Enhancing Treatment Precision: Evaluating The Validity and Reliability of Modified NutricheQ in Detecting Iron Deficiency in Children Aged 1-3 Years in Indonesia

Background Iron deficiency anemia remains a persistent issue in developing countries, including Indonesia, prompting recommendations for routine iron supplementation. However, supplementation is often provided without prior screening, increasing the risk of unnecessary treatment and potential side effects. Validated, non-invasive screening tools could be especially valuable in these settings. This study aimed to evaluate the validity and reliability of a modified NutricheQ Questionnaire for identifying iron deficiency risk in the Indonesian pediatric population. Method A two-step cross-sectional study was conducted among 300 children aged 1–3 years across Jakarta, Indonesia. The first step assessed the validity of the modified NutricheQ Questionnaire, followed by analysing its correlation with dietary iron intake and serum ferritin level. Validity was assessed through Spearman correlation test. Internal consistency was measured using Cronbach’s alpha, and inter-rater reliability was assessed using Cohen’s kappa. Receiver operating characteristic (ROC) analysis was performed to identify optimal threshold scores based on sensitivity, specificity, and area under the curve (AUC). Results The modified questionnaire demonstrated acceptable construct validity, with Spearman correlation coefficients >0.30 for all items except red meat intake. Inter-rater reliability ranged from fair to perfect agreement (Cohen’s kappa: 0.40–1.00). There was moderate correlation with dietary iron intake (r = 0.39, p < 0.01) and serum ferritin levels (r = 0.29, p < 0.01). The 5-question version had an optimal threshold score of 5, yielding a sensitivity of 80.4% and specificity of 48.3% in predicting iron intake; AUC was 0.709 (CI 95%: 0.648-0.770). A shortened 3-question version with a threshold of 4 maintained the same sensitivity (80.4%) and improved specificity (67.0%) for predicting iron intake (AUC = 0.768; 95% CI: 0.712–0.825), with sensitivity of 71.6% and specificity of 61.5% for predicting low serum ferritin (AUC = 0.682; 95% CI: 0.597–0.758). Conclusion The modified NutricheQ Questionnaire demonstrated good validity and reliability for identifying risk factors of iron deficiency in Indonesian children aged 1–3 years. The shortened three-item version, in particular, shows promise as a non-invasive screening tool for use in low-resource settings.