reduced energy contents across the five fast-food chains. To encourage widespread reformulation by the fast-food industry and enhance the impact of labelling legislation, the government should work with industry to set targets for reformulation of nutrient content.

Childhood cancer survivors (CCS) have been shown to practise suboptimal dietary intake and may benefit from nutrition interventions during and after treatment. Cooking classes have become popular for encouraging healthy eating behaviours in community-based programming and academic research; however, literature on teaching cooking classes in CCS is limited. The purpose of the present study was to address the development and implementation of classes for CCS based on a recently developed framework of healthy cooking behaviour.

A conceptual framework was developed from a systematic literature review and used to guide healthy cooking classes for CCS in different settings.

One paediatric cancer hospital inpatient unit, one paediatric cancer in-hospital camp programme and two off-site paediatric cancer summer camp programmes.

One hundred and eighty-nine CCS of varying ages and thirteen parents of CCS.

Seventeen classes were taught at camps and seven classes in the hospital inpatient unit. Healthy cooking classes based on the conceptual framework are feasible and were well received by CCS.

Cooking classes for CCS, both at the hospital and at camp, reinforced the principles of the conceptual framework. Future trials should assess the dietary and anthropometric impact of evidence-based healthy cooking classes in CCS.

Anaemia is a major contributor to the global disease burden and half of pregnant women in India were anaemic in 2016. The aetiology of anaemia is complex, yet anaemia determinants are frequently examined in isolation. We sought to explore how shifts in sociodemographic (wealth, age at pregnancy, education, open defecation, cooking fuel type, household size), programmatic (iron–folic acid tablet consumption, antenatal care visits) and dietary factors (intake of Fe, folic acid, vitamin B12, phytate) predicted changes in anaemia prevalence.

Nutrient levels for eighty-eight food items were multiplied by household consumption of these foods to estimate household-level nutrient supply. A synthetic panel data set was created from two rounds of the District Level Household and Facility Survey (2002–04 and 2012–13) and Household Consumer Expenditures Survey (2004–05 and 2011–12). Ordinary least-squares multivariate regression models were used.

Districts (n 446) spanning north, north-east, central and south India.

Pregnant women aged 15–49 years (n 17 138).

In the model accounting for both non-dietary and dietary factors, increased age at pregnancy (P<0·001), reduced village-level open defecation (P=0·001), consuming more Fe (P<0·001) and folic acid (P=0·018) and less phytate (P=0·002), and urbanization (P=0·015) were associated with anaemia reductions. A 10 mg increase in daily household Fe supply from 2012 levels was associated with a 10 % reduction in anaemia.

Public health interventions to combat anaemia in pregnant women should use a holistic approach, including promotion of delayed marriage, construction and use of toilets, and measures that facilitate adoption of nutrient-rich diets.

Few studies have considered the combined effects of home-related determinants on children's diet. The present study investigated independent associations between sociodemographic and food practice (SFP) characteristics and fruit and vegetable consumption in UK children and the combined effects of SFP on consumption using pattern analysis.

Diet was assessed using 4 d food diaries, SFP were collected using computer-assisted personal interview. Linear regressions were used to test associations; principal component analysis was used to identify patterns of SFP characteristics. Regression of fruit (g/d) and vegetables (g/d) v. component scores of each pattern were performed.

UK National Diet and Nutrition Survey Rolling Programme (2008–2010).

Children aged 1·5–10 years (n 642).

Significant associations were found between fruit and vegetable consumption and household socio-economic status. Pattern 1, which was positively correlated with household structure characteristics, was associated with increased fruit consumption (P < 0·001). Pattern 2, characterised by positive correlations for socio-economic status, fruit availability and organic food purchase, and negatively correlated with household size and the number of children per household, was associated with higher fruit and vegetable consumption (both P < 0·001). Pattern 3, characterised by high frequency of eating out and eating takeaway, was associated with a lower consumption of both fruit (P < 0·012) and vegetables (P < 0·023).

Patterns of SFP determinants may be more informative than individual characteristics in relation to dietary outcomes. Results have public health implications on the healthfulness of meals eaten out of home and in takeaways, as well as the need to reduce diet inequality in larger households with lower socio-economic status.

To validate a picture book for estimation of food portion sizes using two approaches: (i) ‘perception’ of food portions by comparison with a series of food photos; and (ii) ‘conceptualization and memory’, using the same photos to estimate the amount of served food one hour after self-served food portions.

Each partner developed a country-specific picture book based on the so-called EPIC-Soft picture book. Representative and common photo series were chosen achieving approximately 25 % of the original picture book (n 23). Three portions from each photo series were randomly selected.

The study was performed within the Pilot study in the view of a Pan-European dietary survey – Adolescents, adults and elderly (PILOT-PANEU) project.

A sample of adolescents and adults was recruited in five countries: Bulgaria (n 103), Finland (n 34), Germany (n 69), Hungary (n 62) and Portugal (n 77).

Among the portions of the corresponding photo series and depending on the type of food, from 18 % (cheese) to 96 % (ratatouille) of participants chose the correct portions. In the perception study, agreement between the portions shown and reported was substantial (intraclass correlation coefficient (ICC)=0·805) and the mean difference was very low. In the memory study, agreement between the served and reported portions was lower than in the perception study (ICC=0·536). Agreement also seemed to decrease as the appearance of food on the plate differed from food in the picture.

Overall, the picture series selected can be applied in future intake surveys to quantify foods similar to those depicted in the pictures.

In post-transitional countries, obesity disproportionally affects women. Longitudinal studies can detect high-risk groups in whom to target actions. We investigated the magnitude and velocity of BMI changes in Chilean women of reproductive age and evaluated whether these trends vary in specific groups.

Longitudinal study. We measured weight and height in 2007 (baseline) and again in 2010 (follow-up); we estimated change in BMI (weight/height2) within the 3-year period and assessed its relationship with age, years of education and parity, collected at baseline and follow-up using a questionnaire.

Population-based cohort of low- to middle-income Chilean women.

Seven hundred and sixty-one women of reproductive age (mean 32·0 (sd 7·0) years), mothers of children who participate in the Growth and Obesity Cohort Study (GOCS).

At baseline, 61 % of women had BMI ≥ 25·0 kg/m2. After 3 years, women gained on average 2·6 kg and obesity (BMI ≥ 30·0 kg/m2) increased by 23 % (12 % new obesity cases). Women with normal nutritional status gained more BMI than obese women (1·4 v. 0·6 kg/m2, P < 0·001). An increase in parity was positively associated with BMI change, independently of age, nutritional status and education (P < 0·05). Age and education were not associated with BMI change after controlling for other factors (P > 0·05).

In Chile, a post-transitional country, we observed an alarming increase in obesity among women of reproductive age. Our results indicate that in this population actions need to be targeted at all women irrespective of their nutritional status. A key component of these policies should be avoiding excessive weight gain during pregnancy.

To determine risk factors for consumption of soda and other sugar-sweetened beverages (SSB) among 2-year-old children.

The analysis was performed using three linked data sets: the 2004–2005 Oregon Pregnancy Risk Assessment Monitoring Survey (PRAMS); its longitudinal follow-up, 2006–2007 Oregon PRAMS-2; and 2004–2005 Oregon birth certificates.

PRAMS is a surveillance programme supported by the federal Centers for Disease Control and Prevention and implemented by participating state health departments. Using mixed methods, PRAMS surveys women 2–6 months after a live birth. Oregon PRAMS-2 re-interviews respondents shortly after the index child's second birthday. Oregon PRAMS oversamples minority women.

Using monthly cohorts, we randomly selected 5851 women from the 2004–2005 birth certificates. In total 1911 women completed both PRAMS and PRAMS-2. The weighted response rate of PRAMS-2 was 43·5 %.

Almost half of mothers (49·9 %) reported that their child drank SSB on at least 1 d/week. Mothers whose children drank SSB at least once weekly were more likely to have low income (adjusted OR = 2·83, 95 % CI 2·09, 3·83) and to eat out on ≥2 d/week (OR = 2·11 %, 95 % CI 1·66, 2·70). Hispanic and non-Hispanic black women were most likely to report that their child drank SSB at least once weekly.

Half of mothers reported that their 2-year-old children drank SSB at least once weekly. Public health interventions and policies should address childhood SSB consumption including educating health-care providers and parents.

Transition from a low-fat vegetable-rich rural diet to a high-fat Westernised diet is considered a factor in the escalating occurrence of vascular-related diseases and type 2 diabetes in urban black South Africans. Consumption of morogo is a distinguishing feature of rural African diets.

To determine fatty acid profiles and folate contents of three widely consumed, wild-growing, African dark green leafy vegetables (morogo).

GC–MS was applied for analysis of fatty acid composition and a validated microbiological assay conducted to determine folic acid contents of wild-growing morogo sampled from deep rural villages in three different geographical regions of South Africa.

Measured fatty acids ranged from 1610·2 to 2941·6 mg/100 g dry mass, with PUFA concentrations 1·4 to 2·8 times those of SFA. Calculated from the relative percentages of linoleic acid (18:2n-6) and linolenic acid (18:3n-3), the ratio of 18:2n-6 to 18:3n-3 PUFA was 1·0:3·4 to 1·0:8·9. The only MUFA was palmitoleic acid (16:1), measured at 34·7 (sd 0·3) to 79·0 (sd 9·3) mg/100 g dry mass, and the predominant SFA was palmitic acid (16:0), measured at 420·6 (sd 83·3) to 662·0 (sd 21·2) mg/100 g dry mass. Folic acid concentration varied from 72 to 217 μg/100 g fresh sample.

Morogo is low-fat food item high in folate and with 18:3n-3 in excess of 18:2n-6, the proposed anti-inflammatory effects of which may lower risks of vascular-related chronic diseases and type 2 diabetes.

To examine the extent to which food and beverage brands exhibit personalities on Twitter, quantify Twitter users’ engagement with posts displaying personality features and determine advertising spending across these brands on Twitter.

We identified 100 tweets from 10 food and beverage brands that displayed a ‘personality’, and 100 ‘control’ tweets (i.e. a post by that brand on the same day). Our codebook quantified the following personification strategies: (1) humour; (2) trendy language and (3) absence of food product mentions. We used media articles to quantify other personification strategies: (4) referencing trending topics; (5) referencing current events; (6) referencing internet memes and (7) targeting niche audiences. We calculated brands’ number of tweets, re-tweets, ‘likes’, and comments and report the relationship between advertising spending and retweets per follower.

Twitter posts.

Ten food and beverage brands that were described in media articles (e.g. Forbes) as having distinct personalities.

Personality tweets earned 123 013 retweets, 732 076 ‘likes’ and 14 806 comments, whereas control tweets earned 61 044 retweets, 256 105 ‘likes’ and 14 572 comments. The strategies used most included humour (n 81), trendy language (n 80) and trending topics (n 47). The three brands that spent the most on advertising had similar or fewer retweets per follower than the four that spent relatively little on advertising.

Some food and beverage brands have distinct ‘personalities’ on Twitter that generate millions of ‘likes’ and retweets. Some retweets have an inverse relationship with advertising spending, suggesting ‘personalities’ may be a uniquely powerful advertising tool for targeting young adults.

Few published community garden studies have focused on low socio-economic youth living in public housing or used a community-based participatory research approach in conjunction with youth-focused community garden programmes. The objective of the present study was to evaluate the feasibility (i.e. demand, acceptability, implementation and limited-effectiveness testing) of a 10-week experiential theory-based gardening and nutrition education programme targeting youth living in public housing.

In this mixed-methods feasibility study, demand and acceptability were measured using a combination of pre- and post-programme surveys and interviews. Implementation was measured via field notes and attendance. Limited-effectiveness was measured quantitatively using a pre–post design and repeated-measures ANOVA tests.

Two public housing sites in the Dan River Region of south central Virginia, USA.

Forty-three youth (primarily African American), twenty-five parents and two site leaders.

The positive demand and acceptability findings indicate the high potential of the programme to be used and be suitable for the youth, parents and site leaders. Field notes revealed numerous implementation facilitators and barriers. Youth weekly attendance averaged 4·6 of 10 sessions. Significant improvements (P<0·05) were found for some (e.g. fruit and vegetable asking self-efficacy, overall gardening knowledge, knowledge of MyPlate recommendations), but not all limited-effectiveness measures (e.g. willingness to try fruits and vegetables, fruit and vegetable eating self-efficacy).

This community-based participatory research study demonstrates numerous factors that supported and threatened the feasibility of a gardening and nutrition programme targeting youth in public housing. Lessons learned are being used to adapt and strengthen the programme for future efforts targeting fruit and vegetable behaviours.

Water plays a critical role in the production of food and preparation of nutritious meals, yet few studies have examined the relationship between water and food insecurity. The primary objective of this study, therefore, was to examine how experiences of household water insecurity (HWI) relate to experiences of household food insecurity (HFI) among a pastoralist population living in an arid, water-stressed region of northern Kenya.

We implemented the twelve-item Household Water Insecurity Experiences (HWISE, range 0–36) Scale and the nine-item Household Food Insecurity Access Scale (HFIAS, range 0–27) in a cross-sectional survey to measure HWI and HFI, respectively. Data on socio-demographic characteristics and intake of meat and dairy in the prior week were collected as covariates of interest.

Northern Kenya, June–July 2019.

Daasanach pastoralist households (n 136) from seven communities.

In the prior 4 weeks, 93·4 % and 98·5 % of households had experienced moderate-to-severe HWI and HFI, respectively. Multiple linear regression analyses indicated a strong association between HWI and HFI. Each point higher HWISE score was associated with a 0·44-point (95 % CI: 0·22, 0·66, P = 0·003) higher HFIAS score adjusting for socio-economic status and other covariates.

These findings demonstrate high prevalence and co-occurrence of HWI and HFI among Daasanach pastoralists in northern Kenya. This study highlights the need to address HWI and HFI simultaneously when developing policies and interventions to improve the nutritional well-being of populations whose subsistence is closely tied to water availability and access.

To test the hypothesis that diets with high glycaemic index (GI) and glycaemic load (GL) increase the risk of heart disease.

Overall GI and GL were assessed from 7 d diet records or diet history interviews.

Information on hospitalization and death due to CVD and CHD was obtained from the National Register of Cause of Death and the National Register of Patients.

In total 3959 adult Danes were – depending on time of entry – followed for 6–25 years until 31 December 1999.

Overall GI was inversely associated with heart disease in men. The hazard ratios (95 % CI) for the 10th and 90th GI percentiles compared with the median were 1·38 (1·13, 1·68) and 0·90 (0·76, 1·07) for CVD morbidity, 1·45 (1·05, 1·99) and 0·81 (0·62, 1·06) for CVD mortality, and 1·31 (0·97, 1·76) and 0·65 (0·51, 0·84) for CHD morbidity. In male subjects GL was not associated with either outcome. In women no clear association between overall GI and heart disease was found, whereas positive non-linear associations were found for GL: at very high levels of GL, increase in GL was associated with increasing CVD and CHD morbidity.

In men low-GI diets were associated increased risk of heart disease and GL was not associated with heart disease. In women there was no clear association between GI and heart disease, but to some extent a positive association between GL and heart disease was observed as hypothesized.

To understand perspectives of stakeholders during initial district-wide implementation of a Breakfast in the Classroom (BIC) model of the School Breakfast Program.

Qualitative data were collected from twenty-nine focus groups and twenty interviews with stakeholders in a school district early in the process of implementing a BIC model of the School Breakfast Program.

Ten elementary schools within a large, urban school district in the USA that served predominantly low-income, racial/ethnic minority students.

Purposively selected stakeholders in elementary schools that had implemented BIC for 3–6 months: students (n 85), parents/guardians (n 86), classroom teachers (n 44), cafeteria managers (n 10) and principals (n 10).

Four primary themes emerged, which were interpreted based on the Diffusion of Innovations model. School staff had changed their perceptions of both the relative disadvantages and costs related to time and effort of BIC over time; the majority of each stakeholder group expressed an appreciation for BIC; student breakfast consumption varied from day to day, related to compatibility of foods with child preferences; and stakeholders held mixed and various impressions of BIC’s potential impacts.

The study underscores the importance of engaging school staff and parents in discussions of BIC programming prior to its initiation to pre-emptively address concerns related to cost, relative disadvantages and compatibility with child preferences and school routines/workflow. Effectively communicating with stakeholders about positive impacts and nutritional value of the meals may improve support for BIC. These findings provide new information to policy makers, districts and practitioners that can be used to improve implementation efforts, model delivery and outcomes.

The present study assessed the nutrition information displayed on ready-to-eat packaged foods and the nutritional quality of those food products in Thailand.

In March 2015, the nutrition information panels and nutrition and health claims on ready-to-eat packaged foods were collected from the biggest store of each of the twelve major retailers, using protocols developed by the International Network for Food and Obesity/Non-communicable Diseases Research, Monitoring and Action Support (INFORMAS). The Thai Nutrient Profile Model was used to classify food products according to their nutritional quality as ‘healthier’ or ‘less healthy’.

In total, information from 7205 food products was collected across five broad food categories. Out of those products, 5707 (79·2 %), 2536 (35·2 %) and 1487 (20·6 %) carried a nutrition facts panel, a Guideline Daily Amount (GDA) label and health-related claims, respectively. Only 4691 (65·1 %) and 2484 (34·5 %) of the products that displayed the nutrition facts or a GDA label, respectively, followed the guidelines of the Thai Food and Drug Administration. In total, 4689 products (65·1 %) could be classified according to the Thai Nutrient Profile Model, of which 432 products (9·2 %) were classified as healthier. Moreover, among the 1487 products carrying health-related claims, 1219 (82·0 %) were classified as less healthy. Allowing less healthy food products to carry claims could mislead consumers and result in overconsumption of ready-to-eat food products.

The findings suggest effective policies should be implemented to increase the relative availability of healthier ready-to-eat packaged foods, as well as to improve the provision of nutrition information on labels in Thailand.

The current study aimed to (i) describe racial/ethnic disparities in household food and beverage purchases among participants and non-participants in the Supplemental Nutrition Assistance Program (SNAP) and (ii) examine longitudinal associations between SNAP participation and purchases by race/ethnicity.

To describe disparities, we estimated sociodemographic-adjusted mean purchases of seven unhealthy food and beverage groups (e.g. junk food, sugar-sweetened beverages) and four nutrients (e.g. sugar, Na) among white, black and Hispanic SNAP-participating and non-participating households. To examine longitudinal associations, we used multivariable linear regression with household fixed effects.

USA, 2010–2014.

Food and beverage purchases among low-income (≤250 % federal poverty line) US households (n 30 403) participating in the Nielsen Homescan Panel.

Among non-participants, there were significant black–white disparities (i.e. differences favouring white households) in households’ adjusted mean purchases of processed meat, sweeteners, sugar-sweetened beverages, energy and Na. These disparities persisted among SNAP participants. In contrast, the only significant Hispanic–white disparity among non-participants was for Na purchases; this disparity was reduced in magnitude and no longer significant among SNAP-participating households. Additionally, Hispanic households purchased less energy from junk foods than white households, regardless of SNAP status. In longitudinal models accounting for household fixed effects, SNAP participation was associated with increased energy purchased among black households. No other significant longitudinal associations between SNAP and purchase outcomes were observed.

SNAP may not be meeting its potential to improve food and beverage purchases or reduce disparities. Research is needed to identify strategies for ensuring nutritious purchases across all racial/ethnic groups.