2019/09/16 by V.W.L. Chin, Chin, Vincent, Jarod Y. L. Lee +7
Health Professions · Nursing · #Applications (stat.AP) #Child Nutrition and Water Access #FOS: Computer and information sciences #Food Security and Health in Diverse Populations #Methodology (stat.ME)
paper · pdf · doi:10.48550/arxiv.1909.07550
openalex publication_date 2019/09/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Faltering growth among children is a nutritional problem prevalent in low to\nmedium income countries; it is generally defined as a slower rate of growth\ncompared to a reference healthy population of the same age and gender. As\nfaltering is closely associated with reduced physical, intellectual and\neconomic productivity potential, it is important to identify faltered children\nand be able to characterise different growth patterns so that targeted\ntreatments can be designed and administered. We introduce a multiclass\nclassification model for growth trajectory that flexibly extends a current\nclassification approach called the broken stick model, which is a piecewise\nlinear model with breaks at fixed knot locations. Heterogeneity in growth\npatterns among children is captured using mixture distributed random effects,\nwhereby the mixture components determine the classification of children into\nsubgroups. The mixture distribution is modelled using a Dirichlet process\nprior, which avoids the need to choose the "true" number of mixture components,\nand allows this to be driven by the complexity of the data. Because children\nhave individual differences in the onset of growth stages, we introduce\nchild-specific random change points. Simulation results show that the random\nchange point model outperforms the broken stick model because it has fewer\nrestrictions on knot locations. We illustrate our model on a longitudinal birth\ncohort from the Healthy Birth, Growth and Development knowledge integration\nproject funded by the Bill and Melinda Gates Foundation. Analysis reveals 9\nsubgroups of children within the population which exhibit varying faltering\ntrends between birth and age one.\n