Socio-economic resources, young child feeding practices, consumption of highly processed snacks and sugar-sweetened beverages: a population-based survey in rural northwestern Nicaragua

This cross-sectional study was conducted in the municipalities of Santo Tomás del Nance, San Juan de Cinco Pinos, San Pedro del Norte and San Francisco del Norte, Department of Chinandega in rural northwestern Nicaragua with a population of around 25,000 people and about 5,000 households. The area is collectively called Los Cuatro Santos (The Four Saints), where most of the population is engaged in small-scale subsistence farming activities whilst other income-generating opportunities are limited. Most of the mothers in the area were occupied with household activities and childcare and did not have a stable source of income. The Asociación para el Desarrollo Económico y Social de El Espino (APRODESE) is a local non-governmental economic and social development organization that has developed projects in Los Cuatro Santos to improve living standards. To inform the Los Cuatro Santos society about recent progress in the reduction of child mortality, all households in the area also received a pamphlet containing general health messages, as well as recommendations on infant feeding. In 2003, a Health and Demographic Surveillance System (HDSS) was established by APRODESE to monitor the effects of the development projects. Three rounds of HDSS data collection were performed in 2003, 2007 and 2009 respectively. Data for the current study were collected from May to November 2009 and included child feeding and household food insecurity data in addition to the routinely collected socio-demographic information.

The study sample consisted of children aged 0–35 months (0–3 years). It is common for infant and young child feeding studies to focus on children younger than 2 years, given that efforts to promote better nutrition may have more impact during the first 1,000 days of a child’s life [2]. However, we also included children aged 24–35 months because they may also be at risk of undernutrition [19] as was reconfirmed in a recent study [20]. Further, this age category may be more prone to be exposed to HP snacks and SSBs, underlining the relevance to evaluate non-favourable feeding behaviours in such a group.

The organization of data collection started with a group of locally recruited and trained interviewers who visited all households (about 5,000) in the area and collected socio-demographic and household food insecurity data. They identified and listed households with at least one child younger than 3 years. Thereafter, a second group of interviewers revisited the 1,500 households fulfilling this criterion and collected information on infant and young child feeding practices and the consumption of HP snacks and SSBs of the youngest child in the household (Figure 1).

The instruments to assess recommended infant and young child feeding practices were developed in accordance with the WHO core feeding indicators using a food frequency questionnaire (FFQ), recalling the number of times a food or drink item was consumed during the last 24 hours [3]. In all, 70 food items and 11 beverages were organized into seven groups, as suggested in the WHO guidelines [3,4], and were included in the FFQ list. The FFQ was pilot-tested in a nearby community [21] before being used in the survey. The food and beverage items, cookies and crackers, chocolates and candies, salty crisps, carbonated soft drinks, sweetened powdered fruit drinks and coffee, were listed in the FFQ to represent items being high-energy-dense and sugary or salty. To assess meal frequency, a specific question was asked regarding the number of eating episodes per day for the child in accordance with the, at that time, available WHO guidelines [3].

The WHO guidelines were used to construct feeding indicators on exclusive breastfeeding for children aged 0–5 months and on continued breastfeeding for children aged 12–15 months [3]. Complementary feeding indicators intended for children aged 6–23 months (but here also used for the 24- to 35-month-olds) were based on the following seven food groups: (a) grains, roots, tubers; (b) legumes and nuts; (c) vitamin-A fruits and vegetables; (d) other fruits and vegetables; (e) meats; (f) eggs; and (g) dairy products [3]. Fruits and vegetables were categorised and grouped according to their vitamin A content using food composition tables [22]. One point was given to a food group when at least one of its food items was consumed the previous day. The points for the food groups were summed to give a score that ranged from 0–7. A cut-off of four food groups or more was defined as achieving minimum dietary diversity [3]. Further, the minimum meal frequency indicator was created and included breastfed children who had eaten the minimum number of times or more (2 times if aged 6–8 months, and 3 times if aged 9–35 months) and non-breastfed children who ate or were given milk feeds the minimum number of times or more (4 times for children aged 6–35 months) the previous day [3]. The WHO guidelines were also used to construct an overall dietary adequacy indicator. Thus, for breastfed infants, the minimum acceptable diet included those who had at least the minimum dietary diversity and minimum meal frequency. For non-breastfed children to reach minimum dietary adequacy they should, in addition to these criteria, have consumed at least two milk feedings [3].

Further, one HP snack and one SSB variable were developed from counting whether any of the HP snacks and SSBs were consumed at least once, followed by the creation and categorisation into two dichotomous HP snack and SSB variables. Sugary coffee, however, was not included in the SSB variable, as it is a common drink item and deeply rooted in the culture and, as such, may have different determinants than the other items.

Children aged 6–35 months with at least one inadequate complementary feeding practice and concurrent consumption of any HP snack or SSB were defined as being exposed to a “double burden of suboptimal feeding”.

Housing quality was used as a proxy for the household’s overall economic resources. The basic needs approach (i.e. Index of Unsatisfied Basic Needs) has been commonly used in Latin American settings [23] and is partly based on the presence and type of drinking water source, house floor and walls, electricity supply and the presence of a latrine or toilet. Given that we sought to give the same value of importance to each resource level, the pre-coded options were all coded from 1 to 3, where a value of 1 meant the lowest housing quality situation and 3 the highest. For electricity supply, however, the variable was dichotomous (yes/no) and 1 point was given to households who had no electricity, whilst a value of 3 was given to households with electricity. The pre-coded options were summed up to create a housing quality score ranging from 5–15, which was further divided into tertiles; lowest (5–10), middle (11–12) and highest (13–15).

Maternal education was assessed by asking the highest level of schooling completed by the mothers at the time of the survey. The questions were based on a previous survey, with 28 potential pre-coded options included in the questionnaire. These were defined by the non-governmental organization (NGO) APRODESE and were part of the HDSS data collection instrument. The level of education was further divided into three categories. To reflect the Millenium Development Goal 2 of at least 5 years of primary education for all, the lowest level of education was defined as having less than 5 years of schooling. The middle level of education was having 5 to 9 years of schooling, whilst the highest was 10 years or more.

Household food insecurity was measured using the Household Food Insecurity Access Scale (HFIAS), which has been validated in various international settings [24]. The scale showed good internal consistency in our sample with Cronbach’s α = .863 [25]. This instrument consists of nine questions that measure uncertainty around obtaining food, limited access to high-quality foods and reduction of food quantity in the past 4 weeks [24]. The pre-coded options were never (0 points), rarely (once or twice in the past 4 weeks; 1 point), sometimes (3 to 10 times in the past 4 weeks; 2 points) and often (more than 10 times in the past 4 weeks; 3 points). The score of the answers to the questions was summed up (0–27) and thus a higher value signified a worse condition with more household food insecurity [24]. For descriptive purposes of occurrence of food insecurity, the categorisation of food insecurity followed the HFIAS guide and was defined as food secure, mildly food insecure, moderately food insecure and severely food insecure, based on a number of assumptions that were made by the HFIAS authors, given there is no universally accepted approach to defining the cut-offs [24]. For the inferential analyses, we created a continuous variable of food insecurity which was categorised into tertiles of the score defined as lowest (0–7), middle (8–11) and highest (12–27) food insecurity.

Maternal occupation was assessed by an open-ended question that asked the type of occupation for people older than 5 years in the household. The most common occupations of mothers were housewife, teacher, office worker and health worker. Other maternal occupations, but less common, were agriculturist, artisan and student. For the inferential analyses, we divided maternal occupation into two categories: the first being unemployed (e.g. housewife) mothers and the second labelled as employed (e.g. teacher, health worker) mothers.

Prevalence of the feeding practices and consumption of HP snacks and SSBs were described across the three levels of the three underlying resources in the focus of this study. Cross-tabulation with the Pearson’s chi-squared test was used to compare proportions and evaluate the association between maternal education and maternal occupation. Cross-tabulations were also used to compare proportions and determine whether resources were associated with the recommended feeding practices, consumption of HP snacks and/or SSBs and as well as a double burden of suboptimal feeding. Three statistical models were tested to evaluate associations after adjustment for potential confounders and to further understand the independent associations with the resources. The first model included the unadjusted results, the second included only the potential confounders whilst the third model included both the potential confounders and all three resources of interest in order to evaluate their independent associations with the feeding practices, consumption of HP snacks and SSBs and a double burden of suboptimal feeding. Factors associated with both exposure and outcome, which changed the effect estimate by 10% or more were included in the feeding practices, consumption of HP snacks and SSBs and double burden of suboptimal feeding models. These factors were maternal age, child age and municipality. The resources and confounders were introduced to the multivariate logistic regression analyses, which then provided unadjusted and adjusted odds ratios calculated with 95% confidence intervals for the feeding practices, consumption of HP snacks and SSBs and double burden of suboptimal feeding. The variables in the multivariate logistic regression were coded so that an odds ratio below one should be interpreted as a favourable outcome, whilst an odds ratio above one should be regarded as unfavourable. Significance level was defined at <0.05 and all statistical analyses were completed using the Statistical Package for the Social Sciences version 20 [26].

Our study followed the protocol of the Universal Helsinki Declaration. The Biomedical Research Ethics Committee at the University of Nicaragua in León approved the study protocol. Informed consent was obtained from all participating mothers and caretakers.