This is the authors' abstract. We don't add key points for this paper.
Objectives To evaluate the effect of an integrated nutrition education and home gardening intervention on dietary diversity among mothers in central Oromia, Ethiopia using machine learning supported analysis. Methods A community-based quasi-experimental study was conducted from May to October 2025 among 634 participants (interventions=318, and controls=316). The intervention group received twelve nutrition education sessions, and home gardening, while the comparison group received routine maternal and child nutrition services Maternal dietary diversity was assessed using the Minimum Dietary Diversity for Women (MDD-W) indicator based on a 24-hour dietary recall. The intervention effect was estimated using an adjusted Difference-in-Differences regression model controlling for baseline covariates. Multivariable logistic regression identified independent determinants, while seven supervised machine-learning algorithms were evaluated using accuracy, precision, recall, F1-score, Cohen's Kappa, and area under the receiver operating characteristic curve (AUC). Feature importance was assessed using SHAP and the Random Forest Mean Decrease in Gini. Results The adjusted Difference-in-Differences analysis demonstrated a significant intervention effect after controlling for baseline characteristics (Group × Time interaction: β = 0.72, 95% CI: 0.65, 0.86, P < 0.001). Adequate dietary diversity increased from 35.8% to 86.5% in the intervention group compared with 43.4% in the comparison group (χ 2 = 129.53, P < 0.001). Maternal age, educational status, household wealth index, and home gardening were significant determinants. The Random Forest model achieved the highest overall predictive performance (AUC = 0.71; accuracy = 65%). SHAP analysis identified household wealth, maternal age, and educational status as the most influential predictors. Conclusions The findings suggest that integrated nutrition education and home gardening significantly improved maternal dietary diversity. Combining multivariable regression with machine-learning analysis provided complementary evidence for identifying key determinants and predicting women at risk of inadequate dietary diversity, supporting targeted nutrition interventions and public health decision-making.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Yazew et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: