Children from rural households are often deprived of adequate micronutrient intakes either from food or supplementation. The present study examines: (i) the determinants of households’ micronutrient-rich food consumption; and (ii) the combined effect of vitamin A supplementation and micronutrient-rich food consumption on child stunting in households with different food allocation patterns.

Cross-sectional study. Households’ micronutrient-rich food consumption frequency and vitamin A supplementation were used as a proxy measure for child micronutrient intakes. Intra-household food allocation patterns were assessed from caregivers’ perception of the disparity in food distribution within the household. Descriptive statistics and logistic regression were employed in analysing the study’s objectives.

Rural communities in Kwara State, Nigeria.

Included 419 children aged 6–59 months and 413 households.

Owning small livestock and a refrigerator, knowledge of micronutrient-rich foods and higher parental education had strong associations with households’ micronutrient-rich food consumption. Children from households that consumed micronutrient-rich foods and received more diverse diets were less likely to experience stunting. The combined effect of micronutrient-rich food consumption and vitamin A supplementation was stronger on the likelihood of stunting reduction than the separate effect of each.

Assets ownership, human capital and knowledge of micronutrient-rich foods improve consumption of micronutrient-rich foods among the study population. Micronutrient supplementation to children with poor access to micronutrient-rich foods may not substantially enhance child growth unless reinforced through consumption of micronutrient-rich foods. Fruit and vegetable gardening, livestock holdings and nutrition education to parents should be integral parts of community nutrition programming.

To assess interdependent effects of autonomous motivation to limit sugar-sweetened beverage (SSB) consumption in relation to SSB and healthy beverage (HB) intake in mother–adolescent and father–adolescent dyads.

Adopting a dyadic cross-sectional design, the actor–partner interdependence modelling (APIM) approach was used to construct and analyse two APIM for mother–adolescent and father–adolescent dyads. The first model assessed actor effects (individual’s autonomous motivation associated with his/her own beverage intake) and partner effects (individual’s autonomous motivation associated with another family member’s beverage consumption) of autonomous motivation on SSB consumption. The second model assessed actor and partner effects of autonomous motivation on HB intake.

Two Internet-based surveys were completed in participant households.

Data from a demographically representative US sample of parent–adolescent dyads (1225 mother–adolescent dyads, 424 father–adolescent dyads) were used.

In the first model (autonomous motivation on SSB consumption), actor effects were significant for adolescents, but not for parents. Partner effects were significant for mother–adolescent, but not father–adolescent dyads. In the second model (autonomous motivation on HB intake), actor effects were significant for adolescents and parents in all dyadic combinations. Regarding partner effects, adolescent autonomous motivation had a significant effect on HB intake for mothers and fathers. In addition, maternal autonomous motivation had a significant effect on adolescent HB intake. No partner effects for HB were identified for fathers.

We found significant interdependent effects of autonomous motivation in relation to SSB and HB intake in mother–adolescent and father–adolescent dyads for eleven out of sixteen pathways modelled.

To: (i) understand the nutrition attitudes, self-efficacy, knowledge and practices of school food-service personnel (SFP) in Nebraska and (ii) identify potential barriers that schools face in offering healthy school meals that meet the US Department of Agriculture (USDA) nutrition standards.

Convergent parallel mixed-methods study.

Kindergarten–12th grade schools in Nebraska, USA.

SFP (260 survey participants; fifteen focus group participants) working at schools that participate in the USDA National School Lunch Program.

Mixed-methods themes identified include: (i) ‘Mixed attitudes towards healthy meals’, which captured a variety of conflicting positive and negative attitudes depending on the situation; (ii) ‘Positive practices to promote healthy meals’, which captured offering, serving and promotion practices; (iii) ‘Mixed nutrition-related knowledge’, which captured the variations in knowledge depending on the nutrition concept; and (iv) ‘Complex barriers’, which captured challenges with time, support and communication.

The study produced relevant findings to address the barriers identified by SFP. Implementing multicomponent interventions and providing training to SFP may help reduce some of the identified barriers of SFP.

To examine the relationship between protein intake and the risk of incident premenstrual syndrome (PMS).

Nested case–control study. FFQ were completed every 4 years during follow-up. Our main analysis assessed protein intake 2–4 years before PMS diagnosis (for cases) or reference year (for controls). Baseline (1991) protein intake was also assessed.

Nurses’ Health Study II (NHS2), a large prospective cohort study of registered female nurses in the USA.

Participants were premenopausal women between the ages of 27 and 44 years (mean: 34 years), without diagnosis of PMS at baseline, without a history of cancer, endometriosis, infertility, irregular menstrual cycles or hysterectomy. Incident cases of PMS (n 1234) were identified by self-reported diagnosis during 14 years of follow-up and validated by questionnaire. Controls (n 2426) were women who did not report a diagnosis of PMS during follow-up and confirmed experiencing minimal premenstrual symptoms.

In logistic regression models adjusting for smoking, BMI, B-vitamins and other factors, total protein intake was not associated with PMS development. For example, the OR for women with the highest intake of total protein 2–4 years before their reference year (median: 103·6 g/d) v. those with the lowest (median: 66·6 g/d) was 0·94 (95 % CI 0·70, 1·27). Additionally, intakes of specific protein sources and amino acids were not associated with PMS. Furthermore, results substituting carbohydrates and fats for protein were also null.

Overall, protein consumption was not associated with risk of developing PMS.

In recent years, researchers have been working towards creating a standard conceptual framework of food parenting. To understand how parents’ reports correspond with the proposed model, the current study examined parents’ reports of their feeding behaviours in the context of a newly established framework of food parenting.

Cross-sectional, with a two-week follow-up for a subset of the sample. Participants completed a quantitative and qualitative survey to assess food parenting. The survey included items from common food parenting instruments to measure the constructs posited in the framework. Exploratory factor analyses were conducted to ascertain which items related most closely to one another and factors were mapped on to existing constructs.

Online.

Parents of children aged 2·5–7 years (n 496). Of these, 122 completed a two-week follow-up.

Analyses revealed eleven aspects of Structure (monitoring; distraction; family presence; meal/snack schedule; unstructured practices; healthy/unhealthy food availability; food preparation; healthy/unhealthy modelling; rules), ten aspects of Coercive Control (pressure to eat; using food to control emotions; food incentives to eat; food incentives to behave; non-food incentives to eat; restriction for health/weight; covert restriction; clean plate; harsh coercion) and seven aspects of Autonomy Promotion (praise; encouragement; nutrition education; child involvement; negotiation; responsive feeding; repeated offering). Content validity, assessed via parents’ open-ended explanations of their responses, was high, and test–retest reliability was moderate to high. Structure and Autonomy Promoting food parenting were highly positively correlated.

In general, parents’ responses provided support for the model, but suggested some amendments and refinements.

The present paper aimed to demonstrate how 24 h dietary recall data can be used to generate a nutrition-relevant food list for household consumption and expenditure surveys (HCES) using contribution analysis and stepwise regression.

The analysis used data from the 2011/12 Bangladesh Integrated Household Survey (BIHS), which is nationally representative of rural Bangladesh. A total of 325 primary sampling units (PSU=village) were surveyed through a two-stage stratified sampling approach. The household food consumption module used for the analysis consisted of a 24 h open dietary recall in which the female member in charge of preparing and serving food was asked about foods and quantities consumed by the whole household.

Rural Bangladesh.

A total of 6500 households.

The original 24 h open dietary recall data in the BIHS were comprised of 288 individual foods that were grouped into ninety-four similar food groups. Contribution analysis and stepwise regression were based on nutrients of public health interest in Bangladesh (energy, protein, fat, Fe, Zn, vitamin A). These steps revealed that a list of fifty-nine food items captures approximately 90 % of the total intake and up to 90 % of the between-person variation for the key nutrients based on the diets of the population.

The study illustrates how 24 h open dietary recall data can be used to generate a country-specific nutrition-relevant food list that could be integrated into an HCES consumption module to enable more accurate and comprehensive household-level food and nutrient analyses.

Bangladesh, like many emerging economies of South-East Asia, has started to experience a double burden of continuing high rates of undernutrition and increasing rates of overweight and obesity. A lack of assessment of the nutritional shift leaves a gap in current policies: the growing overweight and obesity is yet to be addressed. The present paper investigates the change in nutritional status, particularly the shift in BMI, of Bangladeshi women of reproductive age (15–49 years) and characterizes the vulnerable households for both underweight and overweight status during a period of 10 years (2004–2014).

Generalized linear mixed-effect models were fitted for both urban and rural residents to assess underweight and overweight status.

Bangladesh Demographic and Health Surveys.

Women aged 15–49 years (n 53 077).

The proportion of overweight increased during 2004–2014 from 10·7 to 25·1 % and the proportion of underweight decreased from 32·6 to 18·2 %. Prevalence of underweight status remained high in rural areas and prevalence of overweight increased rapidly in both rural and urban areas, creating a double burden. The significant contributors to this double burden were the change in women’s level of education, increased household wealth, divisional location and rapid urbanization.

The findings indicate that specific cohort- or area-based intervention policy studies in line with the UN Decade of Action on Nutrition are required to address the nutritional double burden in Bangladesh.

To examine mothers’ and young children’s consumption of indigenous and traditional foods (ITF), assess mothers’ perception of factors that influence ITF consumption, and examine the relationship between perceived factors and ITF consumption.

Longitudinal study design across two agricultural seasons. Seven-day FFQ utilized to assess dietary intake. Mothers interviewed to assess their beliefs about amounts of ITF that they or their young children consumed and on factors that influence ITF consumption levels.

Seme sub-County, Kenya.

Mothers with young children.

Less than 60 % of mothers and children consumed ITF at time of assessment. Over 50 % of the mothers reported that their ITF consumption amounts and those of their children were below levels that mothers would have liked for themselves or for their young children. High cost, non-availability and poor taste were top three reasons for low ITF consumption levels. Mothers who identified high cost or non-availability as a reason for low levels of ITF consumption had significantly lower odds of consuming all ITF except amaranth leaves. Mothers who identified poor taste had significantly lower odds of consuming all ITF except green grams and groundnuts. Similar relationships were noted for young children’s ITF consumption levels.

A majority of the mothers reported that they and their children did not consume as much ITF as the mothers would have liked. Further studies should examine strategies to improve availability and affordability of ITF, as well as develop recipes that are acceptable to mothers and children.

To examine the feasibility of using social media to assess the consumer nutrition environment by comparing sentiment expressed in Yelp reviews with information obtained from a direct observation audit instrument for grocery stores.

Trained raters used the Nutrition Environment Measures Survey in Stores (NEMS-S) in 100 grocery stores from July 2015 to March 2016. Yelp reviews were available for sixty-nine of these stores and were retrieved in February 2017 using the Yelp Application Program Interface. A sentiment analysis was conducted to quantify the perceptions of the consumer nutrition environment in the review text. Pearson correlation coefficients (ρ) were used to compare NEMS-S scores with Yelp review text on food availability, quality, price and shopping experience.

Detroit, Michigan, USA.

None.

Yelp reviews contained more comments about food availability and the overall shopping experience than food price and food quality. Negative sentiment about food prices in Yelp review text and the number of dollar signs on Yelp were positively correlated with observed food prices in stores (ρ=0·413 and 0·462, respectively). Stores with greater food availability were rated as more expensive on Yelp. Other aspects of the food store environment (e.g. overall quality and shopping experience) were captured only in Yelp.

While Yelp cannot replace in-person audits for collecting detailed information on the availability, quality and cost of specific food items, Yelp holds promise as a cost-effective means to gather information on the overall cost, quality and experience of food stores, which may be relevant for nutrition outcomes.

To determine undernutrition prevalence in 0–59-month-old children and its determinants during the period 2000–2015 in sub-Saharan Africa.

Ecological study of time series prevalence of undernutrition in sub-Saharan Africa assessed from 2000 to 2015.

Underweight and stunting prevalence from the World Bank database (2000–2015) were analysed. Mixed models were used to estimate prevalence of underweight and stunting. Country-specific undernutrition prevalence variation was estimated and region comparisons were performed. A meta-regression model considering health and socio-economic characteristics at country level was used to explore and estimate the contribution of different undernutrition determinants.

Countries of sub-Saharan Africa.

During 2000–2015, underweight prevalence in sub-Saharan Africa was heterogeneous, ranging between 7 and 40 %. On the other hand, stunting prevalence ranged between 20 and 60 %. In general, higher rates of underweight and stunting were estimated in Niger (40 %) and Burundi (58 %), respectively; while lowest rates of underweight and stunting were estimated in Swaziland (7 %) and Gabon (21 %). About 1 % undernutrition prevalence reduction per year was estimated across sub-Saharan Africa, which was not statistically significant for all countries. Health and socio-economic determinants were identified as main determinants of underweight and stunting prevalence variability in sub-Saharan Africa.

Undernutrition represents a major public health threat in sub-Saharan Africa and its prevalence reduction during the period 2000–2015 was inconsistent. Improving water accessibility and number of medical doctors along with reducing HIV prevalence and poverty could significantly reduce undernutrition prevalence in sub-Saharan Africa

Low- and middle-income countries (LMIC) are increasingly experiencing the double burden of malnutrition. Studies to identify ‘double-duty’ actions that address both undernutrition and overweight in sub-Saharan Africa are needed. We aimed to identify acceptable behaviours to achieve more optimal feeding and physical activity practices among both under- and overweight children in Rwanda, a sub-Saharan LMIC with one of the largest recent increases in child overweight.

We used the Trials of Improved Practices (TIPs) method. During three household visits over 1·5 weeks, we used structured interviews and unstructured observations to collect data on infant and young child feeding practices and caregivers’ experiences with testing recommended practices.

An urban district and a rural district in Rwanda.

Caregivers with an under- or overweight child from 6 to 59 months of age (n 136).

We identified twenty-five specific recommended practices that caregivers of both under- and overweight children agreed to try. The most frequently recommended practices were related to dietary diversity, food quantity, and hygiene and food handling. The most commonly cited reason for trying a new practice was its benefits to the child’s health and growth. Financial constraints and limited food availability were common barriers. Nearly all caregivers said they were willing to continue the practices and recommend them to others.

These practices show potential for addressing the double burden as part of a broader intervention. Still, further research is needed to determine whether caregivers can maintain the behaviours and their direct impact on both under- and overweight.

The present study evaluated the association of two measures of diet quality with BMI and waist circumference (WC), overall and by education level, among Mexican men and women.

We constructed two a priori indices of diet quality, the Mexican Diet Quality Index (MxDQI) and the Mexican Alternate Healthy Eating Index (MxAHEI), which we examined relative to BMI and WC. We computed sex-specific multivariable linear regression models for the total sample and by education level.

Mexico.

Mexican men (n 954) and women (n 1356) participating in the Mexican National Health and Nutrition Survey 2012.

Total dietary scores were not associated with BMI in men and women, but total MxDQI was inversely associated with WC in men (−0·10, 95 % CI −0·20, −0·004 cm). We also found that some results differed by education level in men. For men with the lowest education level, a one-unit increase in total MxDQI and MxAHEI score was associated with a mean reduction in BMI of 0·11 (95 % CI −0·18, 0·04) and 0·18 (95 % CI −0·25, −0·10) kg/m2, respectively. Likewise, a one-unit increase in total MxDQI and MxAHEI score was associated with a mean change in WC of −0·30 (95 % CI −0·49, −0·11) and −0·53 (95 % CI −0·75, −0·30) cm, respectively, in men with the lowest level of education. In women, the association of diet quality scores with BMI and WC was not different by education level.

Our findings suggest that a higher diet quality in men with low but not high education is associated with lower BMI and WC.

We explored how positive and negative life experiences of caregivers are associated with household food insecurity.

The Midlands Family Study (MFS) was a cross-sectional study with three levels of household food security: food secure, food insecure without child hunger and food insecure with child hunger. Ordinal logistic regression analysis was used for analyses of negative and positive life experiences (number, impact, type) associated with food insecurity.

An eight-county region in South Carolina, USA, in 2012–2013.

Caregivers (n 511) in households with children.

Caregivers who reported greater numbers of negative life experiences and greater perceived impact had increased odds of household food insecurity and reporting their children experienced hunger. Each additional negative life experience count of the caregiver was associated with a 16 % greater odds of food insecurity without child hunger and a 28 % greater odds of child hunger. Each one-unit increase in the negative impact score (e.g. a worsening) was associated with 8 % higher odds of food insecurity without child hunger and 12 % higher odds of child hunger. Negative work experiences or financial instability had the strongest association (OR = 1·8; 95 % CI 1·5, 2·2) with child hunger. Positive life experiences were generally not associated with food security status, with one exception: for each unit increase in the number of positive experiences involving family and other relationships, the odds of child hunger decreased by 22 %.

More research is needed to understand approaches to build resilience against negative life experiences and strengthen positive familial, community and social relationships.

To evaluate the dietary diversity and the nutrient contribution of traditional foods (locally cultivated and wild) by conducting a food intake study in rural Ecuador.

Repeated 24 h recalls over a 14 d interval and frequency of consumption served to simulate the usual diet by the Multiple Source Method. Data on missing visits (n 11) were imputed using multivariate imputation by chained equations. The intakes of three macro- and six micronutrients were reported. Nutrient Adequacy Ratios, Mean Adequacy Ratio (MAR), Dietary Species Richness (DSR) and Minimum Dietary Diversity for Women were used as measures of dietary quality. A linear quantile mixed model was used to investigate the association between DSR, local species, MAR, age, education and occupation.

Guasaganda, Cotopaxi (Ecuador).

Rural, indigenous adult women, non-pregnant and not breast-feeding.

The studied diet had MAR of 0·78. Consumption of traditional foods contributed 38·6 % of total energy intake. Daily requirements for protein, carbohydrates, Fe and vitamin C were reached. An extra level of consumption of local species was associated with an increase in median MAR for macronutrients of 0·033 (P < 0·001). On the other hand, an extra level of consumption of local species was associated with an increase in median MAR for micronutrients of 0·052 (P < 0·001).

We found statistical evidence that traditional foods contribute to adequate intakes of macro- and micronutrients and dietary diversification in the studied population. Future public health interventions should promote the cultivation and consumption of traditional foods to increase the quality of the local diet.

The present study aimed to give an overall view of the pattern of high-dose vitamin A supplementation (VAS) coverage in twenty-three sub-Saharan African countries and factors associated with receipt of VAS among children aged 6–59 months.

Cross-sectional data from the twenty-three Demographic and Health Surveys conducted from 2011 to 2015 in twenty-three sub-Saharan African countries were pooled. A multilevel logistic regression model was used to explore factors associated with VAS.

Twenty-three sub-Saharan African countries.

Children (n 215 511) aged 6–59 months.

The overall coverage of VAS among children aged 6–59 months for the surveys included was 59·4 %. In the multivariable analysis, children whose mothers had primary (adjusted OR (aOR)=1·43; 95 % CI 1·39, 1·47) or secondary or above (aOR=1·72; 95 % CI 1·67, 1·77) educational status were more likely to receive VAS than children whose mothers had no formal education. Other factors associated with significantly increased likelihood of VAS were: living in urban areas; children of working mothers; children whose mothers had higher media exposure; children of older mothers v. children of mothers aged 15–19 years; and older children v. children aged 6–11 months. At the country level, lower media exposure was significant and negatively associated with VAS.

Broader VAS coverage is needed according to our data. More efforts are needed to scale up coverage, focusing mostly on groups at risk of non-receipt of vitamin A.

To evaluate differences in children’s eating behaviour in relation to their weight status.

Prospective, cross-sectional study. Anthropometric measures were taken and age- and sex-adjusted BMI percentiles and Z-scores were calculated according to the Centers for Disease Control and Prevention recommendations to assess weight status. Parents completed a questionnaire which included demographic data and the Children’s Eating Behaviour Questionnaire (CEBQ) to assess eating behaviour.

Tuzla Canton, Bosnia and Herzegovina (September 2016–September 2017).

Male and female children aged 3–10 years and one of their parents.

The study sample comprised 2500 children; 6·8 % of them were underweight and 14·4 % were overweight, while there were 14·8 % obese children and 64·0 % had normal weight. The factor analysis of CEBQ revealed an eight-factor solution. Significant differences in CEBQ subscale scores were found within BMI categories for all CEBQ subscales except Food Fussiness. On the other hand, child BMI Z-scores showed a linear increase with the ‘food approach’ subscales of the CEBQ, except the Desire to Drink subscale which was excluded from analysis, and a decrease with ‘food avoidant’ subscales.

The present study suggests that the CEBQ is valuable for identifying specific eating styles that are associated with weight status and can be seen as important and modifiable determinants implicated in the development and maintenance of overweight/obesity as well as underweight.

To assess trends of mortality attributable to child and maternal undernutrition (CMU), overweight/obesity and dietary risks of non-communicable diseases (NCD) in sub-Saharan Africa (SSA) using data from the Global Burden of Disease (GBD) Study 2015.

For each risk factor, a systematic review of data was used to compute the exposure level and the effect size. A Bayesian hierarchical meta-regression analysis was used to estimate the exposure level of the risk factors by age, sex, geography and year. The burden of all-cause mortality attributable to CMU, fourteen dietary risk factors (eight diets, five nutrients and fibre intake) and overweight/obesity was estimated.

Sub-Saharan Africa.

All age groups and both sexes.

In 2015, CMU, overweight/obesity and dietary risks of NCD accounted for 826204 (95 % uncertainty interval (UI) 737346, 923789), 266768 (95 % UI 189051, 353096) and 558578 (95 % UI 453433, 680197) deaths, respectively, representing 10·3 % (95 % UI 9·1, 11·6 %), 3·3 % (95 % UI 2·4, 4·4 %) and 7·0 % (95 % UI 5·8, 8·3 %) of all-cause mortality. While the age-standardized proportion of all-cause mortality accounted for by CMU decreased by 55·2 % between 1990 and 2015 in SSA, it increased by 63·3 and 17·2 % for overweight/obesity and dietary risks of NCD, respectively.

The increasing burden of diet- and obesity-related diseases and the reduction of mortality attributable to CMU indicate that SSA is undergoing a rapid nutritional transition. To tackle the impact in SSA, interventions and international development agendas should also target dietary risks associated with NCD and overweight/obesity.

Optimal nutrition for hospital patients is crucial and routine monitoring of patients’ nutrient intake is imperative. However, personalised monitoring and customised intervention using traditional methods is challenging and labour-intensive, consequently it is often neglected in hospital settings. The present pilot study aimed to examine the reliability and validity of the Dietary Intake Monitoring System (DIMS) against the weighed food method (WFM).

The DIMS 2.0 is composed of an integrated digital camera, weighing scale, radio-frequency identification sensor and WIFI connection for real-time image and weight dietary data acquisition and analysis. The DIMS equipment was used to collect data for a paired set of meals both before and after meal consumption at lunchtime.

Odense University Hospital, Denmark.

Photos and weights of seventeen patient meals were captured.

The results showed a significant correlation between DIMS and WFM for energy (r=0·99, P<0·01) and protein intake (r=0·98, P<0·01). Intraclass correlation coefficients (ICC) revealed a high degree of agreement among the four non-trained assessors for estimates of portion size of each food item before (0·88, P<0·01) and after consumption (0·99, P<0·01). The ICC for energy and protein intake were 0·99 (P<0·01) and 0·99 (P<0·01), respectively. Bland–Altman plots revealed no systematic bias.

Considering the huge benefits associated with routine monitoring, technological advances have made it possible to develop a novel, easy-to-use DIMS that, according to the findings, is a valid alternative for use in hospital settings.

The present study aimed to explore how healthy food choices are translated into everyday life by exploring definitions of healthy food choices, perceptions of own food choice, and healthy food choice drivers (facilitators) and barriers.

An exploratory qualitative study design was employed using semi-structured face-to-face interviews. Convenience sampling was used to recruit participants. Interviews were audio-recorded, transcribed verbatim and cross-checked for consistency. Thematic analysis was used to identify patterns emerging from the data.

Canberra, Australia, October 2015–March 2016.

A total of twenty-three participants aged 25–60 years were interviewed. The mean age was 38 years and the mean BMI was 29·1 kg/m2. All male participants (n 4) were within the healthy weight range compared with 58 % of female participants, with 26 % being overweight and 16 % being obese.

(i) Healthy food choices are important but are not a daily priority; (ii) healthy eating information is known but can be difficult to apply into everyday life; (iii) popular diets are used in attempts to improve healthy eating; and (iv) social media inspires and connects people with healthy eating.

Social media facilitates healthy food choices by providing access to healthy eating information. In addition to Facebook and Instagram, healthy eating blogs were highlighted as a source of nutrition information. Research should consider exploring the use of healthy eating blogs and whether these blogs can be used as a tool by dietitians to communicate procedural healthy eating information more effectively in the future.

We aimed at analysing changes in consumption of selected food groups in the Portuguese population before and after the Great Recession, which hit the country between 2008 and 2013.

We used pooled cross-sectional data from the Portuguese National Health Interview Surveys of 2005/2006 and 2014. We modelled the probability of consumption of soup, fish, meat, potatoes/rice/pasta, bread, legumes, fruit, vegetables and sweets/desserts, as a function of the year, controlling for age, sex and education, using logistic regressions. Then, we stratified the analysis by age group and education level. Analyses were adjusted for survey weights.

Portugal (2005/2006 to 2014).

Adults (n 43273) aged 25–79 years.

From 2005/2006 to 2014, there was a significantly lower consumption of fish, soup, fruit and vegetables. Conversely, the consumption of legumes and sweets/desserts was significantly higher in 2014. The changes in the selected food groups were consistent across most education levels. Among people aged 65 years or above, there were no significant changes in most foods, except an increase in the consumption of legumes and sweets/desserts. In contrast, people aged 25–39 and 40–64 years significantly decreased their intakes of fish and soup and increased their consumption of sweets/desserts.

The consistent results across education levels suggest that changes in dietary habits are not linked to the economic downturn. By contrast, our findings suggest a shift away from foods commonly linked to the Mediterranean diet, particularly among younger people.