Dietary diversity and nutritional status of preschool children in north west province, south africa: A cross sectional study

Preschool children consume diets inadequate to meet their macro and micronutrient requirements, which ultimately affect their nutritional status due to lack of dietary diversity. A cross sectional study was conducted to investigate the association between dietary diversity scores (DDS) and the nutritional status of 379 preschool children in North West Province of South Africa. A 24 h qualitative recall by mothers of their children’s food consumption was used to calculate DDS based on 12-foods groups following Food and Agriculture Organization protocols. DDS was calculated by counting each of 12-food groups and classified as low (≤4), medium (5–8) and high (9–12). The weight and height of children were measured and height-for-age (HAZ), weight-for-age (WAZ) and BMI-for-age (BAZ) z-scores were calculated based on 2006 WHO standards. Stunting, underweight and thinness were defined as HAZ, WAZ and BAZ < −2SD, respectively. Linear and logistic regression analyses were used to assess the association between DDS and the nutritional indicators. Mean age for children was 4 ± 0.7 years, and the prevalence of stunting (29%), underweight (13%) and thinness (6%) was observed. Mean DDS was 4.39 ± 1.55 out of 12-food groups, with a prevalence of 61% and 39% for low and medium DDS, respectively. Cereals (100%) accounted for the main food group consumed, while fish and other seafood (17%) were the least consumed. Consumption of a diversified diet was associated with lower odds of being stunted [AOR = 0.25, 95%CI: 0.10 to 0.92] among the four-year olds, while in the unadjusted model, 5-year-olds had lower odds of being underweight [OR = −0.32, 95%CI: −0.57 to 0.07]. The findings of this study reinforce the importance of continued nutrition education of mothers, caregivers and preschool staff on the need to ensure consumption of diverse food sources in order to improve the nutritional status of children. Further studies are recommended on the association of DDS with the nutritional status, and factors associated with low dietary diversity among preschool children. This was a cross-sectional study conducted to investigate the association between dietary diversity and the nutritional status of preschool children in North West Province, South Africa. The study was conducted from June to December 2019. The study was conducted in Dr. Kenneth Kaunda District, situated in the North West Province of South Africa. The District is one of the four districts of the Province and is divided into three local municipalities; JB Marks (town; Potchefstroom), Maquassi Hills (town; Wolmaransstad) and Matlosana City (town; Klerksdorp). Dr. Kenneth Kaunda District has a population 742,821 people, and the inhabitants mainly speak Setswana. This District was selected as the area of interest for this study, while Klerksdorp and Wolmaransstad as peri-urban areas and were considered due to the scarcity of nutritional studies among children in these areas compared to Potchefstroom. Dr. Kenneth Kaunda District has 116 preschools. We used preschool settings, because most children who have not reached primary school age are under care in these facilities, which parents are reliant on. Furthermore, concerns regarding the food menus at these facilities to achieve adequate dietary specifications have been raised [21]. The study included children aged three to five years attending preschools in Dr. Kenneth Kaunda District, who had no physical disabilities that would have affected their stature, whose mothers gave consent and were available to participate in the study. The total enrolment number of children aged between three and five years in the preschools of Klerksdorp and Wolmaransstad municipalities was estimated at 3600 through personal communication with the heads of preschools, with enrolments ranging from a total number of 40–45 in the unfunded preschools to enrolments of 80–126 in the largest (funded) preschools. No electronic database on the enrolment numbers was available at the time of the study. Rao software was used to calculate a sample size, taking into consideration the population size of 3600 children, a 5% margin of error, and 95% confidence level. A minimum representative sample of 348 was calculated and buffered with 10% to make up for non-responses, and a sample size of 415 was obtained. Multi-stage sampling was used to select facilities and participants. The preschools selected for this study were targeted due to their high enrolment numbers, their location within the district and their funded by the Department of Social Development. Eight of the largest preschools, four from Klerksdorp and four from Wolmaransstad, were randomly selected in the district. Within the selected preschools, simple random sampling of children was carried out. The study adapted a questionnaire that was used in nutritional status studies [34,35], which considered the UNICEF conceptual framework for malnutrition [36]. The questionnaire was validated through content and face validity and a pilot study. Independent translators who speak Setswana as their mother tongue and are conversant with English did forward and backward translations of the questionnaire. The questionnaire comprised sociodemographic questions about the personal information of mothers, such as age, marital status, education level, as well as household information of the income, family size, head of household, availability of electricity, and sanitation. In addition, questions focused on obstetric history and children’s characteristics (childbirth date, sex, birth order, and birth weight) and selected feeding practices, mainly on breastfeeding and the introduction of solid foods. Trained research assistants administered questionnaires to mothers in Setswana, a local language in the study area. Dietary diversity score was calculated based on 24 h recall of mothers of the child’s consumption of 12 food groups within the past 24 h (Food and Agriculture Organization, 2007). The food groups included were based on FAO [37] recommendations, as follows: (i) cereals, (ii) vegetables, (iii) fruits, (iv) meat, (v) eggs, (vi) fish and other sea foods, (vii) legumes, nuts and seeds, (viii) milk and milk products, (ix) oil and fats, (x) sweets, (xi) spices, condiments and beverages, and (xii) tubers and roots. Commonly consumed foods in the area were incorporated into each food group. A dietary diversity score was created based on the mother’s recall of the child’s food intake in the previous 24 h. The response option of “yes” was scored one point if at least two food items in each food group were consumed by the child, whereas half a point was awarded for food items consumed once per group. For food groups not consumed at all, with a response option of “no”, zero (0) points were given. Dietary diversity scores (DDS) was summed up by counting each of the 12-food groups, and classified as low (≤4), medium (5–8) and high (9–12). The cut-offs used were due to a lack of national and international guidelines on which to base cut-offs [38]. Anthropometric measurements of height and weight for each child were determined following standard procedures. The weight of children was measured using an infant electronic digital weighing scale (smart D-quip electronic scale) to the nearest 0.1 kg, while standing height was measured to the nearest 0.1 m using a non-stretch tape measure. All measurements were taken three times, and the average recorded. Nutritional status indices were generated through anthropometry conversion to sex specific Z-scores using WHO Anthro Software. Stunting and underweight were defined as low height-for-age (HAZ) and weight-for-age (WAZ), respectively, below −2SD, while normal nutritional status was Z-scores between −2SD to +2SD. BMI-for-age (BAZ) scores between +1SD and +2SD indicated possible risk of overweight. Overweight was at BAZ >+2SD, obesity at BAZ ≥3SD and wasting/thinness was at <−2SD [39]. Statistical analyses were performed using STATA (Intercooled Stata® Version 14). Skewness–Kurtosis tests for normality were performed to check the distribution of data for children (i.e., age, weight, height, HAZ, WAZ and BAZ). Descriptive statistics for weight and height of the children were computed, compared using a parametric t-test, and results are presented as mean ± SD. Descriptive statistics for HAZ, WAZ and BAZ of the children were computed and compared by sex using the Mann–Whitney test (i.e., non-parametric test), and the results are presented as median (interquartile range (IQR)). A chi-square test was used to compare the prevalence of stunting, underweight and thinness, stratified by sex, and the results are presented as frequency (n) and percentage (%). The descriptive statistics of DDS (i.e., mean ± SD) were compared by sex and age, using independent sample t-test and ANOVA, respectively. The proportions of low and medium DDS were compared by sex and age using a chi-square test. To assess the relationship between dietary diversity and nutritional status indicators (HAZ, WAZ and BAZ, stunting, underweight and thinness), univariate linear regression (for continuous outcomes) and univariate logistic regression (for dichotomous outcomes) models were built. Multivariate models were created, adjusting for the characteristics of children (i.e., sex, age, birth order, term of pregnancy and breastfeeding), maternal (i.e., age, marital status, education level and employment status) and household (family size, monthly income, house type, access to water and toilet type). Models were run for the overall sample of children and in the age subgroups of children (i.e., 3, 4 and 5 years) to assess the influence of dietary diversity on children who may have a higher risk of poor nutritional status. Unadjusted odds ratio (OR) and adjusted odds ratio (AOR) are presented with a 95% confidence interval (CI) and significance was considered at p < 0.05. This study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving human subjects were approved by Sefako Makgatho Health Sciences University Research and Ethics Committee (SMUREC) (SMUREC/H/47/2019: PG). Furthermore, this study received permission from the North West Provincial Department of Social Development Research Committee, South Africa, and written consent was obtained from the mothers.