The present review aimed to identify and synthesize literature on household food insecurity with respect to whether the respondent was male or female.

A systematic review of prevalence studies followed by a meta-analysis was conducted between 28 August 2014 and 19 October 2014 in seven electronic databases. The search was updated in April 2016. The included studies used experience-based measures to assess household food insecurity. Dichotomous measures of food insecurity were used. Pooled odds ratios of household food insecurity prevalence in women v. men were obtained through random-effect modelling. Quality assessment, publication bias diagnostics and subgroup analysis were also performed.

Population-based studies (i.e. non-clinical populations).

Participants aged 18 years or over.

Out of the 5145 articles initially identified, forty-two studies with a total population of 233 153 were included. In general, results showed that the odds for household food insecurity was 40 % higher in studies where women were the respondent (95 % CI 1·27, 1·54; P<0·001). Besides, subgroup analysis revealed that female-headed households were 75% (95 % CI 49–96%) more likely to be food insecure than male-headed households.

Our results confirm the existence of gender differences in reporting household food insecurity. Furthermore, they indicate that households headed by women constitute a segment of the population that is particularly vulnerable to food insecurity.

An absence of food literacy measurement tools makes it challenging for nutrition practitioners to assess the impact of food literacy on healthy diets and to evaluate the outcomes of food literacy interventions. The objective of the present scoping review was to identify the attributes of food literacy.

A scoping review of peer-reviewed and grey literature was conducted and attributes of food literacy identified. Subjects included in the search were high-risk groups. Eligible articles were limited to research from Canada, USA, the UK, Australia and New Zealand.

The search identified nineteen peer-reviewed and thirty grey literature sources. Fifteen identified food literacy attributes were organized into five categories. Food and Nutrition Knowledge informs decisions about intake and distinguishing between ‘healthy’ and ‘unhealthy’ foods. Food Skills focuses on techniques of food purchasing, preparation, handling and storage. Self-Efficacy and Confidence represent one’s capacity to perform successfully in specific situations. Ecologic refers to beyond self and the interaction of macro- and microsystems with food decisions and behaviours. Food Decisions reflects the application of knowledge, information and skills to make food choices. These interdependent attributes are depicted in a proposed conceptual model.

The lack of evaluated tools inhibits the ability to assess and monitor food literacy; tailor, target and evaluate programmes; identify gaps in programming; engage in advocacy; and allocate resources. The present scoping review provides the foundation for the development of a food literacy measurement tool to address these gaps.

To examine the hypothesis that the association between vitamin D deficiency and depressive symptoms is dependent upon total cholesterol level in a representative national sample of the South Korean population.

This was a population-based cross-sectional study.

The Fifth Korean National Health and Nutrition Examination Survey (KNHANES V, 2010–2012).

We included 7198 adults aged 20–88 years.

The incidence of depressive symptoms in individuals with vitamin D deficiency (serum 25-hydroxyvitamin D<20 ng/ml) was 1·54-fold (95 % CI 1·20, 1·98) greater than in individuals without vitamin D deficiency (serum 25-hydroxyvitamin D ≥20 ng/ml). The relationship was stronger in individuals with normal-to-borderline serum total cholesterol (serum total cholesterol<240 mg/dl; OR=1·60; 95 % CI 1·23, 2·08) and non-significant in individuals with high serum total cholesterol (OR=0·97; 95 % CI 0·52, 1·81) after adjustment for confounding variables (age, sex, BMI, alcohol consumption, smoking status, regular exercise, income level, education level, marital status, changes in body weight, perceived body shape, season of examination date and cholesterol profiles).

The association between vitamin D deficiency and depressive symptoms was weakened by high serum total cholesterol status. These findings suggest that both vitamin D and total cholesterol are important targets for the prevention and treatment of depression.

To clarify the pathways between household livestock and child growth by assessing the relationships between consumption of animal-source foods (ASF) and child growth and evaluating the household livestock correlates of child consumption of ASF.

We conducted a longitudinal cohort study of anthropometry and 3 d feeding recalls among children <5 years old between June 2014 and May 2015. In addition, we collected data on wealth, livestock ownership and livestock diseases in the same households. We used linear and negative binomial mixed models to evaluate the relationships between household livestock characteristics, reported consumption of ASF and child growth.

An 1800-household surveillance catchment area in Western Kenya within the structure of human and animal health surveillance systems.

Children (n 874) <5 years old.

Among children >6 months old, reported frequency of egg and milk consumption was associated with increased monthly height gain (for each additional report of consumption over 3 d: adjusted β (95 % CI)=0·010 (0·002, 0·019) cm/month and 0·008 (0·004, 0·013) cm/month, respectively). Poultry ownership was associated with higher reported frequency of egg, milk and chicken consumption (adjusted incidence rate ratio (95 % CI)=1·3 (1·2, 1·4), 1·4 (1·1, 1·6) and 1·3 (1·1, 1·4), respectively). Some livestock diseases were associated with lower reported frequency of ASF intake (livestock digestive diseases-adjusted incidence rate ratio (95 % CI)=0·89 (0·78, 1·00)).

Child height gain was associated with milk and egg consumption in this cohort. ASF consumption was related to both household livestock ownership and animal health.

To evaluate the quality of food choices according to adolescent individual earnings in Brazil.

Adolescents were classified according to their individual earnings as having or not having spending power for their own expenses. Food records from two non-consecutive days of the Brazilian National Dietary Survey (NDS 2008–2009) were used to estimate food intake. Quality of food choices was based on two approaches: (i) the NOVA classification, which classifies processed and ultra-processed foods and drinks as unhealthy food groups; and (ii) traditional classification, with beans, milk, fruits and vegetables as healthy food groups, and soft drinks, sweets, snacks and crackers classified as unhealthy. We compared mean food intake (g/kJ or ml/kJ) according to per capita household income (tertiles) and adolescent individual earnings, with adolescent earnings adjusted for household income, using multiple linear regression.

Brazilian households (n 13 569).

Adolescents aged 14–18 years (n 3673).

Males without individual earnings had higher per capita household income than those with individual earnings. Household income was associated with all three food groups of the NOVA classification and seven of the eight groups of the traditional classification. However, only beans and snacks were consumed in significantly greater quantities by adolescents with individual earnings compared with those without earnings.

Adolescent individual earnings were not the main driver of food choices; however, per capita household income was associated with food choices. The consumption of both healthy and unhealthy items increased with increasing household income.

To assess the impact of the European School Fruit Scheme (SFS) in North Rhine-Westphalia, Germany, on children’s fruit and vegetable (F&V) consumption, in particular frequency.

The study consisted of a pre-test/post-test design with an intervention (eight primary schools) and a control group (two primary schools). Children’s F&V consumption frequency was measured prior to the introduction of the SFS in 2010 and after one year’s delivery of F&V (2011).

Ten primary schools in North Rhine-Westphalia, Germany.

In total, 499 primary-school children aged 6–11 years, 390 in the intervention and 109 in the control schools.

Children highly appreciated the SFS. More than 90 % evaluated the programme positively. Children in the intervention group showed a significant increase in F&V intake frequency from baseline to follow-up, from on average 1·26 (sd 1·37) to 2·02 (sd 1·33) times/d (P=0·000). The intervention variable had a highly significant impact on children’s F&V consumption frequency, even after controlling for gender, age and stay at school for lunch (β=0·773; 95 % CI 0·59, 0·96). The SFS did not induce a reduction of F&V consumption at home. In the control group a non-significant decline in F&V consumption frequency from 1·31 (sd 1·26) to 1·18 (sd 1·34) times/d (P=0·325) was observed.

One year after the programme’s implementation, the SFS led to a significant short-term increase in children’s F&V consumption.

To evaluate whether the lifestyle intervention MetSLIM targeting individuals of low socio-economic status of Turkish, Moroccan and Dutch origin was successful in improving waist circumference and other cardiometabolic risk factors, lifestyle behaviour and quality of life.

A quasi-experimental intervention study (Netherlands Trial Register NTR3721). The intervention group participated in a 12-month combined dietary and physical activity programme. Examinations were performed at baseline and after 12 months. Participants underwent anthropometric measurements and blood withdrawal, and completed questionnaires on dietary intake, physical activity and quality of life.

Socio-economically deprived neighbourhoods in two Dutch cities, involving non-blinded ethnicity-matched and gender-matched research assistants, dietitians and sports instructors.

Mainly Turkish (49 %) and Dutch (36 %) subjects, aged 30–70 years, with a waist-to-height ratio of >0·5 (intervention, n 117; control, n 103). Dropout was 31 %.

At 12 months, the intervention group showed greater improvements than the control group in waist circumference (β=−3·3 cm, 95 % CI −4·7, −1·8, P<0·001) and other obesity measures. Additionally, greater reductions were observed for total cholesterol (β=−0·33 mmol/l, 95 % CI −0·56, −0·10, P=0·005) and LDL cholesterol (β=−0·35 mmol/l, 95 % CI −0·56, −0·14, P=0·001). Dietary changes were significant for fibre intake (β=1·5 g/4184 kJ (1000 kcal), 95 % CI 0·3, 2·7, P=0·016). Compared with the control group, the intervention group reported a decrease in total minutes of physical activity (β=−573 min/week, 95 % CI −1126, −21, P=0·042) and showed improvements in the quality-of-life domains ‘health transition’ and ‘general health’.

MetSLIM was shown to be effective in improving waist circumference, total and LDL cholesterol, and quality of life among Dutch and Turkish individuals living in deprived neighbourhoods.

The Farmers’ Market Fresh Fund Incentive Program is a policy, systems and environmental intervention to improve access to fresh produce for participants on governmental assistance in the USA. The current study examined factors associated with ongoing participation in this matched monetary incentive programme.

Relationship of baseline factors with number of Fresh Fund visits was assessed using Poisson regression. Mixed-effects modelling was used to explore changes in consumption of fruits and vegetables and diet quality.

San Diego, California.

Recipients of Supplemental Nutrition Assistance Program (SNAP), Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and Supplemental Security Income (SSI) who attended participating farmers’ markets from 2010 to 2012 (n 7298).

Among those with participation for ≤6 months, factors associated with increased visits included reporting more daily servings of fruits and vegetables (F&V) at baseline, being Vietnamese or Asian/Pacific Islander, and eligibility because of SNAP/CalFresh or SSI (v. WIC). Among those who came for 6–12 months, being Asian/Pacific Islander, eligibility because of SNAP/CalFresh and enrolling in the autumn, winter or spring were associated with a greater number of Fresh Fund visits. Among those who came for >12 months, being male and eligibility because of SSI were associated with a greater number of visits. Overall, the odds of increasing number of servings of F&V consumed increased by 2 % per month, and the odds of improved perception of diet quality increased by 10 % per month.

Sustaining and increasing Fresh Fund-type programme operations should be a top priority for future policy decisions concerning farmers’ market use in low-income neighbourhoods.

Vitamin A supplementation (VAS) for children aged 6–59 months occurs regularly in most sub-Saharan African countries. The present study aimed to explore child, household and delivery platform factors associated with VAS coverage and identify barriers to compliance in thirteen African countries.

We pooled data (n ~60 000) from forty-four household coverage surveys and used bivariate and multivariable regression analyses to assess the effects of supplementation strategy, rural v. urban residence, child sex, child age, caregiver education and campaign awareness on child VAS status.

Primary caregivers of children aged 6–59 months in thirteen countries.

Door-to-door distribution resulted in higher VAS coverage than fixed-site plus outreach approaches (91 v. 63 %) and was a significant predictor of supplementation in the adjusted model (OR=19·0; 95 % CI 17·2, 21·1; P<0·001). Having been informed about the campaign was the main predictor of VAS in the door-to-door (OR=6·8; 95 % CI 5·8, 7·9; P<0·001) and fixed-site plus outreach (OR=72·5; 95 % CI 66·6, 78·8; P<0·001) groups.

Door-to-door provision of VAS may achieve higher coverage than fixed-site models in the African context. However, the phase-out of door-to-door polio immunization campaigns in most sub-Saharan African countries threatens the main distribution vehicle for VAS. Our findings suggest well-informed communities are key to attaining higher coverage using fixed-site delivery alternatives.

The present study examined food and beverage distributors’ sourcing, placement and promotion of obesogenic (energy-dense, nutrient-poor) product categories from the perspective of small food store owners/managers. The obesogenic product categories of interest were savoury snacks, sugary beverages, sweet snacks, confectionery and frozen treats. Specifically, we examined how frequently distributors sourced these products, and the types of agreements and expectations they had for their placement and promotion. Differences were explored by store size and ethnicity. Fresh produce was used as a comparison when examining differences in frequency of sourcing only, with implications for healthy food access.

Survey research involving in-person interviews.

Four urban areas in the USA: Baltimore, MD; Durham, NC; Minneapolis/St. Paul, MN; and San Diego, CA.

Seventy-two small food store owners/managers, 65 % consent rate.

Most distributors sourced obesogenic products weekly. Agreements to place products were predominantly informal (e.g. handshake) with sweet snack, confectionery and frozen treat distributors, and formal (e.g. contract) with savoury snack and sugary beverage distributors. Free-standing displays were the most common incentive provided by distributors and they expected some control over their placement and pricing. Free/discounted products and signage were also common incentives but slotting fees were not. Smaller stores and ethnic stores were less likely to receive various incentives, but among sweet snack distributors, they were more likely to control the price in ethnic v. non-ethnic stores.

Obesogenic products are ubiquitous. Influencing what is made available to consumers in the retail food environment needs to consider the distributor.

The current qualitative study aimed to identify gender, social and cultural influences on the management and use of unconditional cash transfers as part of a prospective intervention study in Niger.

In February to March 2012, focus group discussions and semi-structured individual interviews were conducted with female caregivers of children aged 6 to 23 months who received unconditional cash transfers. Discussion and interview transcripts were analysed using content thematic analysis.

The study was conducted in the Madarounfa district in Maradi region of Niger.

Among forty-eight intervention villages, fourteen were selected for the qualitative study. Participants were randomly selected from eligible households.

In total, 124 women participated in focus group discussions or interviews. The majority reported giving the cash transfer to the male head of household who primarily managed cash at the household level. Women reported using a portion of the money to purchase foods for the target child. Feeding the household was the primary use of the cash transfer, followed by health care, clothing, gifts or ceremonies.

Gender, social and cultural norms influenced management and usage of the cash transfer at the household level. The results highlight the importance of integrating gender-sensitive indicators into interventions. Information and awareness sessions should be an integral component of large-scale distributions with a special emphasis on gender equality and the importance of women’s empowerment to improve agriculture and family health.

To determine the association of vitamin D status with depression and health-related quality of life among women.

This was a cross-sectional study conducted among women in Kuala Lumpur, Malaysia. Sociodemographic characteristics, physical activity status, perceived depression and health-related quality of life were assessed via a self-administered questionnaire. Fasting blood samples were taken for the analysis of 25-hydroxyvitamin D, parathyroid hormone, fasting blood glucose and full lipid profile. Complex samples multiple logistic regression analysis was performed.

Public secondary schools in Kuala Lumpur, Malaysia.

Seven hundred and seventy female teachers were included.

The mean age of participants was 41·15 (95 % CI 40·51, 41·78) years and the majority were ethnic Malays. Over 70 % of them had vitamin D deficiency (<20 ng/ml or <50 nmol/l) and two-thirds were at risk for depression. In the multivariate analysis, ethnic Malays (adjusted OR (aOR)=14·72; 95 % CI 2·12, 102·21) and Indians (aOR=14·02; 95 % CI 2·27, 86·59), those at risk for depression (aOR=1·88, 95 % CI 1·27, 2·79) and those with higher parathyroid hormone level (aOR=1·13; 95 % CI 1·01, 1·26) were associated with vitamin D deficiency, while vitamin D deficiency was negatively associated with mental health-related quality of life (Mental Component Summary) scores (aOR=0·98; 95 % CI 0·97, 0·99).

Vitamin D deficiency is significantly associated with depression and mental health-related quality of life among women in Kuala Lumpur, Malaysia.

To evaluate the effectiveness of a community-based participatory nutrition promotion (CPNP) programme involving a 2-week group nutrition session in improving child feeding and hygiene practices among caregivers.

Cluster randomized trial. In the intervention area (six clusters), the CPNP programme was added to the context of government nutrition programmes; the control area (six clusters) received the government programme only. Child feeding practices were assessed every 3 months using a 24 h dietary recall questionnaire, and hand washing with soap was assessed every 6 months, over a period of 12 months. Feeding and hygiene measures at each visit were scored and the scores summed up for the entire follow-up period.

Habro and Melka Bello districts, Ethiopia.

Randomly selected mothers with a child aged 6–12 months (n 1790).

A total of 1199 mothers, 629 in the control and 570 in the intervention areas, were assessed at all visits and included in the analysis. Mothers in the intervention area showed higher scores than those in the control area regarding meal frequency (difference: 1·04, 95 % CI 0·35, 1·73), composite feeding score_1 (difference: 1·25, 95 % CI 0·37, 2·13; a summing score of currently breast-feeding, meal frequency and dietary diversity) and composite feeding score_2 (difference: 1·40, 95 % CI 0·49, 2·32; a summing score of meal frequency and dietary diversity). However, there were no differences in the scores of breast-feeding, dietary diversity and hand washing between the two areas (all P>0·05).

The CPNP programme was effective in improving some child feeding behaviours in rural Eastern Ethiopia.

To apply the Theory of Planned Behaviour to examine the relationship between the constructs of background factors and beliefs towards using policy, systems and environmental (PSE) strategies and reported use of PSE strategies to prevent obesity by a group of professional nutrition educators.

Cross-sectional study using self-reported survey.

Cooperative Extension in New York, USA.

Nutrition educators (n 58); survey response rate 100 %.

Nutrition educators’ reported use of PSE strategies to prevent obesity were positively associated with background factors of their community networking and number of staff they managed, their belief of other people’s expectations of them to make PSE changes and the belief that their communities were ready to use PSE strategies; and negatively associated with their belief that individual-level factors contributed to obesity. The relationships among these variables were complicated and their use of PSE strategies occurred only when they utilized their professional networks at a moderately high level (above mean of 5·3 on a scale of 1–7), given that their community was also ready to use PSE strategies.

Nutrition educators’ use of PSE strategies depends on several internal and external factors. Community networking needs to be emphasized as one of the most significant factors contributing to nutrition educators’ work in this area. Organizational and community support should be in place in order to facilitate nutrition educators’ effective use of PSE strategies.

To assess relationships between mothers’ feeding practices (food as a reward, food for emotion regulation, modelling of healthy eating) and mothers’ willingness to purchase child-marketed foods and fruits/vegetables (F&V) requested by their children during grocery co-shopping.

Cross-sectional. Mothers completed an online survey that included questions about feeding practices and willingness (i.e. intentions) to purchase child-requested foods during grocery co-shopping. Feeding practices scores were dichotomized at the median. Foods were grouped as nutrient-poor or nutrient-dense (F&V) based on national nutrition guidelines. Regression models compared mothers with above-the-median v. at-or-below-the-median feeding practices scores on their willingness to purchase child-requested food groupings, adjusting for demographic covariates.

Participants completed an online survey generated at a public university in the USA.

Mothers (n 318) of 2- to 7-year-old children.

Mothers who scored above-the-median on using food as a reward were more willing to purchase nutrient-poor foods (β=0·60, P<0·0001), mothers who scored above-the-median on use of food for emotion regulation were more willing to purchase nutrient-poor foods (β=0·29, P<0·0031) and mothers who scored above-the-median on modelling of healthy eating were more willing to purchase nutrient-dense foods (β=0·22, P<0·001) than were mothers with at-or-below-the-median scores, adjusting for demographic covariates.

Mothers who reported using food to control children’s behaviour were more willing to purchase child-requested, nutrient-poor foods. Parental feeding practices may facilitate or limit children’s foods requested in grocery stores. Parent–child food consumer behaviours should be investigated as a route that may contribute to children’s eating patterns.

The present study aimed to describe change in feeding intentions and predictors of breast-feeding intentions during the course of pregnancy.

Analysis of prospectively collected data from a larger randomized controlled trial of a health education intervention to reduce environmental smoke exposure among women during and after pregnancy.

Participants were recruited from prenatal clinics, but all further communication occurred with participating women living in the community.

Low-income, adult women (n 399) were interviewed during the 16th and 32nd week of pregnancy to ascertain prenatal feeding intentions and breast-feeding knowledge, attitudes and self-efficacy. Characteristics of women by infant feeding intention were assessed along with differences in intention from 16 to 32 weeks of pregnancy and feeding behaviours after delivery. Differences in psychosocial variables between women of different intention for infant feeding were measured. Women in each category of feeding intention were assessed for changes in psychosocial factors by eventual infant feeding behaviour.

Feeding intention early in pregnancy was strongly, but not consistently, associated with feeding intention late in pregnancy, feeding initiation and later feeding patterns. Over one-third of women who were undecided at 16 weeks’ gestation or earlier initiated breast-feeding. Increases in knowledge and improvement in time, social factors and social support barriers were found among those who exclusively breast-fed.

Results indicate that feeding decisions may change during pregnancy. Determining when women make feeding decisions during their pregnancy warrants more research. Interventions to increase breast-feeding intentions should target knowledge, self-efficacy and barriers.

Shifting towards a more sustainable food consumption pattern is an important strategy to mitigate climate change. In the past decade, various studies have optimised environmentally sustainable diets using different methodological approaches. The aim of the present review was to categorise and summarise the different approaches to operationalise the health aspects of environmentally sustainable diets.

Conventional keyword and reference searches were conducted in PubMed, Scopus, Web of Knowledge and CAB Abstracts. Inclusion criteria were: (i) English-language publication; (ii) published between 2005 and October 2015; (iii) dietary data collected for the diet as a whole at the national, household or individual level; (iv) comparison of the current diet with dietary scenarios; and (v) for results to consider the health aspect in some way.

Consumer diets.

Adult population.

We reviewed forty-nine studies that combined the health and environmental aspects of consumer diets. Hereby, five approaches to operationalise the health aspect of the diet were identified: (i) food item replacements; (ii) dietary guidelines; (iii) dietary quality scores; (iv) diet modelling techniques; and (v) diet-related health impact analysis.

Although the sustainability concept is increasingly popular and widely advocated by nutritional and environmental scientists, the journey towards designing sustainable diets for consumers has only just begun. In the context of operationalising the health aspects, diet modelling might be considered the preferred approach since it captures the complexity of the diet as a whole. For the future, we propose SHARP diets: environmentally Sustainable (S), Healthy (H), Affordable (A), Reliable (R) and Preferred from the consumer’s perspective (P).

To identify the features of a nutrition education programme for disadvantaged adults deemed most attractive and useful by participants.

A two-year, multi-method, qualitative evaluation of pre and post data collected from programme participants. Data were imported into NVivo10 for coding to facilitate a thematic analysis.

Western Australia.

Individuals attending the Western Australian FOODcents nutrition education programme that is designed to provide knowledge and skills needed to consume a healthy diet on a budget. Focus groups were conducted several weeks after course completion (five groups, forty-seven participants), observations were conducted during FOODcents sessions (thirty-one observation episodes, 237 participants), and open-ended questions were asked in pre–post hard-copy surveys administered in sessions (n 927) and an online survey administered on average six weeks after course completion (n 114).

The course attributes that were found to be especially important to participants were: (i) user-friendly, practical information that could be immediately translated to their daily lives; (ii) experiential learning that involved direct contact with food products; and (iii) opportunities for social interaction. These aspects of nutrition education were described as being highly influential in the decision to participate in the course, the application of the information in their subsequent food purchase and preparation activities, and their word-of-mouth communications with others about the course.

Incorporating aspects of most importance to participants into nutrition education programme delivery and promotion may increase joining rates, enjoyment, satisfaction with course content and, ultimately, the uptake of recommended behaviours.

The study aimed to evaluate the impact of a 15-month intervention on dietary intake conducted among obesity-prone normal-weight pre-school children.

Information on dietary intake was obtained using a 4 d diet record. A diet quality index was adapted to assess how well children’s diet complied with the Danish national guidelines. Linear regression per protocol and intention-to-treat analyses of differences in intakes of energy, macronutrients, fruit, vegetables, fish, sugar-sweetened beverages and diet quality index between the two groups were conducted.

The Healthy Start study was conducted during 2009–2011, focusing on changing diet, physical activity, sleep and stress management to prevent excessive weight gain among Danish children.

From a population of 635 Danish pre-school children, who had a high birth weight (≥4000 g), high maternal pre-pregnancy BMI (≥28·0 kg/m2) or low maternal educational level (<10 years of schooling), 285 children completed the intervention and had complete information on dietary intake.

Children in the intervention group had a lower energy intake after the 15-month intervention (group means: 5·29 v. 5·59 MJ, P=0·02) compared with the control group. We observed lower intakes of carbohydrates and added sugar in the intervention group compared with the control group after the intervention (P=0·002, P=0·01).

The intervention resulted in a lower energy intake, particularly from carbohydrates and added sugar after 15 months of intervention, suggesting that dietary intake can be changed in a healthier direction in children predisposed to obesity.

Individuals experiencing homelessness are particularly vulnerable to food insecurity. The At Home/Chez Soi study provides a unique opportunity to first examine baseline levels of food security among homeless individuals with mental illness and second to evaluate the effect of a Housing First (HF) intervention on food security in this population.

At Home/Chez Soi was a 2-year randomized controlled trial comparing the effectiveness of HF compared with usual care among homeless adults with mental illness, stratified by level of need for mental health services (high or moderate). Logistic regressions tested baseline associations between food security (US Food Security Survey Module), study site, sociodemographic variables, duration of homelessness, alcohol/substance use, physical health and service utilization. Negative binomial regression determined the impact of the HF intervention on achieving levels of high or marginal food security over an 18-month follow-up period (6 to 24 months).

Community settings at five Canadian sites (Moncton, Montreal, Toronto, Winnipeg and Vancouver).

Homeless adults with mental illness (n 2148).

Approximately 41 % of our sample reported high or marginal food security at baseline, but this figure varied with gender, age, mental health issues and substance use problems. High need participants who received HF were more likely to achieve marginal or high food security than those receiving usual care, but only at the Toronto and Moncton sites.

Our large multi-site study demonstrated low levels of food security among homeless experiencing mental illness. HF showed promise for improving food security among participants with high levels of need for mental health services, with notable site differences.