Abstract

Malnutrition remains a significant public health problem in developing countries, including Indonesia. In addition to stunting that has received a lot of attention, the problem of underweight nutrition in toddlers still requires serious studies and interventions. This study aims to model the nonlinear relationship between age and weight of toddlers, as well as identify factors that affect the incidence of underweight in West Lombok Regency. The semiparametric approach is used by applying piecewise polynomial regression and truncated splines with variations in polynomial orders and number of node points. The research data included toddler weight as a response variable, as well as 22 predictor variables representing the child's health condition, maternal characteristics, feeding practices, and household sanitation. The selection of the best model is made based on the Akaike Information Criterion (AIC) criteria. The results of the analysis showed that the two-order piecewise polynomial model with four knots was the best model, indicated by the lowest AIC value and the coefficient of determination (R²) of 97%. These findings indicate that the problem of underweight in toddlers is complex and multidimensional, requiring an integrated intervention approach across sectors. The semiparametric approach has been shown to be effective in capturing the growth patterns of toddlers that are not linear and dynamic.

Concepts :

Access to Document

Citations by Year