Dietary Patterns, Food Quality, and Nutritional Status of School-Aged Children (8–12 Years) in Thulamela Municipality, Limpopo Province, South Africa: Implications for Public Health

2.1. Study design and setting

A cross-sectional study design was employed to assess eating behaviours, dietary quality, and nutritional status among school-aged children aged 8–12 years in Thulamela Local Municipality, situated in the Vhembe District of Limpopo Province, South Africa. The study design was appropriate for describing nutritional outcomes and examining associations between dietary practices and anthropometric indicators at a single point in time. Thulamela Municipality is one of the four local municipalities in the Vhembe District and comprises both rural and peri-urban settlements, with a total of approximately 227 villages. The municipality is predominantly rural, and the majority of residents rely on subsistence farming, informal employment, social grants, and small-scale trading as primary sources of livelihood. High levels of unemployment and household poverty persist in the area, which may negatively influence food security, diet quality, and access to healthcare services.

The municipality experiences inequities in access to basic services, including potable water, sanitation, and healthcare facilities, particularly in remote rural villages. These structural challenges contribute to vulnerabilities in child nutrition and health. Limited access to diverse and nutrient-rich foods, combined with increased availability of low-cost, energy-dense processed foods, places children at risk of both undernutrition and emerging overnutrition, reflecting the dual burden of malnutrition observed in many rural South African settings. Given these socio-economic and environmental conditions, Thulamela Municipality provides a relevant setting for investigating dietary practices and nutritional status among school-aged children.

2.2. Study Population, sample size, and sampling

The study included school-aged children (8–12 years) from Thulamela Municipality. The sample size was determined using Slovin’s formula based on an estimated population of 139,529, with a 5% margin of error and 95% confidence level, yielding a minimum of 399 participants. An additional 10% was added for non-response, increasing the target to 439; however, 347 children were ultimately included due to attrition during data collection.

A multistage sampling approach was used. Eight villages were randomly selected from 227 in the municipality. Households were then assigned identification numbers and randomly selected until the required sample was reached. Children meeting the age criteria were identified within selected households, and systematic sampling was applied to ensure even distribution. Only children with caregiver consent were included in the study.

2.3. Variables measured

Variables measured were socio-demographic characteristics, eating habits, food quality and anthropometric measurements (weight and height).

2.4. Socio-Demographic characteristics

The socio-demographic characteristics of the parents were estimated via questionnaires given as part of oral interview. The socio-demographic collected data on variables such as gender, education, marital status, income, and occupation.

2.5. Anthropometric status

Anthropometric assessments were carried out in accordance with internationally recognised standard protocols recommended by the International Society for the Advancement of Kinanthropometry (ISAK) to ensure accuracy and consistency (Esparza-Ros et al., 2019). All measurements were conducted by a qualified Biokineticist with appropriate training in anthropometric assessment.

Body weight and height were measured in duplicate using calibrated instruments, and the average of the two readings was recorded for analysis. During measurement, children wore light clothing and no footwear. Height was measured to the nearest 0.1 cm using a portable stadiometer, while body weight was measured to the nearest 0.01 kg using a portable SECA solar digital scale (Model 0213; SECA, Hamburg, Germany). Standardized measurement procedures were followed throughout data collection to minimise measurement error and enhance reliability.

2.6. Definition of thinness, stunting, overweight, and obesity

The nutritional status of the children was assessed using age- and sex-specific anthropometric indicators based on World Health Organization (WHO) growth reference standards (WHO, 2011). Measured weight and height values were converted into Z-scores using the WHO Anthro and WHO AnthroPlus software, which are designed for the assessment of nutritional status in children and adolescents.

Anthropometric classifications were derived from the calculated Z-scores. Stunting was identified using the height-for-age indicator, with values below −2 standard deviations (SD) indicating stunting and values below −3 SD indicating severe stunting. Thinness was determined using body mass index-for-age (BMI-for-age), where Z-scores less than −2 SD denoted thinness and scores less than −3 SD denoted severe thinness. BMI-for-age Z-scores between −2 SD and + 1 SD were classified as normal nutritional status.

At the higher end of the BMI-for-age distribution, Z-scores greater than +1 SD and up to +2 SD were used to classify children as overweight, while Z-scores exceeding +2 SD were categorised as obesity, in accordance with WHO reference guidelines.

2.7. Dietary habits and food quality

Dietary habits of the children were assessed using a structured Food Frequency Questionnaire (FFQ) adapted from the 2005 South African National Food Consumption Survey. The FFQ was designed to capture habitual dietary intake by assessing the frequency of consumption of commonly eaten foods over a specified period. Information collected included meal patterns, such as the frequency of breakfast, lunch, and dinner, the number of meals consumed per day, and snacking behaviours. The questionnaire also assessed consumption frequency across major food groups, allowing for the evaluation of overall dietary quality.

Food quality was assessed based on the diversity and frequency of consumption of nutrient-dense versus energy-dense foods reported in the FFQ. Foods were categorised into key groups, including fruits, vegetables, whole grains, dairy products, sugary foods, fried foods, and processed foods. Higher food quality was indicated by frequent consumption of nutrient-rich food groups such as fruits, vegetables, whole grains, and dairy products, while lower food quality was characterised by frequent intake of highly processed, sugary, and fried foods. This approach enabled the identification of dietary patterns reflective of healthy or poor-quality diets.

The FFQ was selected due to its established validity and reliability in assessing dietary intake among children in South African and similar low- and middle-income settings. Tools derived from the National Food Consumption Survey have been previously validated and widely used to assess habitual food intake and dietary patterns in paediatric populations. The interviewer-administered approach further enhanced reliability by minimising recall errors and addressing literacy limitations, while standardised administration procedures ensured consistency across interviews. Adaptation of the FFQ to the local context helped to maintain content validity by ensuring that commonly consumed foods within the study area were adequately represented.

2.8. Ethical consideration

Ethical approval for the study was obtained from the University of Venda Research Ethics Committee (FHS/25/NUT/03/2603). Permission to conduct the study was also granted by the traditional leaders of the eight selected villages, facilitating community entry and household access. All study procedures complied with the Declaration of Helsinki, Good Clinical Practice guidelines, and applicable South African ethical regulations. Prior to data collection, parents or caregivers and participating children received detailed information about the study. Written informed consent was obtained from parents or guardians, and assent was obtained from the children. For illiterate caregivers, the consent information was explained verbally in Tshivenda or Xitsonga, and consent was documented using a thumbprint.

2.9. Statistical analysis

Quantitative data were analysed using the Statistical Package for the Social Sciences (SPSS) for Windows, version 29 (SPSS Inc., Chicago, IL, USA). Children’s nutritional status was classified using standard anthropometric reference criteria and appropriate cut-off values. Descriptive statistics were initially employed to summarise participants’ socio-demographic characteristics and dietary practices, with categorical variables presented as frequencies and percentages.

Associations between dietary practices and anthropometric status were examined using chi-square (χ²) tests for categorical variables and Kruskal–Wallis tests for variables that were not normally distributed. To assess relationships between dietary habits (including consumption of whole grains, vegetables, fruits, dairy products, sugar-sweetened beverages, and fast foods) and socio-demographic characteristics of parents or caregivers, binary logistic regression analyses were conducted. Dietary variables were dichotomised to reflect consumption versus non-consumption or adequate versus inadequate intake, based on response distributions.

Logistic regression models were used to estimate odds ratios (ORs) with corresponding 95% confidence intervals (CIs), quantifying the strength and direction of associations between socio-demographic factors and dietary habits. Independent variables included parental education level, household income, employment status, marital status, child’s sex, and household food expenditure. Reference categories were specified for all categorical variables. Statistical significance was set at p < 0.05.

3.1. Socio-demographic characteristics of parents

The study sample included slightly more females (53.0%) than males (47.0%). Most parents (91.0%) were highly literate, while a small proportion (4.0%) reported having no formal education. Regarding caregiver marital status, 45.8% were single, 35.4% were married, 10.1% were cohabiting, and 8.6% were widowed. Approximately 76.3% of households reported a low monthly income (<R10,000), with the remaining 23.7% classified as high-income. Additionally, around half of the caregivers were unemployed ( Table 1).

3.2. Eating habits and meal frequency among school-Aged children

Table 2 presents the eating habits and meal frequency patterns of the participating children. Daily breakfast consumption was reported by 63.4% of participants, while 4.9% never ate breakfast. Breakfast frequency was significantly associated with height-for-age (p = 0.001). Similarly, the number of meals consumed per day showed a significant relationship with height-for-age (p = 0.001), with most children (60.8%) eating three or more meals daily. Regular snack consumption was reported by 71.2% of participants and was also significantly correlated with height-for-age (p = 0.021). Lunch and dinner were consumed consistently, with over 87% of children reporting daily intake of these meals.

3.3. Food quality and consumption patterns

Table 3 summarizes the frequency of foods consumed by children and their associations with nutritional status. Overall, consumption of healthy food groups was high, with most children reporting intake of vegetables (85.6%), fruits (78.9%), dairy products (65.2%), and whole grains (59.8%). These food groups were significantly associated with both height-for-age and BMI-for-age: vegetables (p = 0.012), fruits (p = 0.010), dairy products (p < 0.001), and whole grains (p < 0.001). Sugary foods and fried foods were also commonly consumed (73.1% and 64.7%, respectively) and were significantly related to height-for-age (p = 0.033). Additionally, water intake of five or more glasses per day, reported by 71.5% of children, showed a significant association with height-for-age (p = 0.033).

3.4. Nutritional status of school-aged children

Approximately 21.0% of the children were moderately stunted, and 20.5% were severely stunted. The prevalence of overweight and obesity was 42.1% and 32.6%, respectively. Moderate thinness was observed in 0.8% of the children, while 0.3% were classified as severely thin ( Table 4).

3.5. Association between dietary habits and socio-demographic characteristics

Children’s dietary habits were significantly associated with several socio-demographic factors. Consumption of whole grains and vegetables was higher among children whose parents had only primary education compared to secondary education (OR = 3.38, p = 0.038; OR = 3.10, p = 0.039, respectively), and vegetable intake was also greater in households earning more than R20,000 per month (OR = 5.93, p = 0.003). Male children were more likely than females to consume fruits (OR = 1.83, p = 0.004) and dairy (OR = 1.80, p = 0.004), whereas higher household spending on food (>R2,000) was associated with lower dairy (OR = 0.48, p = 0.029) and fast-food consumption (OR = 0.31, p = 0.003). Children from single-parent households were less likely to consume dairy (OR = 0.40, p = 0.014), while sugar-sweetened drink intake was higher among children of unemployed parents (OR = 3.13, p = 0.001), those from higher-income households (OR = 6.61, p = 0.002), and those whose parents had tertiary education (OR = 3.30, p < 0.001). These findings indicate that parental education, household income, employment status, child’s sex, and household food expenditure significantly influence children’s dietary patterns ( Table 5).

Policy, practice, and global health implications

The findings indicate a need to strengthen school nutrition programmes by ensuring provision of balanced, nutrient-dense meals, particularly breakfast, while reducing availability of energy-dense, nutrient-poor foods to help curb increasing overweight and obesity. Nutrition education for parents and caregivers at community level is important to support healthy meal planning, dietary diversity, and consistent eating habits at home. Social protection measures and food assistance for vulnerable households can improve access to healthy foods and reduce dietary inequalities. Regular monitoring and evaluation of feeding programmes are necessary to improve effectiveness and inform policy adjustments.

Although the study was conducted in Thulamela Municipality, the coexistence of undernutrition and overnutrition reflects wider challenges in many low- and middle-income countries. These results add to global evidence on child nutrition and may guide similar interventions in resource-limited settings.