Stunting remains a major nutritional problem that requires accurate and reliable assessment of children’s current growth status. Machine learning can support stunting classification; however, class imbalance and limited model interpretability remain important challenges. This study aims to evaluate the effect of SMOTEENN across classification algorithms with different learning characteristics and interpret the best-performing model using SHAP. Two secondary datasets were integrated, resulting in 41,176 records, of which 41,173 were retained after data cleaning and numerical conversion. Logistic Regression, Decision Tree, Random Forest, and XGBoost were evaluated using original and SMOTEENN-balanced training data. Hyperparameter optimization was performed using RandomizedSearchCV with stratified five-fold cross-validation and ROC-AUC as the optimization metric. The optimized configurations were fixed and applied to both experimental scenarios. Model performance was evaluated using Accuracy, Precision, Recall, F1-Score, and ROC-AUC on an independent test set. Random Forest trained on the original data achieved the best overall performance, with an Accuracy of 0.9600, Precision of 0.9584, F1-Score of 0.9505, and ROC-AUC of 0.9927. SMOTEENN consistently improved Recall across all models but generally reduced the overall performance of tree-based models. The largest improvement occurred in Logistic Regression, with Recall increasing from 0.7058 to 0.8657. SHAP analysis identified Height and Age (Month) as the most influential features. These findings show that the effect of SMOTEENN depends on dataset and algorithm characteristics, while SHAP provides transparent interpretation of current stunting-status classification within the analyzed datasets.
Siti Nurhasanah Nugraha, Muji Ernawati, Evita Fitri et al.
Iron and folic acid supplementation for adolescent girls is essential for preventing anemia in countries such as Indonesia, where the prevalence of anemia ranges from 20% to 40%. The iron supplementation coverage rate in Bekasi City is 13.29%. However, only 8.3% of adolescent girls adhere to iron supplementation. This study aimed to analyze relationships between predisposing, enabling, reinforcing factors and adherence to iron supplementation among adolescent girls in Bekasi, Indonesia. A descriptive cross-sectional study was conducted between June and August 2023 using validated questionnaires to assess predisposing, enabling, and reinforcing factors. A total of 400 participants were selected using the multistage sampling method consisting of cluster random sampling followed by consecutive sampling. Chi-square tests and multiple logistic regression analyses were used to establish the relationship between the variables. The availability of iron supplementation at schools, peer support, family support, attitude, and teacher support were found to be significantly associated with adjusted odds ratios(AOR) consecutively of 4.738 (2.654–8.460), 4.218 (2.150–8.275),2.463 (1.379–4.398), 2.291 (1.273–4.112), and 2.089 (1.058–4.124). The availability of supplements in schools emerged as the strongest factor associated with adherence to iron supplementation. Anemia prevention strategies for adolescent girls should prioritize interventions targeting the most influential determinants of adherence.
Malnutrition is a major nutritional and public-health concern affecting children and adolescents in India. It includes under nutrition, micronutrient deficiencies and over nutrition, with stunting, wasting, underweight, anemia, overweight and obesity representing important manifestations. The nutritional condition of children is influenced by dietary quality, household food security, socioeconomic status, parental education, sanitation, healthcare access and feeding practices. Jaunpur district of Uttar Pradesh provides an important regional context for examining these interrelated determinants. The District Nutrition Profile of Jaunpur prepared using national survey data indicates a substantial burden of child under nutrition and anemia, while more recent national evidence demonstrates the continuing importance of addressing both under nutrition and emerging overweight. This review synthesizes available evidence related to malnutrition, dietary practices, micronutrient deficiencies, socioeconomic determinants and possible Home Science-based interventions relevant to Jaunpur district. Evidence from South Asia indicates that nutritional problems among school-age children and adolescents remain heterogeneous and that inadequate dietary diversity and unhealthy food consumption are important concerns. Nutrition education, dietary diversification, school-based interventions, growth monitoring and family-centered counseling can contribute to improving nutritional outcomes. The review identifies a need for more district-specific research among school-age children in Jaunpur, particularly studies combining anthropometric assessment, dietary diversity, nutritional awareness and socioeconomic variables.
Background: Stunting in the first two years of life remains a major public health concern in Kenya, affecting about one in five children and leading to long-term consequences on cognitive development and economic productivity. Traditional, largely reactive approaches that rely on linear statistical models have a limited ability to capture the complex, non‑linear determinants of stunting. Given that stunting arises from an interplay of immediate, underlying, and structural determinants, machine learning (ML) methods offer a promising alternative. ML techniques can leverage large, heterogeneous datasets to uncover hidden patterns and non‑linear interactions among associated factors, potentially enabling earlier and more accurate identification of children with higher probabilities of stunting. Objective of the study: This study applied supervised machine learning techniques to predict early childhood stunting among children aged 0-23 months and identify associated factors using nationally representative data from the 2022 Kenya Demographic and Health Survey. Methods: Three models—logistic regression (LR), balanced random forest (BRF), and categorical boosting (CatBoost)—were developed and evaluated using metrics aligned with programmatic priorities, including recall, area under the receiver operating characteristic curve (AUC), F1-score, and Brier Score. To address class imbalance, balanced class weights were applied, and models were trained at a threshold of 0.4 to optimize sensitivity. Results: CatBoost achieved the highest overall performance (AUC = 70.8%, recall = 85.8%), while LR remained highly competitive (AUC = 71.5%, Recall = 77.2%), outperforming BRF. Across all three models, consistent predictors included the child’s age, sex, birth size, twin status, maternal stature, household wealth, and geographic location, reflecting the multifactorial nature of stunting and aligning with existing literature. Conclusions: Simpler models, when combined with domain-informed feature engineering, could perform comparably to advanced algorithms while supporting transparency, which is essential for policy-relevant decision-making. The findings also provide methodological support for integrating predictive modelling into nutrition research and highlight opportunities to improve stunting prevention strategies.
Stunting is a long-term nutritional issue and a serious public health issue, especially for young children under two. In order to meet nutrient requirements and promote linear growth, dietary diversification is crucial. This study aimed to determine the relationship between dietary diversity, measured using the Dietary Diversity Score (DDS), and stunting among children aged 6–24 months in the working area of Bagu Primary Health Center, Central Lombok Regency. An analytic observational study with a cross-sectional design was conducted among 43 mothers of children aged 6–24 months using total sampling. Dietary diversity was assessed using a structured questionnaire based on nine food groups recommended by the FAO, while stunting status was determined from anthropometric records in the Maternal and Child Health handbook. Data were analyzed using the Chi-square test. Of the 43 children, 62.8% were stunted and 37.2% were severely stunted. Most children (58.1%) had low dietary diversity, while 41.9% had diverse diets. No significant association was found between dietary diversity and stunting severity (p = 0.655). However, diverse diets showed a lower tendency toward severe stunting (OR = 0.750; 95% CI: 0.212–2.656). The findings highlight the need to improve dietary diversity through locally available food sources.
Baiq Nur Lathifatuzzahrah, Lina Nurbaiti, Wiwin Lastyana et al.
Public health nutrition is shaped by the environments in which people obtain food, make dietary choices, experience nutritional vulnerability, and respond to changing health and social conditions.This third issue of Volume 17 of World Nutrition brings together 21 contributions spanning original research, reviews, commentaries, opinion, and policy perspectives, reflecting the breadth and interconnectedness of contemporary nutrition challenges.Collectively, these contributions emphasise the importance of context, equity, accountability, and evidence-informed action in improving nutrition and health outcomes.A prominent theme is the food environment and the accountability of institutions that influence dietary choices.Sezaki et al., in Access to Nutrition Initiative Global Index and Spotlight Index comparisons: Promoting nation-led frameworks for nutrition accountability, examine the evolution of corporate nutrition accountability frameworks and the growing role of country-specific approaches.Complementing this work, du Plessis et al., in Food and non-alcoholic beverage marketing and sponsorship at secondary schools in the City of Cape Town, South Africa, document extensive food and beverage marketing within schools.Riofrio Jacinto et al., in Food access, purchasing habits, and dietary diversity of peri-urban households in Hanoi, Vietnam, further demonstrate how purchasing behaviour, affordability, proximity, and food assortment can shape dietary diversity.Together, these studies illustrate why nutrition policy must consider not only individual choices but also the environments in which those choices are made.The issue gives considerable attention to nutrition across the life course.Ali et al., in From Breast to Bowl: A quasi-experimental, mixed-methods evaluation of peer learning to improve complementary feeding in rural northeastern Ghana, demonstrate the potential of peer learning to strengthen complementary feeding practices.Kamda et al., in Factors associated with diet adequacy among women attending antenatal care in the conflict-affected Kumba Municipalities, Cameroon, examine the nutritional vulnerabilities women experience in conflict-affected settings.Funduluka et al., in Unveiling dietary patterns among children aged 6-23 months: A framework analysis for perspectives in Chinyunyu Community, Rufunsa District, Zambia, identify poverty, household decision-making, workload, and food availability as important influences on infant and young child feeding.Pencil et al., in Anaemia among in-school adolescents from Harare, Zimbabwe: Prevalence and key drivers, highlight the burden of anaemia among adolescents and the relevance of household education, nutrition knowledge, and dietary practices.Onyeanwu, in Food insecurity among older adults in the United States: A systematic review through the gerontological public health framework, extends attention to older adults, highlighting the interconnected economic, social, environmental, and healthcare determinants of food insecurity and its consequences.
Background/Objectives: Bangladesh carries one of the world's highest estimated childhood lead burdens alongside a substantial undernutrition burden, yet no nationally representative study has examined the association of blood lead levels (BLLs) with anthropometric failure. Methods: We analysed data from the 2025 Bangladesh Multiple Indicator Cluster Survey, including 10,495 children aged 12–59 months with BLLs and anthropometric measurements. Venous blood lead, arsenic, cadmium, and mercury were measured alongside anthropometry. Associations of BLLs with height-for-age (HAZ), weight-for-age (WAZ), and weight-for-height (WHZ) z-scores were estimated using linear regression; associations with stunting, wasting, underweight, and the Composite Index of Anthropometric Failure (CIAF, comprising six disaggregated failure categories) were estimated using modified Poisson regression yielding adjusted prevalence ratios (APRs). Models considered continuous standardised log-transformed BLLs and elevated BLLs (≥5 µg/dL), with and without adjustment for arsenic, cadmium, mercury, and covariates. Results: The prevalence of stunting, wasting, underweight, and CIAF was 25.9%, 14.1%, 26.6%, and 38.2%, respectively. Elevated blood lead prevalence was 38.6 Higher BLLs were significantly associated with lower HAZ (β=-0.075), WAZ (β=-0.062), and WHZ (β=-0.029), and with higher prevalence of stunting (APR=1.09), wasting (APR=1.08), underweight (APR=1.08), and CIAF (APR=1.07), largely unchanged after adjustment for co-exposure. The association of WHZ was not significant with elevated BLLs. BLLs were associated with several CIAF patterns, most strongly the combined stunting-wasting-underweight category (APR =1.18), with significant associations present across every category involving stunting. In exploratory analyses, elevated BLLs were also significantly associated with household soil lead concentration (APR=1.20), occupational contact with lead-related work (APR=1.08), and child pica behaviour (APR=1.18), and were inversely associated with solid fuel use for cooking (APR=0.78). Elevated BLL prevalence varied substantially by district (range 4.4–93.7%), with higher burden concentrated in the central-eastern region. Conclusions: BLLs are independently associated with impaired growth and specific patterns of CIAF among Bangladeshi children. These findings support prioritising lead-specific source control and integrating growth monitoring with environmental exposure surveillance in national child health programming, and suggest CIAF-based surveillance may capture lead's population-level nutritional burden more fully than single indicators alone. Household soil lead, occupational exposure, and child pica behaviour represent modifiable targets for source-control interventions.
Early childhood development (ECD) is the foundation for health, cognitive ability, and social and economic productivity throughout life. Nutritional deprivation and socio-economic inequality continue to constrain developmental potential in low- and middle-income countries. This study aimed to assess the nutritional, socio-economic, and caregiving factors associated with early childhood development among children aged 3–4 years in Bangladesh. Data from the 2019 Bangladesh Multiple Indicator Cluster Survey (MICS), a nationally representative survey, were analyzed. The Early Childhood Development Index (ECDI) was constructed across four domains: literacy-numeracy, physical, social-emotional, and learning. A three-level ordinal outcome (not on track, partially on track, and on track) was derived. Survey-adjusted partial proportional odds models were estimated using Stata 17 to account for the complex sampling design. The proportional odds assumption was evaluated using a design-adjusted Wald test. Among the 8,752 children included in the analysis, 73.3% were developmentally on track, 23.6% were partially on track, and 3.2% were not on track. After adjustment, children without functional disabilities had substantially higher odds of being developmentally on track (AOR = 3.93). Stunted children had 17% lower odds of optimal developmental outcomes (AOR = 0.83). Higher maternal education (AOR = 1.17 per level increase), greater household wealth (AOR = 1.11 per quintile increase), attendance in early childhood education programs (AOR = 1.53), female sex (AOR = 1.40), and increasing age were positively associated with developmental attainment. The interplay of nutritional, socio-economic, and caregiving factors significantly shapes early childhood development in Bangladesh. Developmental outcomes can be improved through interventions addressing child undernutrition, educational inequality, and disability inclusion. A comprehensive and equitable policy framework is necessary to maximize opportunities for children's early development.
Introduction: Stunting is a major public health problem that affects the growth and development of children under five. Functional foods in child-friendly forms may provide an alternative nutritional approach to support linear growth and stunting prevention. Objective: This study aimed to evaluate the potential of red ginger extract gummy candy as a functional food to support linear growth and stunting prevention among children under five. Method: This quasi-experimental study used intervention and control groups. Children in the intervention group received red ginger extract gummy candy, while those in the control group received a placebo for one month. Changes in height and weight were assessed before and after the intervention. Result: The Shapiro–Wilk test showed that the anthropometric data were not normally distributed (p<0.05). In the intervention group, height increased significantly after the intervention (p=0.005), whereas the change in body weight was not statistically significant (p=0.161). In the control group, neither body weight (p=0.138) nor height (p=0.059) showed statistically significant changes. However, the Mann–Whitney U test showed no statistically significant difference between the intervention and control groups in changes in body weight (p=0.970) or height (p=0.701).. Conslusion: Red ginger gummy candy was associated with a significant increase in height within the intervention group, but the change was not significantly different from the control group. Further studies with larger samples and longer interventions are needed to determine its potential in supporting linear growth.
Introduction: Appropriate complementary feeding from 6–23 months is essential for child growth and development, yet dietary diversity remains inadequate in many resource-constrained settings. Understanding contextual influences on feeding practices is important for improving child nutrition. Objective: This study examined qualitative evidence to identify dietary patterns and factors shaping complementary feeding among children aged 6–23 months in Chinyunyu, Rufunsa District, Zambia. Methods: This Framework Analysis applied Gale et al. (2013) and the WHO Infant and Young Child Feeding framework to interpret qualitative data. The study was conducted in drought-prone Chinyunyu Ward, Rufunsa District, Zambia. Participants included 24 women aged 18–49 years with a child under two years and seven mothers-in-law who participated in Photovoice and focus group discussions, alongside three husbands and four community leaders who participated in in-depth interviews. Results: Caregivers generally described awareness of recommended feeding practices; however, poverty, limited maternal decision-making power, and heavy workloads were perceived to constrain adherence. Diets cited were dominated by cereal-based porridges prepared with groundnuts, sugar, or salt. Groundnuts were reported as the main protein source, and reliance on them may have limited dietary diversity. Furthermore, increasing dependence on commercial foods was also perceived to undermine optimal complementary feeding. Conclusions: Despite awareness of recommended feeding practices, poverty, gendered resource control, and heavy workloads were reported to constrain adherence to WHO recommendations. Empowering women, improving livelihoods, and promoting affordable dietary diversity, including legumes beyond groundnuts, may strengthen complementary feeding and child nutrition.
Priscilla Funduluka, Mercy Mukuma, Regina Murphy Keith et al.
