Menu
Menu
Menu
Menu
In this study, we investigated the relationship between agricultural biodiversity and dietary diversity of children and whether factors such as economic access may affect this relationship. This paper is based on data collected in a baseline cross-sectional survey in November 2013. The study population comprising 1200 mother-child pairs was selected using a two-stage cluster sampling. Dietary diversity was defined as the number of food groups consumed 24 h prior to the assessment. The number of crop and livestock species produced on a farm was used as the measure of production diversity. Hierarchical regression analysis was used to identify predictors and test for interactions. Whereas the average production diversity score was 4.7 ± 1.6, only 42.4% of households consumed at least four food groups out of seven over the preceding 24-h recall period. Agricultural biodiversity (i.e. variety of animals kept and food groups produced) associated positively with dietary diversity of children aged 6–36 months but the relationship was moderated by household socioeconomic status. The interaction term was also statistically significant [β = −0.08 (95% CI: −0.05, −0.01, p = 0.001)]. Spearman correlation (rho) analysis showed that agricultural biodiversity was positively associated with individual dietary diversity of the child more among children of low socioeconomic status in rural households compared to children of high socioeconomic status (r = 0.93, p < 0.001 versus r = 0.08, p = 0.007). Socioeconomic status of the household also partially mediated the link between agricultural biodiversity and dietary diversity of a child’s diet. The effect of increased agricultural biodiversity on dietary diversity was significantly higher in households of lower socioeconomic status. Therefore, improvement of agricultural biodiversity could be one of the best approaches for ensuring diverse diets especially for households of lower socioeconomic status in rural areas of Northern Ghana.
The study was undertaken in resource-poor households in rural areas of Northern Ghana where the primary occupation is farming. The study area is characterized by high poverty and recurrent droughts and floods which predispose communities to increased vulnerability to food insecurity and malnutrition. The Ghana Living Standards Survey Round 6 Report showed that the regions where the study was conducted have higher proportions of households in the lowest quintile than in the highest quintile [23]. The majority of the people have agriculture as their main occupation while some are involved in trading. The main staple foods including maize, sorghum, millet and yam are usually harvested from October through December. Although the food security situation is usually good during harvest time, child care tends to suffer because of lack of time on the part of rural mothers. A high proportion of rural mothers work daily away from home, and therefore frequently face challenges to the care of children. The rainfall pattern is unimodal and the period is usually short and lasts from May to August, followed by a long dry season (September – April) with dry harmattan winds. This paper is based on analysis of data which were collected in a baseline survey prior to an intervention study. The intervention package focused on nutrition behavior change communication (BCC) for improved child and maternal nutrition. In the intervention communities where the International Institute of Tropical Agriculture (IITA) developed and promoted better agronomic practices, they received a nutrition education package in addition to routine health services. The educational sessions were mainly messages that promote better health and nutrition, focusing on 1) appropriate complementary feeding such as use of thicker instead of thinner porridges; 2) use of animal-source foods; 3) dietary diversity; and 4) personal hygiene. In the comparison communities, eligible households were those that had no previous exposure to IITA program activities but they also received general health and nutrition messages at monthly growth monitoring sessions. The baseline survey report has been reported elsewhere [24] but, briefly, a community-based cross-sectional cluster survey was carried out in November 2013. The study population comprised mothers/primary caregivers and their children. A stratified, two-stage sample design in which the primary sampling units (communities) were selected with probability proportional to size within each of the five districts was used. Households were selected using random systematic sampling within each cluster. In each selected cluster, a complete list of all households was compiled, and systematic random sampling was used to select eligible households. The primary outcome variable used to estimate the sample size was the population proportion of chronic malnutrition (25.0 %) in the study area (Nutrition Surveillance Report, 2013, Unpublished). This outcome indicator was used to calculate a sample size of 1200 (600 per intervention and comparison areas). A sample size of 288 was required to ensure that the estimated prevalence of the main outcome variable was within plus or minus 5% of the true prevalence at 95% confidence level. Assuming a correction factor of 2 (the ‘design effect’) for cluster sampling, the sample size was increased to 576. A non-response rate of 5% and other unexpected events (e.g. damaged/incomplete questionnaire) was factored in the sample size determination and so the sample size is adjusted to 600 for 25 intervention communities. The same number of children was selected from comparison communities using probability proportionate to size (PPS). The Emergency Nutrition Assessment (ENA) software was used to randomly select the required number of clusters. The main outcome variable for this study was dietary diversity score of households and farm production diversity as an explanatory variable. The independent covariates were maternal, child and household characteristics. Child’s age was categorized into 6–8 months, 9–11 months, 12–23 months and 24–36 months. A brief description of main independent and dependent variables is as follows: As in previous studies, agricultural biodiversity was measured by the number of food groups grown and/or types of animals raised for food [25–27]. Households recalled all food groups and livestock grown/reared during the previous agricultural season were collected from both mother and father in each household through interviews. Agricultural biodiversity score at the household level was therefore calculated by summing the number of food groups and/or types of animals raised for food and sale. If a household produces several varieties of food crops that belong to the same food groups, the production diversity score will be smaller than the simple species count. Agricultural production diversity was also categorized (livestock only, crops only, crops and livestock, and nothing) and tested for association against minimum dietary diversity. Dietary diversity of the child was measured as per WHO guidelines [28,29]. The seven foods groups used for calculation of WHO minimum dietary diversity indicator are: 1) grains, roots and tubers; 2) legumes and nuts; 3) dairy products; 4) flesh foods; 5) eggs; 6) vitamin A rich fruits and vegetables; and 7) other fruits and vegetables. The dietary diversity score (DDS) was calculated by summing the number of food groups consumed by the child as reported over the 24-h recall period. From the dietary diversity score, the minimum dietary diversity indicator was constructed. Minimum dietary diversity is the proportion of children who ate at least four or more varieties of foods from the seven food groups in a 24-h time period [28,29]. A household wealth index based on household assets and housing quality was used as a proxy indicator for socioeconomic status (SES) of households. Principal Component Analysis (PCA) was used to determine household wealth index from information collected on housing quality (floor, walls, and roof material), source of drinking water, type of toilet facility, the presence of electricity, type of cooking fuel, and ownership of modern household durable goods (e.g. bicycle, television, radio, motorcycle, sewing machine, telephone, cars, refrigerator, mattress, bed, computer and mobile phone) [30–33]. The analysis of data took into account the complex design of multi-stage cluster surveys. All quantitative data were coded for statistical analysis using SPSS Complex Samples module for Windows 18.0 (SPSS Inc., Chicago). This was done in order to make statistically valid population inferences and computed standard errors from sample data. Design weights were added to each district’s sample data (i.e. total population divided by number of respondents) to perform weighted analysis. Bivariate associations were made between agricultural biodiversity and individual dietary diversity of children using Spearman rank correlation coefficients. We conducted three-step moderated hierarchical multiple regression analyses to determine independent predictors and moderators of dietary diversity of the child. Multicollinearity was investigated by using the variance inflation factor (VIF). A VIF (the reciprocal of the tolerance statistics) of greater than 5 is generally considered evidence of multicollinearity. Potential effect modification (statistical interaction) was investigated to ascertain whether the relationship between agricultural biodiversity and individual dietary diversity of children was moderated by socioeconomic status of household. Effect modification was identified and adjusted for through using three-step moderated hierarchical multiple regression analyses. The main covariate predictor variables (household wealth index, age group of child and household size) were entered in the first step. In the second step we added the main explanatory variable of interest (i.e. agricultural production diversity) and the interaction term (moderation) was added in the third step. The interaction term comprised the product of the centered agrobiodiversity score and centered household wealth index. Also, mediation analysis which provides a better understanding of the causal chain by which an independent variable (X) influences a dependent variable (Y) through a mediator (M) [34] was used to assess whether socioeconomic status of the household mediates the link between agricultural biodiversity and dietary diversity of a child’s diet. The study protocol was approved by the Scientific Review Committee of the School of Allied Health Sciences, University for Development Studies, Ghana. Ethics clearance was obtained from the Institutional Review Board (IRB) of the Tamale Teaching Hospital, Ghana (Ref no. TTH/10/11/15/01). Participation in the study was voluntary and no incentives were provided. Verbal informed consent was sought from all the study participants before the commencement of any interview. The study was not harmful to any study participant. Study participants were free to withdraw from the study at any time without any penalty.
N/A
